Early Experience and The Functional Calibration of The Stress-Response Systems

Western University
Scholarship@Western
Electronic Thesis and Dissertation Repository
11-19-2021 1:00 PM
Early experience and the functional calibration of the stress-

response systems
Niki Hosseini-Kamkar, The University of Western Ontario
Supervisor: Morton, J Bruce, The University of Western Ontario

A thesis submitted in partial fulfillment of the requirements for the Doctor of Philosophy degree
in Psychology
© Niki Hosseini-Kamkar 2021
Follow this and additional works at: https://ir.lib.uwo.ca/etd
Part of the Cognitive Neuroscience Commons
Recommended Citation
Hosseini-Kamkar, Niki, "Early experience and the functional calibration of the stress-response systems"
(2021). Electronic Thesis and Dissertation Repository. 8252.
https://ir.lib.uwo.ca/etd/8252
This Dissertation/Thesis is brought to you for free and open access by Scholarship@Western. It has been accepted
for inclusion in Electronic Thesis and Dissertation Repository by an authorized administrator of
Scholarship@Western. For more information, please contact wlswadmin@uwo.ca.
Abstract
Individuals exposed to adversities in childhood are at a greater risk of developing

various diseases as adults, including cardiovascular disease and cancer (Felitti et al. 1998).
These findings have sparked an interest in examining biological mechanisms that might
explain the link between exposure to adversity and disease. To date, evidence has linked
adversity to the function of the hypothalamic-pituitary-adrenal (HPA) axis. More recently,
adversity has been associated with the function of the mesolimbic dopamine pathway as well.
This thesis uses a variety of techniques to explore the association between adversity
and the function of the HPA axis and mesolimbic dopamine pathway. Current evidence
examining exposure to adversity and the function of the HPA axis and mesolimbic dopamine
pathway remains highly contradictory—with some studies reporting heightened cortisol and
dopamine reactivity, and others reporting blunted reactivity. I hypothesized that the mixed
findings may be related to inconsistent definitions of what constitutes “adversity”. The term
“adversity” is typically used broadly and ranges from moderate stressors (e.g., work-related
stress) to extremely traumatic events (e.g., sexual abuse). This thesis makes a clear
distinction between trauma and adversity to help resolve the apparent contradictions in the
literature.
In Chapter 2, p-curve meta-analysis was used to explore the literature linking trauma versus
adversity to cortisol reactivity. The results provided support for associations between trauma
and blunted cortisol reactivity, and moderate adversity and heightened cortisol reactivity.
In Chapter 3, a systematic review of the human Positron Emission Tomography (PET)

imaging studies was conducted to examine the relationships between adversity versus trauma
and dopamine reactivity. The results mirror the findings from Chapter 2, demonstrating that
trauma is associated with blunted dopamine reactivity, while adversity is associated with
heightened dopamine reactivity.
ii
Chapter 4 uses behavioural and functional magnetic resonance imagining (fMRI) methods to
explore the link between adversity, impulsivity, reward-learning, and ventral striatal
reactivity to rewards. The findings demonstrate that adversity is associated with impulsivity
and potentiated reward-learning, and that the association between adversity and reward-
learning was partially mediated by ventral striatal reactivity. Overall, these findings support
an inverted U-shaped relationship between severity of adversity and cortisol and dopamine
reactivity.
Keywords
Adversity, trauma, cortisol, dopamine, HPA axis, mesolimbic dopamine pathway, plasticity,
p-curve, meta-analysis
iii
Summary for Lay Audience
Do adverse life-experiences leave a lasting imprint on the brain? Likewise, does

experiencing adversity have consequences for health and well-being? Seminal work by Felitti
and colleagues (1998) showed that individuals exposed to adversities in childhood are at a
greater risk of developing various diseases as adults, including heart disease and cancer. As a
result of these findings, researchers began investigating how adversity becomes “biologically
embedded into the brain”. To date, evidence has linked adversity to the function of two
neural systems:
1.) the hypothalamic-pituitary-adrenal (HPA) axis involved in stress responses and
cortisol secretion
2.) the dopamine pathway involved in reward-learning and decision-making
This thesis addressed how adverse life-experiences are related to the function of the HPA
axis and dopamine pathway. There is a wealth of data showing that adversity is related to
differences in the function of the HPA axis and dopamine pathway, but what is unclear, is if
adversity is related to heightened cortisol and dopamine levels, or reduced cortisol and
dopamine levels. One reason for the mixed findings is that the term “adversity” is broad and
can range from moderate stressors (e.g., work-related stress) to extreme trauma (e.g., sexual
abuse). Here, I differentiated between moderate adversity and extreme trauma and I
examined how each is related to differences in the function of the HPA axis and dopamine
pathway. In Chapter 2, I examined how adversity and trauma are related to cortisol function,
and I found that trauma was related to reduced cortisol reactivity, while adversity was related
to heightened cortisol reactivity. Chapter 3 investigated the link between adversity and
trauma and the dopamine pathway; like the findings in Chapter 2, I found that adversity was
related to heightened dopamine levels, while trauma was related to reduced dopamine levels.
Finally, in Chapter 4, 9-12-year-old children participated in a neuroimaging study and we
obtained measures of adversity from their parents. The results showed that children who
experienced more adversity were more impulsive, learned faster from rewards, and the
reward regions within the dopamine pathway reacted more to receiving rewards compared to
children who had experienced less adversity.
iv
Co-Authorship Statement
The work presented in this doctoral thesis was designed and written in collaboration with my
advisor, Dr. J. Bruce Morton. While I am the primary author of each of the chapters
presented in this work, Dr. Morton has contributed to the design of projects, interpretation of
the findings, as well as in revising and editing the work presented in this thesis.
v
Acknowledgments
First and foremost, I would like to express my thanks and gratitude to my advisor—Dr. J
Bruce Morton. Thank you for being supportive, kind, and insightful. Thank you for always
believing in me and giving me the freedom to pursue research questions that were of my own
interest; but above all else, thank you for being patient and understanding through it all.
I would also like to express my thanks to my advisory committee—Drs. Jesicca Grahn and
Daniel Ansari. Dr. Grahn, you have always been a role-model for me and you’ve inspired me
as a female scientist on many occasions. Dr. Ansari your enthusiasm for science is a
motivator for any graduate student that crosses your path, and my work in Chapter 2 (p-
curve) would not have been possible without your help and feedback.
I would like to thank Bea Goffin who was always a constant source of support through the
darkest of days. Likewise, I’d like to thank Anna Matejko who taught me how to write an
NSERC application, how to write my first conference abstract, and many other “grad school
firsts”—you’re amazing and I am so lucky to have met you. I’d like to express my gratitude
to friends and colleagues in the lab and department (past and present), including Dr. Mazen
El-Baba, Dr. Isu Cho, Dr. Cassandra Lowe, Samantha Goldsmith, and Daniel J. Lewis—
thank you for being my support system and for getting me through some of the most difficult
days.
Words cannot express my gratitude to my family – my mom (Shohreh Shahi) who taught me
what it truly means to be strong and resilient—mom, I look up to you for everything. My dad
(Javad Kamkar) who is patient and kind and inspires me to learn to be a better person. And of
course, my amazing, wonderful, incredible sister –Nellie Kamkar. Nellie, I wouldn’t be able
to get through anything in life, including writing this thesis, without your help and support.
I’m pretty sure that at this point, you could defend this research better than I could
considering the countless times you’ve listened to me talk about it and present it. Thank you
for EVERYTHING. Of course, my friends who are also family—Elena, Krista, Jolene, and
Ali—thank you too for tolerating my inordinate levels of stress throughout the years.
A very special thank you to my friends and colleagues –Mel Colindras, Marley Zia, Belal
Zia, Erica Tobias, and Andrew Moreland for assisting with data collection.
vi
Table of Contents
Abstract ............................................................................................................................... ii
Co-Authorship Statement.................................................................................................... v
Acknowledgments.............................................................................................................. vi
Table of Contents .............................................................................................................. vii
List of Tables (where applicable) ..................................................................................... xii
List of Figures (where applicable) ................................................................................... xiii
List of Appendices (where applicable) ........................................................................... xvii
Chapter 1 ............................................................................................................................. 1
1 General Introduction ...................................................................................................... 1
1.1 The long-term impact of early-life adversity .......................................................... 1
1.2 Adaptive phenotypic plasticity ............................................................................... 2
1.2.1 Individual differences in phenotypic plasticity ........................................... 3
1.3 The role of the central nervous system ................................................................... 4
1.3.1 The plasticity of the HPA axis .................................................................... 5
1.3.2 The plasticity of the dopamine system........................................................ 6
1.4 Distinguishing adversity and trauma ...................................................................... 8
1.5 Aim of current thesis and summary of findings...................................................... 9
1.5.1 Chapter 2 – The differential calibration of the HPA axis as a function of

trauma versus adversity: A systematic review and p-curve meta-analyses 9
1.5.2 Chapter 3 – Stress, glucocorticoids, and the mesolimbic dopamine

pathway: How early-life experiences become biologically embedded into
the dopamine system ................................................................................. 11
1.5.3 Chapter 4 – Ventral striatal reactivity links adversity and reward

processing in children ............................................................................... 12
1.6 Adaptive plasticity: Stress-response Adaptations to Variability in Early-life

experiences (SAVE) model................................................................................... 12
1.7 References ............................................................................................................. 15

vii
Chapter 2 ........................................................................................................................... 22
2 The differential calibration of the HPA axis as a function of trauma versus adversity:
A systematic review and p-curve meta-analyses ......................................................... 22
2.1 Introduction ........................................................................................................... 22
2.1.1 The stress response system ....................................................................... 23
2.2 Theoretical frameworks for understanding the relationship between adversity and
HPA axis function ................................................................................................. 25
2.2.1 Allostatic load and overload: Wear and tear on the body ......................... 25
2.2.2 The General Adaptation Syndrome (G-A-S) Model ................................. 27
2.2.3 Adaptationist perspectives ........................................................................ 28
2.3 Empirical findings on the association between life-experiences and HPA axis
reactivity ............................................................................................................... 30
2.4 Sources of variability and confounding factors .................................................... 31
2.4.1 Cortisol measurements .............................................................................. 31
2.4.2 Trauma versus adversity ........................................................................... 33
2.5 Current studies: p-curve meta-analyses ................................................................ 35
2.6 Methods................................................................................................................. 37
2.6.1 Study 1: Data sources and study selection ................................................ 37
2.6.2 Study 2: Data sources and study selection ................................................ 43
2.6.3 P-curve data abstraction ............................................................................ 48
2.7 Results ................................................................................................................... 48
2.7.1 Study 1.1: Traumatic experiences and cortisol hyper-reactivity p-curve . 48
50
2.7.2 Study 1.2: Traumatic experiences and cortisol hypo-reactivity p-curve .. 52
2.7.3 Study 2.1: Adverse experiences and cortisol hyper-reactivity p-curve .... 55
2.7.4 Study 2.2: Adverse experiences and cortisol hypo-reactivity p-curve ..... 56
2.8 Discussion ............................................................................................................. 59
viii
2.8.1 Conclusion ................................................................................................ 64
2.9 References ........................................................................................................... 121
Chapter 3 ......................................................................................................................... 141
3 Stress, glucocorticoids, and the mesolimbic dopamine pathway: How early-life

experiences become biologically embedded into the dopamine system .................... 141
3.1 Introduction ......................................................................................................... 141
3.1.1 The mesolimbic dopamine pathway: A brief review .............................. 142
3.1.2 Stress and drugs: Animal studies ............................................................ 143
3.1.3 Stress and drugs: Human studies ............................................................ 145
3.1.4 Mechanisms of action: The link between the HPA axis and the mesolimbic
dopamine pathway .................................................................................. 146
3.1.5 Mechanisms of action: The role of glucocorticoids ................................ 147
3.1.6 Stress and dopamine: Animal literature .................................................. 149
3.1.7 The link between the HPA axis and the mesolimbic dopamine pathway in
humans .................................................................................................... 151
3.1.8 Theoretical frameworks and models ....................................................... 152
3.1.9 Stress-response Adaptations to Variability in Early-life experiences

(SAVE) model ........................................................................................ 153
3.1.10 Current study and hypotheses ................................................................. 154
3.2 Methods............................................................................................................... 155
3.2.1 Data sources and study selection ............................................................ 155
3.2.2 Screening and inclusion .......................................................................... 155
3.3 Results ................................................................................................................. 158
3.4 Discussion ........................................................................................................... 161
3.1 Conclusion .......................................................................................................... 164
3.2 References ........................................................................................................... 166
Chapter 4 ......................................................................................................................... 182
4 Ventral striatal activity links adversity and reward processing in children ............... 182
ix
4.1 Introduction ......................................................................................................... 182
4.2 Methods............................................................................................................... 185
4.2.1 Method Study 1 ....................................................................................... 185
4.3 Results – Study 1 ................................................................................................ 187
4.4 Methods Study 2 ................................................................................................. 188
4.4.1 Participants .............................................................................................. 188
4.4.2 Measure of reward-based learning .......................................................... 189
4.4.3 MRI data acquisition ............................................................................... 190
4.4.4 fMRI data pre-processing ....................................................................... 190
4.4.5 Event-related modeling ........................................................................... 191
4.4.6 Reward-related Region of Interest (ROI) analysis.................................. 191
4.5 Results – Study 2 ................................................................................................ 192
4.5.1 Behavioral results.................................................................................... 192
4.5.2 ROI activation, relation to behavior........................................................ 194
4.5.3 Adversity-behavior association partially explained by VS activation .... 195
4.6 Discussion ........................................................................................................... 196
4.7 References ........................................................................................................... 200
Chapter 5 ......................................................................................................................... 209
5 General Discussion..................................................................................................... 209
5.1 Developmental plasticity and adverse life-experiences ...................................... 209
5.2 Summary of thesis and common themes ............................................................ 210
5.2.1 Aims of current thesis ............................................................................. 210
5.2.2 Summary of findings............................................................................... 210
5.2.3 Adversity versus trauma ......................................................................... 212
5.3 Integration of findings with available theory ...................................................... 213
5.3.1 The Biological Sensitivity to Context (BSC) model ............................... 213

x
5.3.2 Theories on the plasticity of the mesolimbic dopamine pathway ........... 214
5.3.3 Stress-response Adaptations to Variability in Early-life Experiences

(SAVE)..................................................................................................... 214
5.4 Implications......................................................................................................... 215
5.4.1 Health consequences related to cortisol and dopamine reactivity .......... 215
5.4.2 Adaptive advantages of high-risk behaviours ......................................... 216
5.5 Limitations and future directions ........................................................................ 217
5.6 Final Remarks ..................................................................................................... 219
5.7 References ........................................................................................................... 220
Tables ............................................................................................................................. 254
Table S1.......................................................................................................................... 254
Curriculum Vitae ............................................................................................................ 255
xi
List of Tables (where applicable)
Table 2.1: P-curve Disclosure Table of Trauma-Cortisol Hypo-reactivity Literature ........... 77
Table 2.2: P-curve Disclosure Table of Adversity-Cortisol Hyper-reactivity Literature ..... 103
Table 2.3: P-curve Disclosure Table of Adversity-Cortisol Hypo-reactivity Literature ...... 112
Table 3.1: Summary of all studies reporting associations between adverse life-experiences
and dopaminergic function ......................................................Error! Bookmark not defined.
Table 3.2: Overview of human PET studies of adversity versus trauma and dopamine
reactivity ..................................................................................Error! Bookmark not defined.
xii
List of Figures (where applicable)
Figure 2.1: Graphical depiction of relationship between severity of adversity and HPA axis
reactivity as proposed by allostatic load/overload models (Danese & McEwen, 2012). In
healthy individuals, exposure to trauma is associated with blunted HPA axis reactivity; in
contrast, in individuals with comorbid mental illnesses, trauma is related to heightened HPA
axis reactivity. ......................................................................................................................... 26
Figure 2.2: Graphical depiction of the relationship between duration of adversity and HPA
axis reactivity as proposed by the G-A-S model (Selye, 1946). According to this model, acute
stressors result in a phase of hypercortisolism; however, if exposure to adversity persists
long-term, hypocortisolism will occur. ................................................................................... 28
Figure 2.3: Graphical depiction of the relationship between severity of adversity and HPA
axis reactivity as proposed by the BSC model (Boyce & Ellis, 2005). According to this
model, HPA hyper-reactivity emerges under both protected environments with low adversity
and extremely stressful environments. .................................................................................... 30
Figure 2.4: Graphical depiction of Study 1 and Study 2 aims. The orange curvilinear function
represents the predictions that can be made about the relationship between adversity and
HPA axis reactivity based on the G-A-S model. The U-shaped function in black represents
the predictions that can be made concerning the relationship between severity of adversity
and HPA axis reactivity based on the BSC model. In Study 1 (outlined in red), we will use p-
curve to determine whether the literature reporting that exposure to trauma is related to
heightened cortisol reactivity or blunted cortisol reactivity contains evidential value. In Study
2 (outlined in yellow), we will use p-curve to assess whether exposure to moderate adversity
is related to heightened or blunted cortisol reactivity. ............................................................ 37
Figure 2.5: Flowchart - PsycINFO database search (trauma and cortisol reactivity) ............. 40
Figure 2.6: Flowchart - PubMed database search (trauma and cortisol reactivity) ................ 41
Figure 2.7: Number of articles included from each database ................................................. 42
Figure 2.8: Flowchart - PsycINFO database search (adversity and cortisol reactivity) ......... 45
xiii
Figure 2.9: Flowchart - PubMed database search (adversity and cortisol reactivity) ............. 46
Figure 2.10: Number of articles included from each database ............................................... 47
Figure 2.11: Trauma-Cortisol Hyper-reactivity p-curve. The blue line shows the distribution
of p-values from the data (22 statistically significant results). For example, 41% of the
significant p values were between .01 and .02. The red dotted line shows the expected
distribution of p values if there were no effect. The green dashed line shows the distribution
of p values if an effect existed with studies powered at 33%. N = 22 p values, p-curve is not
significantly right-skewed ....................................................................................................... 50
Figure 2.12: Dropping the lowest and highest p values on the significance of right-skewness
of the full p-curve (top row), half p-curve (middle row), and the test for flatness relative to
33% power (bottom row). The red solid line demonstrates the significance threshold of p =
.05. The filled circle demonstrates the results reported in the text, the open circle
demonstrates the results relative to the significance level if (k) number of p-values were
dropped. .................................................................................................................................. 51
Figure 2.13: Trauma-Cortisol Hypo-reactivity p-curve. The blue line shows the distribution
of p-values from the data (48 statistically significant results). For example, 40% of the
significant p values were below .01. The red dotted line shows the expected distribution o of
p values if there were no effect. The green dashed line shows the distribution of p values if an
effect existed with studies powered at 33%. N = 48 p values, p-curve is significantly right-
skewed..................................................................................................................................... 53
dropped. .................................................................................................................................. 54
Figure 2.15: Adversity-Cortisol Hyper-reactivity p-curve. The blue line shows the
distribution of p-values from the data (15 statistically significant results). For example, 53%
of the significant p values were below .01. The red dotted line shows the expected
xiv
distribution of p values if there were no effect. The green dashed line shows the distribution
of p values if an effect existed with studies powered at 33%. N = 15 p values, p-curve is
significantly right-skewed. ...................................................................................................... 55
dropped. .................................................................................................................................. 56
Figure 2.17: Cortisol hypo-reactivity p-curve. The blue line shows the distribution of p-
values from the data (16 statistically significant results). For example, 31% of the significant
p values were between .01 and .02. The red dotted line shows the expected distribution of p
values if there were no effect. The green dashed line shows the distribution of p values if an
effect existed with studies powered at 33%. N = 16 p values, p-curve is not significantly
right-skewed. ........................................................................................................................... 57
dropped. .................................................................................................................................. 58
Figure 3.1: Flowchart - Early-life adversity and dopamine function.................................... 157
Figure 4.1: Associations Between Adversity and Reward-Based Learning and Decision-
Making. (A) Adversity is positively correlated with reward-related learning r(38) = 0.547, p
< .05(B) Adversity is not significantly correlated with loss-related learning r(38) = 0.189, p =
.242 (C) Adversity is positively correlated with impulsive decision-making r(38) = 0.35, p <
.05.......................................................................................................................................... 188
xv
Figure 4.2: Observed choice to gain-pair and loss-pair stimuli. Over the course of the trials,
participants learned to select the stimulus that more frequently results in a reward (pink) and
avoid the stimulus that more frequently results in a loss (blue) ........................................... 193
Figure 4.3: Associations Between Adversity And Reward-Related and Loss-Related

Learning. (A) Adversity was positively correlated with gain-pair accuracy (pink) r(24) =
0.39, p < 0.05, but not loss-pair accuracy, r(24) = 0.25, p =.218 (blue). (B) Adversity was
positively correlated with learning rate to gains (pink) r(24) = 0.40, p < 0.05, but not learning
rate to losses r(24) = 0.1, p = .627 (blue). (C and D) The association between adversity and
gain-pair accuracy was fully mediated by learning rate to gains. ACME = Average causal
mediation effect; ADE = Average direct effect. ................................................................... 194
Figure 4.4: Neuroimaging Results. (A and B) Neurosynth meta-analysis of 671 studies that
included the word “reward”. Both the VS and vmPFC are preferentially related to the term
“reward”. Based on the meta-analysis ROIs were created for both the VS [-12,10,-9; 12, 10,
-9] as well as the vmPFC [2,62,-10; -2,62,-10], visible in blue. (C and D) Reward-relate
activity correlated positively with learning rate to gains in the VS r(24) = 0.54, p < 0.05, and
vmPFC r(24) = 0.47, p < 0.05............................................................................................... 195
Figure 4.5: Mediation Analysis. (A and B) Reward-related activity in the VS partially

mediated the relationship between adversity and learning rate to gains. ACME = Average
causal mediation effect; ADE = Average direct effect. ........................................................ 196
xvi
List of Appendices (where applicable)
Appendix A: Supplementary Results for Chapter 2 ............................................................. 227
Appendix B: Supplementary Table for Chapter 3 ................................................................ 236
Appendix C: Supplementary Materials for Chapter 4 .......................................................... 249
xvii
1
Chapter 1
1 General Introduction
1.1 The long-term impact of early-life adversity
Accumulating evidence demonstrates that adversity experienced early in life has
long-lasting consequences for health and disease (Anda et al., 2006; Chapman et al.,
2004; Dube et al., 2003; Edwards et al., 2003; Felitti et al., 1998; Shonkoff, Boyce, &
McEwen, 2009; Slavich & Irwin, 2014). Seminal work by Vincent Felitti and colleagues
(1998) showed that there is a strong dose-response relationship between exposure to
adversity in childhood and the incidence of various diseases in adulthood, including
cardiovascular disease, lung disease, and cancer. To further demonstrate the long-term
impact of adversity on health and well-being, in a sample of over 17,000 participants, a
graded relationship was found between adverse childhood experiences and 18 different
outcomes across multiple domains. Specifically, there was a dose-response relationship
between adverse childhood experiences (ACE) scores and affective, somatic, substance
abuse, memory, sexual, and aggression-related outcomes (Anda et al., 2006). Additional
studies have revealed that the number of adverse childhood experiences are related to
depressive disorders (Chapman et al., 2004; Slavich & Irwin, 2014), mental health scores
(Edwards et al., 2003), and the risk of drug initiation and addiction (Dube et al., 2003). In
terms of the impact of adverse life-experiences on psychological health and well-being,
evidence suggests that exposure to adversities is considered to be the strongest proximal
risk factor for the development of depressive disorders (Slavich & Irwin, 2014). Indeed,
depressed individuals have a 2.5-fold greater likelihood of having experienced a major
life event prior to the onset of depression relative to non-depressed controls (Monroe,
Slavich, & Georgiades, 2009; Slavich & Irwin, 2014). To determine whether adversities
are causally linked to health-related outcomes, Bradford Hill criteria for causation was
used in a systematic review and meta-analysis examining the health-consequences of
exposure to physical abuse, emotional abuse, and neglect (Norman et al., 2012). The
authors concluded that there is indeed a causal relationship between childhood
maltreatment and mental disorders, drug use, suicide attempts, sexually transmitted
2
infections, and risky sexual behaviour (Norman et al., 2012). Overall then, overwhelming
evidence using various methodologies in large samples of participants demonstrates that
adversities experienced early in childhood have a strong impact on various health-related
outcomes. These findings suggest that early-life adversities may become “biologically
embedded” in the developing individual (Berens, Jensen, & Nelson, 2017; Hertzman et
al., 1999) and emphasize the plasticity and sensitivity of developing organisms to the
quality of the early-life environment.
1.2 Adaptive phenotypic plasticity

Phenotypic plasticity is defined as the tendency of a single genotype to give rise
to multiple phenotypes in response to varying environmental conditions (Fusco &
Minelli, 2010; Kelly, Peanhuis, & Stoehr, 2011; Sahoo, Subbalakshmi, & Jolly, 2020).
Phenotypic plasticity can be adaptive in that it allows an organism to develop a
phenotype in response to the conditions of the local environment, thereby allowing
organisms to cope with environmental variation (Snell-Rood, 2012). Contextual plasticity
refers to the extent to which a phenotype varies as an immediate response to variation in
the environment (Stamps, 2016). Developmental plasticity, on the other hand, refers to
the extent to which an individual’s present phenotype is affected by past experiences
(Stamps, 2016). Thus, in developmentally plastic individuals, the neural and hormonal
state of the organism at the present time can be affected by the quality of the early-life
environment (Stamps et al., 2016).
The obvious advantage of plasticity is that it allows organisms to alter their

phenotypes in response to the demands of the external environment; however, phenotypic
plasticity is a double-edged sword in that it can come with substantial costs to the
organism. The costs of plasticity include the energetic costs associated with sensory and
regulatory mechanisms of plasticity, the production costs of environmentally induced
structures and processes, the cost of acquiring information from the environment, and the
phenotypic “imprecision” related to developing a phenotype based on experience (Auld,
Agrawal, & Relyea, 2010; Belsky & Pluess, 2013; DeWitt, Sih, & Wilson, 1998)). For
example, modification of a phenotype based on experiences very early in development
may result in a “mismatch” between the developed phenotype and the environment that
3
the organism finds itself in later in life (Belsky & Pluess, 2013). Taken together then,
these considerations of the costs associated with phenotypic plasticity underscore the
notion that developmental malleability should not be considered as an absolute advantage
to all developing organisms (Belsky & Pluess, 2013).
The diathesis-stress perspective suggests that certain individuals are more

vulnerable to the negative effects of unstable or adverse environments. According to the
diathesis-stress model, sensitivity is viewed as vulnerability for developing maladaptive
outcomes when exposed to adversity (Monroe & Simons, 1991; Pluess, 2015). In
contrast to the diathesis-stress model, Belsky and Pluess (2009) propose that individuals
with greater phenotypic plasticity are more sensitive to both the negative consequences of
adverse environments and the beneficial effects of supportive and enriching environments
(Belsky & Pluess, 2009). Furthermore, this model suggests that exposure to adverse life
events does not “dysregulate” development, but rather results in responses that were once
biologically adaptive under conditions of adversity (Belsky & Pluess, 2009). Critically,
“adaptive responses” in this context refers to behaviours that increase reproductive
fitness; however, these behaviours may still constitute high-risk strategies that jeopardize
health (Belsky & Pluess, 2009; Shonkoff et al., 2009). Indeed, life-history theory
demonstrates that after exposure to adverse early-life experiences, females may mature
early in order to increase reproductive fitness. However, early sexual maturation carries
numerous risks to the individual including early sexual behavior, early first birth,
increased risk of sexually transmitted infections, and breast cancer (Belsky & Pluess,
2013; Ellis, 2004). This observation demonstrates that environmental experiences can
regulate the rate of development to serve reproductive fitness goals, albeit at a cost to the
individual in terms of physical health (Belsky & Pluess, 2013).
1.2.1 Individual differences in phenotypic plasticity

Current evolutionary and developmental theory suggests that there are significant
individual differences in plasticity, and plasticity should be considered as a phenotype in
its own right, with some individuals being more susceptible to environmental variation,
and others being less malleable (Belsky & Pluess, 2013; but see Stamps, 2016).
Consistent with this idea, Belsky’s differential susceptibility model (DST) proposes that
4
individuals differ in their sensitivity to the environment, with some being more and others
being less susceptible to both the negative and positive factors in the environment
(Belsky, 1997; Belsky, Bakermans-Kranenburg, & Van IJzendoorn, 2007). According to
the DST, individual differences in susceptibility reflect two alternative developmental
strategies: the plastic strategy results in adaptation to the environment, whereas the fixed
strategy reflects relative inertia to variation in the environment (Pluess, 2015). The
developmentally plastic strategy has the advantage of allowing organisms to modify their
phenotypes in response to signals from the local environment, although this comes at a
cost in terms of resources and energy expenditure (Belsky & Pluess, 2013). The fixed
strategy, on the other hand, reflects a highly canalized developmental strategy, where
there is a trade-off between the potential benefit of thriving in more than one environment
against the costs of developmental plasticity (Belsky & Pluess, 2013). Initially,
individual differences in susceptibility to environmental factors were thought to reflect
variation in genotypic differences (i.e., the expression or lack thereof of “plasticity
genes”); however, more recently, evidence suggests that high susceptibility may also
develop in response to prenatal and postnatal environmental factors themselves (Pluess,
2015). That is, heightened sensitivity to the environment may be induced by
environmental factors and do not necessarily reflect genetically-derived differences
(Pluess, 2015).
1.3 The role of the central nervous system

Heightened environmental sensitivity is likely associated with changes in the
structure and function of the central nervous system. These changes in the central nervous
system allow environmental events to have a greater impact on the developing organism
(Pluess, 2015). The sensitivity of the nervous system to environmental events is reflected
in physiological reactivity (e.g., stress reactivity), and in behavioural outcomes in
response to environmental variation and stress (Pluess, 2015). Indeed, the brain is the
central organ that interprets and responds to variation in environmental signals (McEwen,
2008). Furthermore, the brain is the primary target of stress in that it changes both
structurally and functionally in response to stressors (McEwen, 2008). When faced with a
stressor or adversity, the brain is the key organ that determines what is threatening or
5
stressful, and initiates behavioural and physiological responses aimed at coping with and
overcoming the stressor (McEwen, 2008). Therefore, when assessing phenotypic
plasticity in response to environmental variation, it is critical to consider the plasticity of
regions within the central nervous system that play a major role in interpreting and
responding to stressful experiences. Indeed, the stress-response systems including the
hypothalamic-pituitary adrenal (HPA) axis have long been considered to be plastic in the
face of variation in the quality of the early-life environment (Liu et al., 1997; Meaney,
2010). More recently, the mesolimbic dopamine pathway has also been hypothesized to
exhibit plasticity in response to early-life experiences (Gatzke-Kopp, 2011).
1.3.1 The plasticity of the HPA axis

In examining the link between exposure to early-life adversity and subsequent health-
related outcomes, a consideration of plastic neural circuitry that exhibits sensitivity to
stressful experiences is necessary. Of particular interest in this regard, are systems that
are dense with glucocorticoid receptors, including the prefrontal cortex, hippocampus,
amygdala, and notably, the HPA axis (Arnsten, 2011; Berens et al., 2017; Liu et al.,
1997; Tottenham & Galvan, 2016). The HPA axis consists of a negative feedback loop
between the hypothalamus, pituitary gland, and adrenal glands (Berens et al., 2017; Del
Rey, Chrousos, & Besedovsky, 2008). The primary function of the HPA axis is the
coordinated neuroendocrine release of glucocorticoids –cortisol in humans and
corticosterone in animals—in response to stressful situations. Subsequently,
glucocorticoid production enables increased availability of energy resources (e.g.,
glucose) to enable “fight-or-flight” mechanisms in response to acute stressors (Del Rey et
al., 2008). Therefore, HPA axis activation is an adaptive response under acutely stressful
conditions in that it allows the organism sufficient energy resources to overcome the
immediate stressor (Boyce & Ellis, 2005; Selye, 1950). After chronically stressful
conditions however, the set-point of the HPA axis may become permanently altered (See
Chapter 2). The enduring change in the HPA axis as a function of exposure to adversity
demonstrates a case of “developmental plasticity”; here, adverse early-life experiences
alter the neuroendocrine response of the HPA axis later in life.
6
Therefore, when exploring the association between early-life adversity and

developmental plasticity within the central nervous system, special consideration of the
HPA axis is required for the following reasons:
1.) The primary functional role of the HPA axis is to release glucocorticoids in
response to acute stressors.
2.) Rodent literature (Liu et al., 1997; Meaney 2010) suggests that the HPA axis is a
highly plastic system that is developmentally sensitive to variation in the quality
of the early-life environment. Consistent with the animal literature, human
association studies also demonstrate developmental plasticity within the HPA axis
as a function of exposure to early-life adversity (See Chapter 2). What remains
outstanding however, is not whether the HPA axis is indeed sensitive to
environmental influences, but rather, the direction of change—in other words,
does the HPA axis become hyper-or hypo-reactive after exposure to early-life
adversity? Empirical evidence to date remains mixed and has found evidence for
both hyper and hypo-reactivity of the HPA axis after exposure to early-life
adversity (See Chapter 2 – Tables 2.1-2.4).
1.3.2 The plasticity of the dopamine system

The mesolimbic dopamine pathway consists of a complex circuitry involving the
amygdala, hippocampus, medial prefrontal cortex (mPFC), nucleus accumbens, ventral
tegmental area (VTA) and ventral pallidum; the flow of information between these
regions is modulated by dopaminergic neurotransmission (Haber & Knutson, 2010;
Pierce & Kumaresan, 2006). The functions of the mesolimbic dopamine pathway are vast
and include reinforcement learning, decision-making, reward-processing, and the
regulation of motor control (Wise, 2004). Like the HPA axis, Gatzke-Kopp (2011)
proposed that the mesolimbic dopamine pathway also exhibits developmental plasticity in
response to variation in the quality of the early-life environment. In fact, extensive animal
literature exists that demonstrates enduring changes within the mesolimbic dopamine
pathway as a function of exposure to various stressors, including physical stressors (e.g.,
footshock and tailpinch) as well as social stressors (e.g., isolate rearing, social defeat by
7
an aggressive intruder; Haney et al., 1995; LeSage, Stafford, & Glowa, 1999; Piazza &
Le Moal, 1998; Schenk et al., 1987; Sinha 2000).
Moreover, stress, glucocorticoids, and the mesolimbic dopamine pathway have long
been proposed to form a physiological chain—this link suggests that variation in HPA
axis function may have downstream implications of dopaminergic transmission (Piazza &
Le Moal, 1996). While empirical human evidence linking early-life adversity to
dopamine levels is limited (See Chapter 3), several considerations listed below
demonstrate that the mesolimbic dopamine pathway may indeed exhibit developmental
plasticity to early-life experiences in much the same way as the HPA axis.
1.) Risk behaviours involving the mesolimbic dopamine pathway, including
impulsivity, risky sexual behaviours, the consumption of high-fat, high-sugar
foods, and substance abuse problems appear to be at the center of mechanisms
linking early-life adversity to later negative health outcomes (Duffy, McLaughlin,
& Green, 2018; Felitti et al., 1998; Garrido, Weiler, & Taussig, 2018; Stevens et
al., 2011). If behaviours involving the mesolimbic dopamine pathway mediate the
well-documented association between adversity and negative health outcomes,
then it suggests that the mesolimbic dopamine pathway must itself be sensitive to
adverse early-life experiences.
2.) There is a physiological chain linking stress, the HPA axis, and the mesolimbic
dopamine pathway. As discussed above, the HPA axis is involved in
glucocorticoid release in response to acutely stressful conditions. Interestingly,
empirical evidence demonstrates that a substantial number of dopamine neurons
express glucocorticoid receptors (Harfstrand et al., 1986; Hensleigh & Pritchard,
2013). The presence of glucocorticoid receptors on dopamine neurons provides a
biological mechanism that explains the link between stress, the HPA axis, and the
mesolimbic dopamine pathway. Moreover, in animal models, glucocorticoid
administration potentiates dopamine release, while administration of agents that
block the function of glucocorticoids results in blunted dopamine release
(Campbell & Carroll, 2001; Goeders & Clampitt, 2002; See Marinelli & Piazza,
2002 for a review).
8
3.) Consistent with the notion that the mesolimbic dopamine pathway is plastic to
early-life adversity, rodent models confirm that exposure to both acute and
chronic stressors are related to variation in dopamine levels within the mesolimbic
pathway (See Douma & de Kloet, 2020; Holly & Miczek, 2016 for reviews).
Taken together then, it appears that both the HPA axis and the mesolimbic dopamine
pathway demonstrate developmental plasticity in response to variation in environmental
quality. What remains to be determined however, is whether adverse life-experiences are
associated with heightened or blunted cortisol and dopamine levels.
1.4 Distinguishing adversity and trauma

The human literature reporting associations between adversity and cortisol reactivity
is highly mixed, with evidence available in support of both heightened and blunted
cortisol levels after exposure to adverse life-experiences (See Chapter 2 – Tables 2.1-
2.4). Likewise, human PET studies investigating the relationship between adverse life-
experiences and dopamine levels also provide conflicting results, with some studies
reporting heightened and others reporting blunted dopamine levels after exposure to
adversity (See Chapter 3 – Table 3.1). Therefore, while it is apparent that both the HPA
axis and the mesolimbic dopamine pathway are plastic to environmental adversity, the
nature of the association between adversity and cortisol/dopamine levels remains unclear.
Perhaps one factor that has resulted in the mixed empirical findings is unclear definitions
of what types of events constitute adversity versus trauma (Krupnik, 2019). Currently,
there are no clear conceptual boundaries in the available literature to define and
differentiate “adversity” versus “trauma”. For example, in some cases trauma is defined
narrowly using Criterion A of trauma-related disorders in the DSM-V, while in other
cases it is defined so broadly that virtually any adverse event ranging from extreme
trauma (e.g., sexual abuse, maltreatment) to moderate forms of adversity (e.g., low SES,
death of a family pet) are all considered to be “adverse life events” (Krupnik, 2019). To
reduce variability across studies owing to broad conceptual definitions of “adverse life
events”, Krupnik (2019) suggests that “adversity” and “trauma” should be differentiated
based on both the severity of stressors and their tendency to overwhelm self-regulatory
9
functions. According to this model then, stressors become traumatic when their severity
overwhelms self-regulatory processes (Krupnik, 2019).
1.5 Aim of current thesis and summary of findings

The overall aim of the current thesis is to examine whether and how the HPA axis
and the mesolimbic dopamine system exhibit developmental plasticity after exposure to
adverse life-experiences. Critically, in this thesis, I will use a narrow definition to
differentiate between adversity and trauma in examining the available literature. Here,
trauma will be defined using Criterion A of trauma-related disorders using DSM-V.
These events include exposure to death, threatened death, actual or threatened serious
injury, and actual or threatened sexual violence (e.g., sexual abuse, war-related trauma,
physical violence). In contrast, adversity is defined as negative life events that do not fit
the criteria for traumatic events, examples include low SES, changing residences, and
work-related stress. The overall goal of the current thesis is to use a variety of methods to
examine whether exposure to moderate adversity versus extreme trauma results in
differential calibration of the HPA axis and the mesolimbic dopamine pathway.
Specifically, I determine whether differentiating between adversity and trauma helps
resolve the mixed empirical findings. The main findings of this thesis demonstrate that
after exposure to moderate levels of adversity, both the HPA axis and mesolimbic
dopamine pathway become hyper-reactive (i.e., heightened cortisol and dopamine
reactivity). In contrast, after exposure to severe trauma, both the HPA axis and
mesolimbic dopamine pathway become hypo-reactive (i.e., blunted cortisol and
dopamine reactivity). Based on these results, I propose the Stress-response Adaptations
to Variability in Early-life experiences (SAVE) model. The SAVE model is based on the
established link between the HPA axis and mesolimbic dopamine pathway, and predicts
an inverted-U shaped relationship between severity of adversity and both HPA axis and
dopamine reactivity.
1.5.1 Chapter 2 – The differential calibration of the HPA axis as a

function of trauma versus adversity: A systematic review and
p-curve meta-analyses
10
1.5.1.1 Aims and objectives

Although there is substantial evidence suggesting that early-life adversity is
associated with altered HPA axis function, in humans, the direction of the association
remains unclear. Some studies have found that adversity is associated with heightened
cortisol reactivity, while others have reported blunted cortisol reactivity. Recent meta-
analyses have also yielded contradictory findings (Bunea, Szentagotai-Tatar, & Miu,
2017; Fogelman & Canli, 2018). Therefore, the literature on how adverse life experiences
are related to HPA axis reactivity remains mixed. P-curve meta-analyses can determine
which set of contradictory findings are more likely to be correct in that they have not
been subject to p-hacking—thus, p-curve is a powerful tool for resolving discrepant
findings across studies. To resolve the contradictory results in the literature, we
conducted a comprehensive review and p-curve meta-analysis. We included studies
reporting associations between negative life experiences and cortisol reactivity.
Importantly, we made a distinction between “trauma” (e.g., sexual abuse,
institutionalization, maltreatment) versus “adversity” (e.g., low SES, parental conflict,
workplace stress).
1.5.1.2 Main findings and conclusions

In Study 1, I reviewed associations between severe trauma and cortisol reactivity,
and found that the literature reporting a relationship between trauma and blunted cortisol
reactivity contains evidential value. In contrast, in Study 2, I reviewed associations
between moderate levels of adversity and cortisol reactivity, and I found that the
literature reporting heightened cortisol reactivity contains evidential value. Therefore, the
severity of negative life experiences (i.e., trauma versus adversity) appears to be a critical
factor in determining the direction of HPA axis regulation. These results provide
evidence in support of theoretical models predicting an inverted U-shaped association
between severity of adversity and HPA axis reactivity, and they highlight the importance
of distinguishing between trauma and adversity (Hosseini-Kamkar, Lowe, & Morton,
2021).
11
1.5.2 Chapter 3 – Stress, glucocorticoids, and the mesolimbic

dopamine pathway: How early-life experiences become
biologically embedded into the dopamine system

The mesolimbic dopamine pathway is hypothesized to be developmentally plastic in
response to variation in environmental quality (Gatzke-Kopp, 2011). Furthermore, the
link between adversity and health-related outcomes is partially explained by variation in
health-risk behaviours involving the dopaminergic system. Therefore, an understanding
of the underlying biological mechanisms that explain the links between stress, the HPA
axis, and the mesolimbic dopamine pathway is warranted. This chapter has two primary
objectives:
1.) To review animal models exploring the associations between stress, HPA axis
reactivity, and mesolimbic dopamine function. The primary goal of this section is
to explore the underlying biological mechanisms explaining the link between
stress, the HPA axis, and the mesolimbic dopamine pathway.
2.) To systematically review the available human literature using PET imaging to
examine the association between early-life adversity and mesolimbic dopamine
reactivity. Similar to the methods in Chapter 2, I differentiate between “adversity”
and “trauma” using DSM-5 criteria when examining the literature.

Evidence obtained from the systematic review of human PET imaging studies
assessing the relationship between adverse life-experiences and dopamine reactivity
demonstrates that moderate adversity is related to a hyperdopaminergic profile, while
trauma is related to a hypodopaminergic profile. Similar to the findings in Chapter 2,
these results provide support for an inverted-U shaped relationship between severity of
adversity and dopamine reactivity.
12
1.5.3 Chapter 4 – Ventral striatal reactivity links adversity and

reward processing in children
The main purpose of Chapter 4 was to conduct an empirical study investigating
the link between exposure to adversity and dopamine-mediated behavoiurs including
reward-related learning and impulsivity. Furthermore, I conducted an fMRI study to
determine whether adversity is related to variation in BOLD responses within regions of
the mesolimbic dopamine pathway, including the ventral striatum (Kamkar et al., 2017).

The results showed that adversity is associated with potentiated learning from
rewards and impulsivity in children 9-12 years of age. In other words, children who
experienced more adversity were more impulsive and learned faster from rewards,
notably, learning from negative outcomes was unrelated to adversity. Moreover, using
fMRI we showed that the association between adversity and reward processing is
partially mediated by ventral striatal reactivity to rewards. While I did not measure
dopamine levels directly, these findings demonstrate that indeed, the regions within the
mesolimbic dopamine pathway exhibit functional sensitivity to experiences of adversity
early in life (Kamkar et al., 2017).
1.6 Adaptive plasticity: Stress-response Adaptations to

Variability in Early-life experiences (SAVE) model
Based on the findings of this thesis and the well-established link between stress,
HPA axis reactivity, and mesolimbic dopamine reactivity, I propose the SAVE model.
The SAVE model predicts an inverted-U shaped relationship between severity of
adversity and both HPA axis and dopamine reactivity. The SAVE model is based on the
biological role that glucocorticoid release plays in glucose metabolism, and the tendency
of the body to “save” resources unless it is necessary and efficient to utilize them. When
exposed to stressful experiences, glucocorticoids are produced and result in increased
availability of energy resources to facilitate “fight or flight” responses (Del Rey et al,
2008). Under chronically stressful or traumatic conditions, “fight or flight” may not be
13
feasible, and thus, the organism’s physiology may adapt in a manner that does not waste
energy resources. If the organism perceives the situation as completely inescapable, there
is little utility in releasing glucocorticoids and mobilizing energy resources unnecessarily
(Cabib & Puglisi-Allegra, 2012). However, under moderate levels of adversity, it may be
adaptive for the organism to mobilize energy resources to facilitate coping with the
stressor –this is reflected in heightened HPA axis and dopamine reactivity. Thus, the
adaptive significance of the SAVE model rests under the assumption that it is not adaptive
for organisms to waste energy resources.
In response to environmental stressors, two patterns of behaviour consistently emerge:

1.) Active coping strategies: some members tend to be bold, aggressive and
impulsive
2.) Passive coping strategies: other members appear to avoid stressors, behave
fearfully and cautiously (Cabib & Puglisi-Allegra, 2012).
According to DST theory, these individual differences in coping strategies may be the
result of differences in susceptibility to environmental stressors.
In contrast to the SAVE model, the biological sensitivity to context model (BSC)
proposed by Boyce and Ellis (2005) presents a U-shaped function to describe the
relationship between severity of adversity and HPA axis reactivity. According to the
BSC model, highly stress-reactive phenotypes emerge under both extremely stressful and
highly protected environments (Boyce & Ellis, 2005). Indeed, the biological sensitivity to
context model (BSC) claims that physiological reactivity is a plasticity factor, and more
physiologically reactive individuals are said to be more susceptible to both positive and
negative factors in their environment (Boyce & Ellis, 2005). Therefore, according to the
BSC model, high plasticity phenotypes emerge under both very protected and highly
stressful conditions. The SAVE model, directly challenges this notion—if biological
reactivity is indeed a plasticity factor, then according to empirical evidence and the SAVE
model, high plasticity phenotypes emerge under moderately adverse environmental
conditions. Once again, it appears adaptive to mobilize energy resources as indicated by
heightened biological reactivity under moderately adverse conditions where sensitivity to
14
environmental cues and subsequent resource mobilization may be fruitful. Under

protected and extremely traumatic conditions, where escape and avoidance are unlikely,
it would not be adaptive for organisms to waste energy resources associated with the
costs of plasticity, including attempts at obtaining information from the environment and
the expenditure biological resources required to cope with the stressor.
Active coping strategies require the release of dopamine to face the stressor—
reflected in heightened aggressive and impulsive behaviours (Cabib & Puglisi-Allegra,
2012). In contrast, passive coping strategies appear to be related to blunted release of
glucocorticoids and dopamine (Cabib & Puglisi-Allegra, 2012). According to the SAVE
model then, we expect individuals to exhibit active coping strategies in response to
moderately adverse conditions, and passive coping strategies in response to extremely
stressful or traumatic conditions. According to DST theory, these individual differences
in coping strategies may be the result of differences in susceptibility to environmental
stressors (Pluess, 2015). By extension then, moderately adverse life-experiences are
associated with heightened biological reactivity (e.g., cortisol and dopamine reactivity)
and reflect heightened plasticity to environmental signals. In contrast, extreme trauma is
associated with blunted biological reactivity (e.g., cortisol and dopamine reactivity) and
reflect reduced plasticity to environmental signals. According to the SAVE model, these
differences in plasticity and biological reactivity reflect adaptive responses geared at
saving biological resources, thus organisms expend the necessary energy required for
active coping with stressors only when environmental conditions signal that the stressor
is indeed escapable/avoidable.
15
1.7 References
Anda, R. F., Felitti, V. J., Bremner, J. D., Walker, J. D., Whitfield, Ch., Perry, B. D.,
Dube, Sh. R., & Giles, W. H. (2006). The enduring effects of abuse and related
adverse experiences in childhood: A convergence of evidence from neurobiology
and epidemiology. European Archives of Psychiatry and Clinical Neuroscience,
256(3), 174–186. https://doi.org/10.1007/s00406-005-0624-4
Arnsten, A. F. T. (2011). Prefrontal cortical network connections: Key site of
vulnerability in stress and schizophrenia. International Journal of Developmental
Neuroscience : The Official Journal of the International Society for
Developmental Neuroscience, 29(3), 215–223.
https://doi.org/10.1016/j.ijdevneu.2011.02.006
Auld, J. R., Agrawal, A. A., & Relyea, R. A. (2010). Re-evaluating the costs and limits of
adaptive phenotypic plasticity. Proceedings of the Royal Society B: Biological
Sciences, 277(1681), 503–511. https://doi.org/10.1098/rspb.2009.1355
Belsky, J, Jonassaint, C., Pluess, M., Stanton, M., Brummett, B., & Williams, R. (2009).
Vulnerability genes or plasticity genes?. Molecular Psychiatry, 14(8), 746–754.
https://doi.org/10.1038/mp.2009.44
Belsky, Jay, & Pluess, M. (2013). Beyond risk, resilience, and dysregulation: Phenotypic
plasticity and human development. Development and Psychopathology, 25(4 Pt
2), 1243–1261. https://doi.org/10.1017/s095457941300059x
Belsky, Jay. (1997). Variation in Susceptibility to Environmental Influence: An
Evolutionary Argument. Psychological Inquiry, 8(3), 182–186.
https://doi.org/10.1207/s15327965pli0803_3
Belsky, Jay, & Pluess, M. (2009). The Nature (and Nurture?) of Plasticity in Early
Human Development. Perspectives on Psychological Science, 4(4), 345–351.
https://doi.org/10.1111/j.1745-6924.2009.01136.x
Berens, A. E., Jensen, S. K. G., & Nelson, C. A. (2017). Biological embedding of
childhood adversity: From physiological mechanisms to clinical implications.
BMC Medicine, 15(1), 135. https://doi.org/10.1186/s12916-017-0895-4
16
Boyce, W. T., & Ellis, B. J. (2005). Biological sensitivity to context: I. An evolutionary–

developmental theory of the origins and functions of stress reactivity.
Development and Psychopathology, 17(02).
https://doi.org/10.1017/S0954579405050145
Bunea, I. M., Szentágotai-Tătar, A., & Miu, A. C. (2017). Early-life adversity and
cortisol response to social stress: A meta-analysis. Translational Psychiatry,
7(12), 1274. https://doi.org/10.1038/s41398-017-0032-3
Cabib, S., & Puglisi-Allegra, S. (2012). The mesoaccumbens dopamine in coping with
stress. Neuroscience and Biobehavioral Reviews, 36(1), 79–89.
https://doi.org/10.1016/j.neubiorev.2011.04.012
Campbell, U. C., & Carroll, M. E. (2001). Effects of ketoconazole on the acquisition of
intravenous cocaine self-administration under different feeding conditions in rats.
Psychopharmacology, 154(3), 311–318. https://doi.org/10.1007/s002130000627
Chapman, D. P., Whitfield, C. L., Felitti, V. J., Dube, S. R., Edwards, V. J., & Anda, R.
F. (2004). Adverse childhood experiences and the risk of depressive disorders in
adulthood. Journal of Affective Disorders, 82(2), 217–225.
https://doi.org/10.1016/j.jad.2003.12.013
Del Rey, A., Chrousos, G., & Besedovsky, H. (2008). The hypothalamus-pituitary-
adrenal axis. Elsevier.
Dewitt, T., Sih, A., & Wilson, D. S. (1998). DeWitt TJ, Sih A, Wilson DS. Costs and
limits of phenotypic plasticity. Trends Evol Ecol 13: 77-81. Trends in Ecology &
Evolution, 13, 77–81. https://doi.org/10.1016/S0169-5347(97)01274-3
Douma, E. H., & de Kloet, E. R. (2020). Stress-induced plasticity and functioning of
ventral tegmental dopamine neurons. Neuroscience and Biobehavioral Reviews,
108(oa7, 7806090), 48–77. https://doi.org/10.1016/j.neubiorev.2019.10.015
Dube, S. R., Felitti, V. J., Dong, M., Chapman, D. P., Giles, W. H., & Anda, R. F. (2003).
Childhood Abuse, Neglect, and Household Dysfunction and the Risk of Illicit
Drug Use: The Adverse Childhood Experiences Study. Pediatrics, 111(3), 564–
572. https://doi.org/10.1542/peds.111.3.564
17
Duffy, K. A., McLaughlin, K. A., & Green, P. A. (2018). Early life adversity and health-
risk behaviors: Proposed psychological and neural mechanisms. Annals of the
New York Academy of Sciences, 1428(1), 151–169.
https://doi.org/10.1111/nyas.13928
Edwards, V. J., Holden, G. W., Felitti, V. J., & Anda, R. F. (2003). Relationship Between
Multiple Forms of Childhood Maltreatment and Adult Mental Health in
Community Respondents: Results From the Adverse Childhood Experiences
Study. American Journal of Psychiatry, 160(8), 1453–1460.
https://doi.org/10.1176/appi.ajp.160.8.1453
Ellis, B. (2004). Timing of Pubertal Maturation in Girls: An Integrated Life History
Approach. Psychological Bulletin, 130, 920–958. https://doi.org/10.1037/0033-
2909.130.6.920
Felitti, V. J., Anda, R. F., Nordenberg, D., Williamson, D. F., Spitz, A. M., Edwards, V.,
Koss, M. P., & Marks, J. S. (1998). Relationship of Childhood Abuse and
Household Dysfunction to Many of the Leading Causes of Death in Adults.
American Journal of Preventive Medicine, 14(4), 245–258.
https://doi.org/10.1016/S0749-3797(98)00017-8
Fogelman, N., & Canli, T. (2018). Early life stress and cortisol: A meta-analysis.
Hormones and Behavior, 98, 63–76. https://doi.org/10.1016/j.yhbeh.2017.12.014
Fusco, G., & Minelli, A. (2010). Phenotypic plasticity in development and evolution:
Facts and concepts. Philosophical Transactions of the Royal Society B: Biological
Sciences, 365(1540), 547–556. https://doi.org/10.1098/rstb.2009.0267
Garrido, E. F., Weiler, L. M., & Taussig, H. N. (2018). Adverse Childhood Experiences
and Health-Risk Behaviors in Vulnerable Early Adolescents. The Journal of Early
Adolescence, 38(5), 661–680. https://doi.org/10.1177/0272431616687671
Gatzke-Kopp, L. M. (2011). The canary in the coalmine: The sensitivity of mesolimbic
dopamine to environmental adversity during development. Neuroscience and
Biobehavioral Reviews, 35(3), 794–803.
Goeders, N., & Clampitt, D. (2002). Potential role for the hypothalamo-pituitary-adrenal
axis in the conditioned reinforcer-induced reinstatement of extinguished cocaine
18
seeking in rats. Psychopharmacology, 161(3), 222–232.

https://doi.org/10.1007/s00213-002-1007-4
Haber, S. N., & Knutson, B. (2010). The Reward Circuit: Linking Primate Anatomy and
Human Imaging. Neuropsychopharmacology, 35(1), 4–26.
https://doi.org/10.1038/npp.2009.129
Haney, M., Maccari, S., Le Moal, M., Simon, H., & Vincenzo Piazza, P. (1995). Social
stress increases the acquisition of cocaine self-administration in male and female
rats. Brain Research, 698(1–2), 46–52. https://doi.org/10.1016/0006-
8993(95)00788-R
Harfstrand, A., Fuxe, K., Cintra, A., Agnati, L. F., Zini, I., Wikstrom, A. C., Okret, S.,
Yu, Z. Y., Goldstein, M., & Steinbusch, H. (1986). Glucocorticoid receptor
immunoreactivity in monoaminergic neurons of rat brain. Proceedings of the
National Academy of Sciences, 83(24), 9779–9783.
https://doi.org/10.1073/pnas.83.24.9779
Hensleigh, E., & Pritchard, L. M. (2013). Glucocorticoid receptor expression and sub-
cellular localization in dopamine neurons of the rat midbrain. Neuroscience
Letters, 556, 191–195. https://doi.org/10.1016/j.neulet.2013.09.067
Holly, E. N., & Miczek, K. A. (2016). Ventral tegmental area dopamine revisited: Effects
of acute and repeated stress. Psychopharmacology, 233(2), 163–186.
https://doi.org/10.1007/s00213-015-4151-3
Hosseini-Kamkar, N., Lowe, C., & Morton, J. B. (2021). The differential calibration of
the HPA axis as a function of trauma versus adversity: A systematic review and p-
curve meta-analyses. Neuroscience & Biobehavioral Reviews, 127, 54–135.
Kamkar, N. H., Lewis, D. J., van den Bos, W., & Morton, J. B. (2017). Ventral striatal
activity links adversity and reward processing in children. Developmental
cognitive neuroscience, 26, 20-27.
Kelly, S. A., Panhuis, T. M., & Stoehr, A. M. (2012). Phenotypic Plasticity: Molecular
Mechanisms and Adaptive Significance. In R. Terjung (Ed.), Comprehensive
Physiology (p. c110008). John Wiley & Sons, Inc.
https://doi.org/10.1002/cphy.c110008
19
Krupnik, V. (2019). Trauma or adversity?. Traumatology, 25(4), 256-

261. https://doi.org/10.1037/trm0000169
LeSage, M. G., Stafford, D., & Glowa, J. R. (1999). Preclinical research on cocaine self-
administration: Environmental determinants and their interaction with
pharmacological treatment. Neuroscience & Biobehavioral Reviews, 23(5), 717–
741. https://doi.org/10.1016/S0149-7634(99)00015-9
Liu, D., Diorio, J., Tannenbaum, B., Caldji, C., Francis, D., Freedman, A., ... & Meaney,
M. J. (1997). Maternal care, hippocampal glucocorticoid receptors, and
hypothalamic-pituitary-adrenal responses to stress. Science, 277(5332), 1659-
1662.
Marinelli, M., & Piazza, P. V. (2002). Interaction between glucocorticoid hormones,
stress and psychostimulant drugs*. European Journal of Neuroscience, 16(3),
387–394. https://doi.org/10.1046/j.1460-9568.2002.02089.x
McEwen, B. S. (2008). Central effects of stress hormones in health and disease:
Understanding the protective and damaging effects of stress and stress mediators.
European Journal of Pharmacology, 583(2–3), 174–185.
https://doi.org/10.1016/j.ejphar.2007.11.071
Meaney, M. J. (2010). Epigenetics and the Biological Definition of Gene × Environment
Interactions. Child Development, 81(1), 41–79. https://doi.org/10.1111/j.1467-
8624.2009.01381.x
Monroe, S., & Simons, A. (1991). Diathesis stress theories in the context of life stress
research: Implications for the depressive disorders. Psychological Bulletin, 110,
406–425. https://doi.org/10.1037//0033-2909.110.3.406
Monroe, S. M., Slavich, G. M., & Georgiades, K. (2009). The social environment and life
stress in depression. Handbook of depression, 2(1), 340-60.
Norman, R. E., Byambaa, M., De, R., Butchart, A., Scott, J., & Vos, T. (2012). The
Long-Term Health Consequences of Child Physical Abuse, Emotional Abuse, and
Neglect: A Systematic Review and Meta-Analysis. PLOS Medicine, 9(11),
e1001349. https://doi.org/10.1371/journal.pmed.1001349
20
Piazza, P. V., & Le Moal, M. (n.d.). Pathophysiological basis of vulnerability to drug

abuse: Role of an interaction between stress, glucocorticoids, and dopaminergic
neurons. Annual Review of Pharmacology and Toxicology, 36(1), 359–378.
Pierce, R. C., & Kumaresan, V. (2006). The mesolimbic dopamine system: The final
common pathway for the reinforcing effect of drugs of abuse? Neuroscience &
Pluess, M. (2015). Individual Differences in Environmental Sensitivity. Child
Development Perspectives, 9(3), 138–143. https://doi.org/10.1111/cdep.12120
Sahoo, S., Subbalakshmi, A. R., & Jolly, M. K. (2020). 1—The fundamentals of
phenotypic plasticity. In H. Levine, M. K. Jolly, P. Kulkarni, & V. Nanjundiah
(Eds.), Phenotypic Switching (pp. 1–21). Academic Press.
https://doi.org/10.1016/B978-0-12-817996-3.00001-3
Selye, H. (1950). Stress and the General Adaptation Syndrome. British Medical Journal,
1(4667), 1383–1392.
Shonkoff, J. P., Boyce, W. T., & McEwen, B. S. (2009). Neuroscience, Molecular
Biology, and the Childhood Roots of Health Disparities: Building a New
Framework for Health Promotion and Disease Prevention. JAMA, 301(21), 2252–
2259. https://doi.org/10.1001/jama.2009.754
Sinha, R., Fuse, T., Aubin, L. R., & O'Malley, S. S. (2000). Psychological stress, drug-
related cues and cocaine craving. Psychopharmacology, 152(2), 140-148.
Slavich, G. M., & Irwin, M. R. (2014). From stress to inflammation and major depressive
disorder: a social signal transduction theory of depression. Psychological
bulletin, 140(3), 774.
Snell-Rood, E. C. (2012). Selective Processes in Development: Implications for the Costs
and Benefits of Phenotypic Plasticity. Integrative and Comparative Biology,
52(1), 31–42. https://doi.org/10.1093/icb/ics067
Stamps, J. A. (2016). Individual differences in behavioural plasticities. Biological
Reviews, 91(2), 534–567. https://doi.org/10.1111/brv.12186
21
Stevens, S. B., Brice, C. S., Ale, C. M., & Morris, T. L. (2011). Examining depression,
anxiety, and foster care placement as predictors of substance use and sexual
activity in adolescents. Journal of Social Service Research, 37(5), 539-554.
Tottenham, N., & Galván, A. (2016). Stress and the adolescent brain: Amygdala-
prefrontal cortex circuitry and ventral striatum as developmental targets.
Neuroscience & Biobehavioral Reviews, 70, 217–227.
Wise, R. A. (2004). Dopamine, learning and motivation. Nature Reviews Neuroscience,
5(6), 483–494. https://doi.org/10.1038/nrn1406
22
Chapter 2
2 The differential calibration of the HPA axis as a function

of trauma versus adversity: A systematic review and p-
curve meta-analyses
2.1 Introduction
Adverse childhood experiences including child maltreatment, neglect, and
community violence have been linked to increased risk of negative health outcomes later
in life. In one seminal study, Felitti et al. (1998) obtained data from 13,494 adults who
completed the adverse childhood experiences (ACE) questionnaire and a standardized
medical evaluation. The authors found a strong dose response relationship between
exposure to abuse or household dysfunction during childhood and risk of disease later in
life, including ischemic heart disease, cancer, and chronic lung diseases (Felitti et al.,
1998). The findings highlighted strong and cumulative negative impacts of adverse early
life experiences on later health outcomes and led to detailed empirical inquiry into how
life experiences become biologically embedded (Hertzman, 1999).
Of particular interest in this regard was understanding how adverse early life
experiences impact “stress-sensitive” systems of the brain such as the hippocampus and
amygdala involved in learning, memory, and emotion regulation (Tottenham & Sheridan,
2010), as well as the prefrontal cortex (PFC), an area critical for decision-making,
executive functioning, and top-down control of limbic regions (Arnsten, 2009; Berens,
Jense, & Nelson, 2017). These regions are dense with glucocorticoid receptors,
suggesting that their development and function might be altered as a consequence of
environmentally-induced alterations in cortisol release. Indeed, rodent models have
demonstrated that offspring of mothers that showed more licking and grooming had
reduced levels of ACTH and corticosterone in response to acute stress (Liu et al., 1997).
This plasticity of the HPA axis to early life programming is possible through epigenetic
modification of glucocorticoid receptor genes (Meaney, 2010). Thus, rodent models have
demonstrated that the HPA axis is a highly plastic biological system with demonstrated
23
sensitivity to variation in the quality of the early-life environment (Liu et al., 1997;
Meaney, 2010).
Although early-life adversity has consistently been shown to predict HPA axis
dysregulation in humans, the nature of the association remains unclear (See Tables 2.1-
2.4 for summaries of the conflicting literature). Several factors, including differences in
how HPA axis reactivity is measured and differences in how adverse early life
experiences are operationalized, potentially contribute to discrepant findings between
studies. However, even when explicit criteria are used to define trauma versus adversity,
and outcomes are confined to selected measures (e.g., cortisol reactivity to a stress
induction), available evidence remains conflicting. For example, studies of the
association between traumatic life experiences (as defined by DSM-V criteria) and
cortisol reactivity to a laboratory stressor report conflicting findings, with some studies
reporting cortisol hyper-reactivity and others reporting cortisol hypo-reactivity (See
Tables 2.1-2.2). One powerful method for resolving such discrepancies is p-curve meta-
analyses (Simonsohn et al., 2014a, 2015); therefore, we have conducted a comprehensive
review and used p-curve meta-analytic techniques to examine associations between
adversity and cortisol reactivity.
2.1.1 The stress response system

Environmental events that signal threat to an organism’s survival result in a stress
response, defined as a set of neural, endocrine, and behavioural responses that regulate
homeostasis within metabolic and physiological functions. This coordinated
physiological response is made possible by the stress response system and leads to an
increase in heart rate and blood pressure, and the metabolic release of glucose and
nutrients, better enabling the organism to overcome the acute stressor (Selye, 1950;
Boyce & Ellis, 2005). The stress response system is composed of two anatomically
separate but functionally interconnected circuits – the corticotropin-releasing factor
(CRF) system and the locus-coeruleus-norepinephrine (LC-NE) system (Boyce & Ellis,
2005; McEwen, 1998). The CRF system is involved in the regulation of the HPA axis as
well as the corticolimbic circuitry of the amygdala. The LC-NE activation of the
hypothalamus, on the other hand, results in the activation of the autonomic nervous
24
system (ANS) and initiates fight or flight responses to stress. The simultaneous
activation of the CRF and LC-NE systems, in addition to their projections to the
amygdala, anterior cingulate cortex, and mesocorticolimbic dopaminergic systems allows
for behavioural responses to stress (Boyce & Ellis, 2005). The sympathetic branch of the
ANS reacts quickly to threat to optimize fight or flight responses through the release of
catecholamines; the HPA axis, on the other hand, is comprised of a slower response of
endocrine signals ultimately leading to the release of glucocorticoids (Ewing-Cobbs et
al., 2017).
Evidence of plasticity within the HPA axis has been obtained from rodent studies
reporting epigenetic modifications of glucocorticoid receptors as a function of variation
in the quality of the early-life environment (Liu et al., 1997; Meaney, 2010). In humans,
plasticity within the HPA-axis is demonstrated via observations that the reactivity
threshold of the HPA-axis changes in response to chronic adversity (See Tables 2.1-2.4).
Therefore, when considering the long-term biological embedding of adversity, the HPA-
axis is an important physiological system to consider, not only because of its central role
in responding to acute stressors, but also due to its high degree of plasticity. The HPA
axis is comprised of a feedback loop between the hypothalamus, the pituitary gland, and
the adrenal glands. In response to stress, CRF is released from the hypothalamus; CRF
then stimulates the pituitary system to release adrenocorticotropic hormone (ACTH) that
ultimately results in the adrenal cortical secretion of glucocorticoids – cortisol in humans
and corticosterone in animal species. Glucocorticoid secretion subsequently forms a
negative feedback loop to the hypothalamus to inhibit the secretion of CRF (Berens et al
2017; Del Rey, Chrousos, & Besedovsky, 2008). The HPA axis is central to regulating a
host of homeostatic systems in the body including the metabolic, cardiovascular, and
immune systems, thereby allowing organisms to adapt to their changing environments.
Furthermore, glucocorticoid production results in increased availability of glucose to
facilitate “fight-or-flight” mechanisms in response to acute stress (Del Rey et al., 2008).
Thus, in the face of a stressor, neuroendocrine signals redirect energy resources to allow
the organism to overcome the stressor (Selye, 1950; Boyce and Ellis, 2005). Chronic
exposure to adversity however, can disrupt the set-point of HPA activity, resulting in
either hyper or hypo-reactivity (Yehuda et al., 2010). While HPA axis activation is
25
adaptive under acutely stressful situations, chronic activation of the HPA axis, for
example under conditions of long-term adversity or trauma, may lead to negative health
outcomes and various pathologies (McEwen & Wingfield, 2003).
2.2 Theoretical frameworks for understanding the

relationship between adversity and HPA axis function
Various theoretical frameworks have been proposed to understand the impact of
adversities on HPA axis function. Broadly, there are two categories of theoretical
perspectives: (1) “Toxic stress” models that view the consequences of chronic and
extreme adversities as ultimately harmful to the body (McEwen & Wingfield, 2003;
Nelson, 2017; Selye, 1946; Shonkoff & Garner, 2012); and (2) “Adaptationist”
perspectives that view the calibration of the HPA axis in response to changing
environmental conditions as evolutionarily adaptive (Del Giuidice, Ellis, & Shirtcliff,
2011).
2.2.1 Allostatic load and overload: Wear and tear on the body
Allostatic load and overload explain how prolonged exposure to stressors can
have potentially negative consequences on physiology. While allostasis has been defined
as the process of maintaining homeostasis (Sterling & Eyer, 1988), allostatic load or
overload is the wear and tear on the body associated with prolonged or dysregulated use
of allostasis (McEwen & Stellar, 1993; McEwen, 1998; McEwen & Wingfield, 2003).
The primary mediators of allostasis include the HPA axis, catecholamines, and cytokines,
and the allostatic state of an organism refers to the sustained activity of these primary
mediators (McEwen & Wingfield, 2003). Allostatic overload is further subdivided into
Types I and II, where Type 1 overload is related to an emergency response to an acute
stressor, and Type II overload is the stress response induced by chronic stressors that are
not immediately dangerous (McEwen & Wingfield, 2003).
In the short term, allostatis has protective effects for an organism. For example,
the release of glucocorticoids allows for the conversion of protein and lipids to
carbohydrates allowing the organism enough energy resources to escape a potentially
dangerous acute stressor (McEwen & Wingfield, 2003). If an allostatic state is sustained
26
over a long period of time, however, the mediators of allostasis can become dysregulated
leading to allostatic overload and can have damaging consequences for the body. For
example, chronically elevated levels of glucocorticoids impede the function of insulin,
and the combination of elevated insulin and glucocorticoids results in the storage of body
fat and the formation of atherosclerotic plaques in coronary arteries (McEwen &
Wingfield, 2003). Therefore, if allostatic load is not relieved and glucocorticoid secretion
is chronically elevated, serious pathophysiological effects can occur (McEwen &
Wingfield, 2003). According to this view then, stress reactivity is composed of a
unidirectional, linear biological response to stressors, and chronically sustained stress
responses are associated with stress-related pathophysiology. To illustrate, Danese and
McEwen (2012) conducted a review and summarized the neural, endocrine, and immune
correlates of childhood maltreatment; according to their summary, experiences of
maltreatment are associated with blunted HPA axis reactivity in healthy individuals and
heightened HPA axis reactivity in individuals with mental illnesses (See Figure 2.1).
Moreover, the review summarizes the potentially negative consequences of adverse
childhood experiences on allostasis, allostatic load, and age-related diseases (Danese &
McEwen, 2012).
Figure 2.1: Graphical depiction of relationship between severity of adversity and

HPA axis reactivity as proposed by allostatic load/overload models (Danese &
27
McEwen, 2012). In healthy individuals, exposure to trauma is associated with

blunted HPA axis reactivity; in contrast, in individuals with comorbid mental
illnesses, trauma is related to heightened HPA axis reactivity.
2.2.2 The General Adaptation Syndrome (G-A-S) Model

The first model to propose a curvilinear relationship between exposure to stress and
stress reactivity was Hans Selye’s General Adaptation Syndrome (G-A-S) model (Selye,
1946). According to the G-A-S there are three phases of the stress response: the alarm
stage, the resistance stage, and the exhaustion stage. The alarm stage describes the
organism’s immediate response to an acute stressor and involves cortisol release by the
HPA axis as well as ANS fight-or-flight reactions to the acute stressor. In the resistance
stage, while the ANS response returns to normal, the body continues to produce elevated
levels of cortisol and glucose to maintain arousal should a stressor reoccur. Eventually
however, the body’s resources are depleted, and the organism reaches the exhaustion
stage marked by a drop in the release of cortisol and glucose (Guilliams & Edwards,
2010; Selye, 1946). Therefore, the G-A-S suggests that chronic stressors will result in the
HPA axis moving from a hyper-reactive system to one that becomes under-responsive
(Selye, 1946). The G-A-S model predicts that initial exposure to stressors results in a
phase of hypercortisolism, reflective of the body’s attempts to overcome the stressor.
Critically, if the hypercortisolism phase is prolonged, it may ultimately result in
modifications within the HPA axis resulting in reduced cortisol production, and
ultimately hypocortisolism (Guilliams & Edwards, 2010; Selye, 1946). As depicted in
Figure 2.2, the G-A-S model proposes an inverted U-shaped function to describe the
stress response within individuals over time (Guilliams & Edwards, 2010; Selye, 1946).
Like allostatic load/overload models, the G-A-S model considers the long-term impact of
adversity on stress-reactivity as a contributing factor to the development of disease
(Selye, 1946). For example, according to the G-A -S, various pathologies including
hypertension, rheumatoid arthritis, allergies, and diabetes are by-products of these
endocrine reactions (Selye, 1950).
28
Figure 2.2: Graphical depiction of the relationship between duration of adversity

and HPA axis reactivity as proposed by the G-A-S model (Selye, 1946). According to
this model, acute stressors result in a phase of hypercortisolism; however, if
exposure to adversity persists long-term, hypocortisolism will occur.
2.2.3 Adaptationist perspectives

The view that chronically elevated stress reactivity results in pathophysiology has
been challenged by adaptationist perspectives suggesting that high reactivity phenotypes
may be protective and beneficial under specific environmental conditions and may even
result in normative or improved health outcomes (Boyce & Ellis, 2005). In a study of the
effects of cardiovascular and immunologic reactivity on respiratory illness incidence,
highly reactive children in high-stress families had significantly higher incidence of
respiratory illness as compared to children with lower cardiovascular and immunologic
reactivity. Importantly however, equally reactive children in low-stress families were the
healthiest of all children (Boyce, Chesney, et al., 1995). Based on these observations,
Boyce and Ellis (2005) proposed the Biological Sensitivity to Context (BSC) model which
states that reactivity does not always reflect a disadvantageous response that puts the
organism at risk for developing pathophysiology, but may reflect a form of conditional
adaptation (Boyce & Ellis, 2005). The BSC is an evolutionary-developmental theory with
the hypothesis that there is a curvilinear, U-shaped relationship between exposure to
29
adversity and the development of stress-reactivity profiles – with highly stress-reactive

phenotypes emerging under both extremely stressful and highly protected environments
(Boyce & Ellis, 2005). Extremely stressful environments, it is claimed, beget increased
vigilance to threats while protective environments foster increased susceptibility to
resources. Paradoxically, then, highly reactive phenotypes that demonstrate vigorous
and/or persistent autonomic, adrenocortical, or other biological response to stress are
thought to emerge under both protected and extremely stressful environments.
The Adaptive Calibration Model (ACM) of stress responsivity is an extension of

the BSC and proposes that individual differences in stress reactivity reflect conditional
adaptations –the evolved ability of organisms to modify their phenotypes and
developmental trajectories to fit the conditions of the local environment (Del Giudice et
al., 2011). Like the BSC perspective, according to this view, individual differences in
stress reactivity are considered adaptive mechanisms rather than the result of pathological
processes (Del Giudice et al., 2011). The central premise here is that across development,
information is encoded into the stress response systems that feeds back on the long-term
calibration of the system itself. This plasticity of the stress response systems allows for
patterns of reactivity that are optimal for the developmental environment of that organism
(Boyce & Ellis, 2005; Del Giudice et al., 2011). The ACM expands upon the BSC in that
it also predicts a curvilinear U-shaped relationship between adversity and HPA axis
reactivity; however, due to sex differences in life-history trade-offs, in environments
characterized by traumatic stress, the ACM predicts blunted reactivity in males (Del
Giudice et al., 2011). Therefore, both the BSC and ACM models predict a curvilinear
relationship between exposure to adversity and stress reactivity; moreover, they challenge
the notion that exposure to adversity results in dysregulation of biological functioning
and derails normal development (Del Giudice et al., 2011).
30
Figure 2.3: Graphical depiction of the relationship between severity of adversity and
HPA axis reactivity as proposed by the BSC model (Boyce & Ellis, 2005). According
to this model, HPA hyper-reactivity emerges under both protected environments
with low adversity and extremely stressful environments.
2.3 Empirical findings on the association between life-

experiences and HPA axis reactivity
Evidence for an association between early-life adversity and HPA axis reactivity in
humans is inconsistent (See Tables 2.1-2.4 for summaries of the conflicting literature),
with both heightened (See Table 2.1) and blunted cortisol reactivity (See Table 2.2) to a
social stress task evident in individuals exposed to traumatic life events. Likewise, in
individuals exposed to moderate levels of adversity, some studies have reported
heightened cortisol reactivity (See Table 2.3), while others have reported blunted cortisol
reactivity (See Table 2.4). HPA axis hyper-reactivity would be suggestive of resistance or
attenuated sensitivity to glucocorticoid negative feedback, while HPA axis hypo-
reactivity would be suggestive of exaggerated HPA axis suppression following chronic
activation (Berens et al 2017). Whether exposure to adversities, however, result in the
HPA-axis becoming hyper-reactive or hypo-reactive remains unclear with research
supporting both hyper-reactive and hypo-reactive profiles.
31
Two meta-analyses have reviewed the relationship between adverse early-life

experiences and cortisol reactivity to a laboratory stressor in humans. One found that
early-life adversity was associated with blunted cortisol release in response to social
stress (Bunea, Szentagotai-Tatar, & Miu, 2017). A more recent meta-analysis of 54
studies investigated the association between early-life adversity and various
measurements of cortisol including cortisol awakening response (CAR), baseline cortisol,
non-stressed cortisol over time, and cortisol reactivity to a stressor. In this meta-analysis
however, there was no significant relationship between early-life adversity and any of the
cortisol measurements, including cortisol reactivity (Fogelman & Canli, 2018). Taken
together then, it is unclear how – or even whether – early-life adversity and cortisol
reactivity are associated. Both meta-analyses reported substantial effect size
heterogeneity across studies and suggested that inconsistent operationalizations of
adversity and cortisol measurements may have contributed to varying effect size
estimates. Therefore, to more accurately characterize the association between adverse
life-experiences and cortisol reactivity, methodological differences in cortisol
measurements, as well as variation in defining the terms “adversity” and “trauma” need
to be taken into consideration when reviewing available literature.
2.4 Sources of variability and confounding factors

2.4.1 Cortisol measurements
Studies vary considerably in the type and timing of cortisol measurements.
Cortisol is the end-product of the HPA axis; however, variation in cortisol measurements
do not necessarily reflect dysregulation at other points within the HPA axis (Nicolson,
2008). The conclusion is often made that the HPA axis has become dysregulated based
solely on results obtained from cortisol values; however, even when cortisol levels are
within normal ranges, other probes within the HPA axis may reflect abnormalities. For
example, elevated CRH or ACTH secretion may be coupled with decreased adrenal
sensitivity; therefore, while CRH and ACTH levels may be elevated, cortisol values may
be within normal ranges (Nicolson, 2008). Despite the observation that variation in
cortisol levels do not reflect all changes within the HPA axis, cortisol measurements are
the most common way to assess HPA axis function. This is largely due to the relatively
32
non-invasive procedures involving the collection of salivary cortisol samples. That said,
there is considerable variation in how cortisol data can be obtained, including
measurements obtained from urine, serum, saliva, hair, and nails (Guilliams & Edwards,
2010; Nejad & Ghaseminezhad, 2016; Nicolson, 2008). One additional benefit of
obtaining salivary cortisol measurements is that the cortisol obtained from saliva is in the
unbound form (free fraction used to measure HPA axis status). In contrast, the cortisol
obtained through serum is bound to corticosteroid binding globulin (CBG), and the free
cortisol must be separated from the bound cortisol in serum samples making its analysis
more complicated (Guilliams & Edwards, 2010).
In addition to the variability introduced from the type of cortisol sample collected,
there are other sources of variability including whether the cortisol measure is a basal
measure of cortisol or a reactivity measure to an acute stressor. Basal measures of cortisol
are used to assess spontaneous cortisol secretions, including circadian patterns of HPA
axis activity; they can be measured via 24-hour urinary sampling or repeated salivary
sampling (Nicolson, 2008). One measure of circadian rhythm often used is the cortisol
awakening response (CAR). Cortisol levels increase sharply during the first 30-40
minutes of waking, and the CAR measures this increase in cortisol levels. The CAR may
reflect an important individual difference measure of cortisol as it appears to be relatively
stable within individuals and has a genetic component (Nicolson, 2008; Wust et al.,
2000). Despite this relative intra-individual stability, it appears to have some degree of
sensitivity to short-term environmental influences. For example, the CAR is sensitive to
the impact of stress related to a workday versus a weekend and therefore, can vary
depending on the day of the week (Kunz-Ebrecht et al., 2004; Nicolson, 2008).
Moreover, a p-curve meta-analysis of 709 findings from 212 studies assessed the
relationship between psychosocial functioning and CAR (Boggero et al., 2017). The
analyses demonstrated that the greatest evidential value existed for findings associating
worse psychosocial functioning with lower CAR (Boggero et al., 2017).
In contrast to basal measures of cortisol, reactivity measures assess cortisol

response to acute stressors. Reactivity measures can assess cortisol response to
pharmacological stimulation (e.g., Dex/CRH tests), physical stress (e.g., physical exercise
33
or cold pressor test), emotional induction stressors (e.g., watching an emotional film, or
discussing past trauma), or psychosocial stressors (e.g., Trier Social Stress Test - TSST).
The type of task used to assess cortisol reactivity can introduce variability in the results
obtained from such studies; further variability is also introduced in the form of timing and
number of saliva samples collected (Nicolson, 2008). Among psychosocial stressors
however, those that involve social-evaluative threat and are uncontrollable produce the
greatest and most consistent increases in cortisol (Nicolson, 2008). Despite the utility of
psychosocial stressors in providing reliable measures of cortisol reactivity, empirical
findings on associations between life-experiences and cortisol reactivity to the
psychosocial stress remain mixed (See Tables 2.1-2.4).
2.4.2 Trauma versus adversity

One major source of variation when examining relations between early-life
experiences and HPA axis function is the lack of a clear definition of trauma. More
troublesome is the fact that there are no conceptual boundaries to determine what
constitutes “adversity” versus “trauma”. (See Krupnik, 2019 for an excellent summary).
In some cases, a narrow definition of trauma is used in accordance with Criterion A of
trauma-related disorders as defined in the DSM-V. Whereas in other cases, trauma is
defined broadly and can constitute virtually any negative life event that results in lasting
changes (Krupnik, 2019). As a result of the unclear definitions and boundaries of trauma
and adversity, the definition of an adverse event may range from extremely negative
experiences (e.g., maltreatment or sexual violence), to more mild negative experiences
(e.g., the loss of a family pet or changing residences). Still others have proposed that
whether a negative event is traumatic or not could depend on subjective versus objective
experiences; in other words, trauma could be defined as an individual’s subjective
experience of an event rather than exposure to it (Krupnik, 2019).
To help differentiate trauma from normal stress responses and adversity, Krupnik
(2019) suggests that stressful experiences be categorized into the following: (a)
normative stress response (NSR), (b) pathogenic stress response (PSR), and (c) traumatic
stress response (TSR). According to this categorization, in normative stress, the organism
returns to the initial homeostatic state after the stressful experience. In pathogenic stress
34
(synonymous with adversity), Type II allostatic overload may occur and results in a less
adaptive homeostatic state after the stressful experience. Finally, in traumatic stress, the
organism experiences a breakdown in self-regulatory functions and transitions to Type I
allostatic overload (Krupnik, 2019). According to this categorization then, trauma and
adversity may be operationalized as a function of both the severity of the stressor/s and
the strength of self-regulatory function (Krupnik, 2019). Critically, according to this
model, any stressor can become traumatic given that its severity overwhelms self-
regulatory processes (Krupnik, 2019).
In the literature, trauma is often operationalized using DSM-V criteria, while using a
diagnostic manual, such as the DSM-V allows for a precise and narrow definition, one
disadvantage is that it introduces the confound of psychopathology. Adversities
experienced in early childhood are a risk factor for the development of stress-related
psychopathology in adulthood (Fogelman & Canli 2018; Green et al., 2010; Heim et al
2008; Nemeroff, 2004). Therefore, the impact of traumatic life experiences on stress
reactivity is often confounded with the presence of comorbid post-traumatic stress
disorder (PTSD) or major depressive disorder (MDD). For instance, it becomes difficult
to tease apart the relative effects of maltreatment versus PTSD status on HPA axis
reactivity, as many studies compare maltreated individuals with PTSD to non-maltreated
individuals without PTSD (Bremner et al., 2003; Danese & McEwen, 2012; Schalinski et
al., 2015). One review reported that overall, maltreated individuals exhibit heightened
basal cortisol levels and tend to have blunted cortisol reactivity to laboratory
psychosocial stress tasks. Interestingly however, the review suggests that
psychopathology moderates the relationship between adversity and HPA axis reactivity.
In particular, healthy participants with a history of adverse life experiences produce a
blunted cortisol response to a laboratory stressor, while individuals with PTSD exhibit
heightened cortisol reactivity (Danese & McEwen, 2012). This underscores the
importance of delineating between trauma and adversity, while taking psychopathology
into account when evaluating the impact of early-life adversities on HPA axis reactivity.
35
2.5 Current studies: p-curve meta-analyses

Different theoretical frameworks make distinct predictions about the relationship
between adversity and HPA axis reactivity. For example, the G-A-S model (Selye, 1946)
and the review by Danese and McEwen (2012), predict that chronic exposure to traumatic
life events should be related to blunted HPA axis reactivity, at least in individuals without
any comorbid mental illnesses. In contrast, the BSC (Boyce & Ellis, 2005) and ACM
models (Del Giudice et al., 2011) make the opposite predictions – namely that exposure
to traumatic experiences should be related to heightened HPA axis reactivity.
Empirically, the literature to date has provided contradictory results, with many studies
reporting associations between adverse life-experiences and cortisol hyper-reactivity (See
Tables 2.1 and 2.3), and others reporting cortisol hypo-reactivity (See Tables 2.2 and
2.4). Recent meta-analyses have also reported conflicting results (Bunea et al., 2017;
Fogelman & Canli, 2018); therefore, it is unclear whether the empirical data supports a
link between adverse life-experiences and HPA-axis reactivity.
P-curve is the distribution of statistically significant p values (< .05) for a set of
findings, and its shape (whether it is right or left-skewed) is diagnostic of evidential value
(Simonsohn, Nelson, & Simmons, 2014a). A set of findings contains evidential value
only when selective reporting can be ruled out as the sole explanation for those findings,
and only right-skewed p-curves, that is, those with more small (e.g., .01s) than large (e.g.,
.04s) significant p values, reflect evidential value (Simonsohn et al., 2014a). Importantly,
p-curves that are not right-skewed indicate that the findings lack evidential value, and
those that are left-skewed (containing more studies with p values around .04) are
suggestive of p-hacking (Simonsohn et al., 2014a). While conducting research, scientists
are faced with many decisions including the number of samples to collect, the number of
variables to operationalize, and the number of analyses that should be conducted. If these
decisions are not made prior to data analysis, researchers can continue to analyze
different variables or numbers of participants until they reach the desired significance
threshold. These practices increase the odds of reaching the significance threshold and
are referred to as p-hacking (Simonsohn et al., 2014a). Conducting a p-curve meta-
analysis may have many uses, including determining which set of contradictory findings
36
has not been subject to p-hacking and is therefore, more likely to be correct (Simonsohn
et al., 2014b; 2015).
Here, we use p-curve to examine the conflicting findings regarding the association
between early-life experiences and HPA axis reactivity. Critically, we make a clear
distinction between traumatic life experiences and adverse life experiences, and we
predict differential reactivity within the HPA-axis as a function of the severity of adverse
life-experiences. Furthermore, we have only included studies in these analyses that assess
salivary cortisol reactivity to a laboratory-based stressor (e.g., TSST), and thus we have
reduced some of the variability introduced as a result of differences in cortisol data
collection. As shown in Figure 2.4, Study 1 assesses the literature reporting associations
between traumatic life experiences and salivary cortisol reactivity. In Study 2, we
examine the literature reporting associations between moderately adverse life experiences
and salivary cortisol reactivity. When assessing each subset of results, if a right-skewed
distribution of p-values is found, the conclusion can be drawn that there is evidential
value in the findings and the results of the experiments have not been subject to selective
reporting. Finding a left-skewed distribution on the other hand, would indicate a lack of
evidential value and be suggestive of selective reporting. For example, in Study 1, if we
find that the literature reporting associations between trauma and cortisol hypo-reactivity
contains evidential value, while the literature reporting a relationship between trauma and
cortisol hyper-reactivity does not contain evidential value, we can have more confidence
in the first subset of findings. Therefore, using p-curve to conduct these analyses will
help clarify the inconsistencies in the literature and will allow us to determine which set
of findings is more likely to be correct.
37
Figure 2.4: Graphical depiction of Study 1 and Study 2 aims. The orange curvilinear
function represents the predictions that can be made about the relationship between
adversity and HPA axis reactivity based on the G-A-S model. The U-shaped
function in black represents the predictions that can be made concerning the
relationship between severity of adversity and HPA axis reactivity based on the BSC
model. In Study 1 (outlined in red), we will use p-curve to determine whether the
literature reporting that exposure to trauma is related to heightened cortisol
reactivity or blunted cortisol reactivity contains evidential value. In Study 2
(outlined in yellow), we will use p-curve to assess whether exposure to moderate
adversity is related to heightened or blunted cortisol reactivity.
2.6 Methods
2.6.1 Study 1: Data sources and study selection
We conducted a thorough search of all human studies investigating the association
between traumatic life experiences and cortisol reactivity to a psychosocial stressor from
1999 to 2020. We limited our search to only include articles after 1999 because the most
influential theoretical models to motivate this literature were published in the 2000’s.
The terms “trauma”, “adversity”, “stress”, “maltreatment”, and “early-life stress” are
often used interchangeably in the literature; similarly, the terms “HPA-reactivity”,
38
“cortisol reactivity” and “stress reactivity” are also used interchangeably (Pre-registration
osf.io/9ehsn). Therefore, we conducted a PsycINFO database search using the following
terms:
((trauma or stress or adversity or early-life stress or maltreatment) and (HPA-reactivity or

cortisol reactivity or stress-reactivity or cortisol hyper-reactivity or cortisol hypo-
reactivity)) [mp=title, abstract, heading word, table of contents, key concepts, original
title, tests & measures, mesh].
The results yielded 2673 articles, we excluded articles published prior to 1999,
reducing the results to 2447 articles. We imposed a further limit to include only peer-
reviewed journals and human literature, which reduced the results to 1663 articles. These
1663 articles went to abstract review from the PsycINFO database search. For the
purposes of this study, we were interested in human studies of traumatic life experiences.
In accordance with DSM-5, the experience must have included qualifying events for
post-traumatic stress disorder (PTSD). Furthermore, we were interested in salivary
cortisol reactivity to a laboratory stressor, thus basal or diurnal cortisol measurements and
plasma or urine cortisol measurements were excluded. Based on these inclusion criteria,
we selected 51 articles from the PsycINFO search to include in the p-curve analyses (See
Figure 2.5).
To be thorough, we also conducted a PubMed database search using the following terms:
(((((((((trauma[Title/Abstract]) OR stress[Title/Abstract]) OR early-life

stress[Title/Abstract]) OR adversity[Title/Abstract]) OR maltreatment[Title/Abstract])
AND HPA-reactivity[Title/Abstract]) OR cortisol reactivity[Title/Abstract]) OR stress-
reactivity[Title/Abstract]) OR cortisol hyper-reactivity[Title/Abstract]) OR cortisol hypo-
reactivity[Title/Abstract] AND (Humans[Mesh]).
The search yielded 2645 results, we excluded studies that were not conducted
with human participants, and 1827 articles remained. These 1827 articles were reviewed
for relevance according to their title. Of the 1827 articles, 59 articles went to full abstract
review. When duplicates of the PsycINFO search results were removed and inclusion
39
criteria were applied, 12 additional articles were added to the p-curve analyses (See
Figure 2.6). An additional 5 articles were included from the reference sections of other
articles in the searches (See Figure 2.7).
The following inclusion criteria were applied:

1.) Must be an empirical study with human participants
2.) Must measure traumatic life experiences in accordance with DSM-5 criteria:
Exposure to: death, threatened death, actual or threatened serious injury,
or actual or threatened sexual violence, in the following way(s): direct
exposure; witnessing the trauma; learning that a relative or close friend
was exposed to a trauma; indirect exposure to aversive details of the
trauma, usually in the course of professional duties (e.g., first responders,
medics) (Diagnostic and Statistical Manual of Mental Disorders, Fifth
Edition [DSM–5]; American Psychiatric Association, 2013).
3.) Must include salivary cortisol reactivity to a laboratory stressor
2.6.1.1 Study 1: Study sample

The final pool included 68 unique articles published from 1999-2020. Of these 68
articles, 2 articles were included in both the hyper and hypo-reactivity p-curve analyses,
because one portion of the results showed hyper-reactivity profiles, while the other
showed hypo-reactivity profiles. For instance, in one study, trauma experienced before
age 11 resulted in cortisol hyper-reactivity, while trauma experienced after age 11
resulted in cortisol hypo-reactivity (Bosch et al., 2012), therefore, this study was included
in both the hyper-reactivity and hypo-reactivity p-curve analyses. Thus, the hypo-
reactivity p-curve analysis included 48 results and the hyper-reactivity p-curve analysis
included 22 results.
40
PsycINFO Keyword Search Results Excluded (published prior to 1999)

N = 2,673 N = 226
Identification
Excluded (not peer reviewed, not in

Records Identified
English language, or animal studies)
N = 2,447
N = 784
Excluded (reviews, meta-analyses,

Records Identified
N = 1,663 or perspectives)
N = 251

Records Identified Excluded or perspectives)
(did not measure trauma)
N = 1,412 NN==1,249
251
Screening
Excluded (did not measure trauma)

Records Identified Excluded (didN not
= 1,249
measure salivary
N = 163 cortisol reactivity to a lab stressor)
N = 90
Excluded (reported trend level p-

Eligibility
values or did not report statistics on

Records Identified relationship between trauma and
N = 73 cortisol reactivity)
N = 22
Total Records Included

Included
N = 51
Figure 2.5: Flowchart - PsycINFO database search (trauma and cortisol reactivity)
41
PubMed Keyword Search Results Excluded (animal studies)

Identification N = 2,645 N = 818
PubMed Keyword Search Results

N = 2,645 Excluded (animal studies)
Title Review N = 818
N = 1827
Title Review
N = 1827
Abstract Review Excluded (duplicates of PsycINFO
N = 59 Search)
N = 43
Excluded (duplicates of PsycINFO

Screening
Records Identified Excluded (didSearch)

not measure salivary
N = 16 cortisol reactivity
N = 43
to a lab stressor)
N=2
Records Identified Excluded (did not measure salivary

N =Identified
Records 16 cortisol reactivity
Excluded to a lab stressor)
(did not measure trauma)
N = 14
Abstract Review NN==22
N = 59
Excluded (did not measure trauma)
Records Identified
N=2
N = 14
Eligibility
Records Identified
N = 12
Records Identified
N = 12
Included

N = 12
Figure 2.6: Flowchart - PubMed database search (trauma and cortisol reactivity)
42
PsycINFO Records Included (k = 51)

(
N = 51
PsycINFO Records Included

Records Entered Records Entered
N = 51 Records Entered
into Cortisol Hyper- into Cortisol Hypo- into Cortisol Both
reactivity p-curve reactivity p-curve p-curves
k = 14
k = 35 k=2
Records Entered
into Cortisol Hyper- Records Entered Records Entered
reactivity p-curve PubMedinto
Records Included
Cortisol Hypo- (k = 12) into Cortisol Both
N = 14 reactivity p-curve p-curves
N = 35 N=2
PubMed Records Included
N = 12
Records Entered Records Entered

into Cortisol Hyper- into Cortisol Hypo-
reactivity p-curve reactivity p-curve
k=6
k=6
Records Entered
into Cortisol Hyper- Records Entered
Records
reactivity found from Reference Sections
p-curve (k = 5) Hypo-
into Cortisol
N=6 N = 5 reactivity p-curve
N=6
Records found from Reference Sections
N=5
Articles Entered
into Cortisol Hypo-
reactivity p-curve
k=5
Figure 2.7: Number of articles included from each database

43
2.6.2 Study 2: Data sources and study selection

We conducted a search of human studies investigating the association between
moderately adverse life experiences and cortisol reactivity to a psychosocial stressor from
1999 to 2020. We conducted a PsycINFO database search using the following terms:
(((stress or adversity) and HPA-reactivity) or cortisol reactivity or stress-reactivity or

cortisol hyper-reactivity or cortisol hypo-reactivity). ab.
The search yielded 2515 results, we excluded articles published prior to 1999, reducing
the results to 2350 articles. We imposed a further limit to include only English-language,
peer-reviewed articles with human participants, which reduced the results to 1597
articles. These 1597 articles went to abstract review from the PsycINFO database search.
For the purposes of this study, we were interested in human studies of moderately
adverse life experiences (e.g., low SES, daily hassles, parental conflict). To ensure that
we included only articles with moderately adverse life experiences, articles that measured
traumatic experiences were excluded. Trauma was defined in accordance with DSM-5
criteria for qualifying events for post-traumatic stress disorder (PTSD). Once again, we
were interested in salivary cortisol reactivity to a laboratory stressor as the outcome
measure. Based on these inclusion criteria, we selected 23 articles from the PsycINFO
search to include in the p-curve analyses (See Figure 2.8).
We also conducted a PubMed database search using the following terms:
((((((Stress[Title/Abstract]) OR (Adversity[Title/Abstract])) AND (HPA-

reactivity[Title/Abstract])) OR (cortisol-reactivity[Title/Abstract])) OR (stress-
reactivity[Title/Abstract])) OR (cortisol hyper-reactivity[Title/Abstract])) OR (cortisol
hypo-reactivity[Title/Abstract])
The search yielded 2800 results, we excluded studies that were published prior to 1999
resulting in 2681 remaining articles. Next, we excluded articles that were not written in
the English language, were not peer-reviewed or were animal studies, resulting in 1784
articles. Of these 1784 articles, 1268 were duplicates of the PsycINFO search and were
therefore excluded, resulting in 516 articles that went to abstract review. After the
44
abstract review, based on the inclusion criteria, 3 additional articles were included in the
p-curve analyses (See Figure 2.9). Finally, an additional 4 articles were included from the
reference sections of other articles (See Figure 2.10).

1. Must include empirical studies with human participants
2. Must measure moderately adverse life experiences (excluding severe
adversity/trauma)
3. Must include salivary cortisol reactivity to a laboratory stressor
2.6.2.1 Study 2: Study sample

The final pool included 30 unique articles published from 1999-2020. Of these 30
articles, one article was included in both the hyper and hypo-reactivity p-curve analyses,
because for males, adverse life experiences resulted in blunted cortisol reactivity,
whereas for females, adverse life experiences resulted in heightened cortisol reactivity
(Fearon et al., 2017). Therefore, this study was included in both the hyper-reactivity and
hypo-reactivity p-curve analyses. In total, the hypo-reactivity p-curve analysis included
16 results, and the hyper-reactivity p-curve analysis included 15 results.
45
PsycINFO Keyword Search Results Excluded (published prior to 1999)

N = 2,515 N = 165
Identification PsycINFO Keyword Search Results

N = 2,515 Excluded (published prior to 1999)
Records Identified N = 165
N = 2,350
N = 753
Records Identified
N = 2,350
Excluded
English (reviews,
language, or meta-analyses,
animal studies)
Records Identified
or perspectives)
N = 753
N = 1,597
N = 221
Records Identified
N = 1,597 Excluded (reviews, meta-analyses,
Records Identified Excluded (did not measure
or perspectives)
N = 1,376 moderate adversity)
N = 221
Screening
N = 1,236
Records Identified Excluded (did not measure

N = 1,376
Records Identified moderate
Excluded (did notadversity)
measure salivary
N = 140 cortisol reactivity
N = 1,236
to a lab stressor)
N = 91
Records Identified Excluded (did not measure salivary

N = 140 Excluded
cortisol (reported
reactivity to atrend level p-
lab stressor)
Eligibility
values or did not

N =report
91 statistics on
Records Identified relationship between adversity and
N = 49 cortisol reactivity)
N = 26

Records Identified values or did not report statistics on
N = 49 relationship between adversity and
Included

cortisol reactivity)
N = 23
N = 26
Figure 2.8: Flowchart - PsycINFO database search (adversity and cortisol

reactivity)
46
PubMed Keyword Search Results Excluded (published prior to 1999)

N = 2,800 N = 119
PubMed Keyword Search Results

N = 2,800 Excluded (not
Excluded (published prior to 1999)
peer reviewed, not in
N = 2,681
Identification
N = 897
Records Identified
N = 2,681 Excluded (duplicates of PsychInfo
Records Identified English language, or animal studies)
search
N = 897
N = 1,784
N = 1268
Records Identified
Excluded (duplicates of PsychInfo
N = 1,784 Excluded (reviews, meta-analyses,
Records Identified search
or perspectives)
N = 1268
N = 516
N = 129
Records Identified
N = 516
Records Identified Excluded (did not measure moderate
or perspectives)
N = 387 adversity)
N = 129
N = 332
Screening
Records Identified Excluded (did not measure moderate

N = Identified
Records 387 Excluded (didadversity)
not measure salivary
N = 55 cortisol reactivity to a lab stressor)
N = 332
N = 46
Records Identified Excluded

Excluded(did not measure
(reported salivary
trend level p-
N =Identified
Records 55 cortisolorreactivity to a lab stressor)
Eligibility
values did not report statistics on

N=9 N = 46
relationship between adversity and
N=6
Records Identified
N=9
Included
Total Records Included values or did not report statistics on

N=3 relationship between adversity and
N=6
Figure 2.9: Flowchart - PubMed database search (adversity and cortisol reactivity)
47
PsycINFO Articles Included (k = 23)
N = 23
Articles Entered Articles Entered Articles Entered

into Cortisol Hyper- into Cortisol Hypo- into Cortisol Both
reactivity p-curve reactivity p-curve p-curves
k = 14
k=8 k=1
PubMed Articles Included (k = 3)
N=3
Articles Entered
into Cortisol Hypo-
reactivity p-curve
k=3
Articles found from Reference Sections (k = 4)
N=4
Articles Entered
into Cortisol Hypo-
reactivity p-curve
k=4
Figure 2.10: Number of articles included from each database

48
2.6.3 P-curve data abstraction

We separated the articles from both the PsycINFO and PubMED searches into the
following four groups:
(1) The group of articles reporting that traumatic experiences are related to
subsequent cortisol hyper-reactivity;
(2) The group of articles reporting that traumatic experiences are related to
subsequent cortisol hypo-reactivity;
(3) The group of articles reporting that adverse experiences are related to subsequent
cortisol hyper-reactivity; and
(4) The group of articles reporting that adverse experiences are related to subsequent
cortisol hypo-reactivity
Following open-science guidelines, we created four p-curve disclosure tables—one

for each set of results (See Tables 2.1-2.4). The disclosure tables include the study
reference, hypothesis of interest to the researchers, sample characteristics, measure of
trauma, stress induction task, cortisol measurement, data analysis plan, the actual result
subjected to p-curve analysis, the exact p-value, and the conclusion drawn by the authors
of the paper. We followed the methods provided by Simonsohn et al. 2014a to conduct
the p-curve analysis. Whenever there were more than two relevant p-values, we selected
the p-value reported first; and in ANOVAs if an attenuation was predicted, the interaction
statistic was selected for p-curve analysis (Simonsohn et al., 2014a). After creating the
disclosure tables and obtaining all relevant results, we conducted four separate p-curves
for each set of results. Conducting these p-curve analyses would determine which set of
results, if any, contains evidential value. All p-curves were calculated using an online tool
(http://www.p-curve.com/app4/).
2.7 Results
2.7.1 Study 1.1: Traumatic experiences and cortisol hyper-
reactivity p-curve
We first aimed to determine whether a history of experiencing traumatic events was
related to heightened cortisol reactivity. Based on our research results, 22 studies reported
49
evidence of cortisol hyper-reactivity to a stressor after having experienced traumatic life

events. To determine whether evidential value is present in this set of findings, we
conducted a p-curve analysis of all 22 reported results. As outlined in Simonsohn et al.
(2014a; 2015), a set of results contains evidential value if either the half p-curve (p-
values below .025) is significantly right-skewed at p < .05, or if both the half p-curve and
full p-curve are significantly right-skewed at p < .10. The p-curve for this set of results
was not significantly right-skewed (half: z = -1.12, p = .1321; full: z = -1.62, p = .0531).
Neither condition was met in this case (See Figure 2.11); therefore, the p-curve did not
indicate evidential value. Moreover, as shown in Figure 2.12, these results do not hinge
upon a few extreme observations of p-values. If we remove the smallest and largest p-
values, the conclusion drawn from p-curve would not be much different from what we
observed when we included all 22 p-values in the analysis. In other words, if we drop the
largest p-values from this set of literature, the resultant p-curve would still not be
significantly right-skewed.
50
Figure 2.11: Trauma-Cortisol Hyper-reactivity p-curve. The blue line shows

the distribution of p-values from the data (22 statistically significant results).
For example, 41% of the significant p values were between .01 and .02. The
red dotted line shows the expected distribution of p values if there were no
effect. The green dashed line shows the distribution of p values if an effect
existed with studies powered at 33%. N = 22 p values, p-curve is not
significantly right-skewed
51
Figure 2.12: Dropping the lowest and highest p values on the significance of right-
skewness of the full p-curve (top row), half p-curve (middle row), and the test for
flatness relative to 33% power (bottom row). The red solid line demonstrates the
significance threshold of p = .05. The filled circle demonstrates the results reported
in the text, the open circle demonstrates the results relative to the significance level
if (k) number of p-values were dropped.
According to the guidelines provided by Simonsohn et al (2014a; 2015), a follow-

up test should be conducted when p-curve is not significantly right-skewed, namely,
testing whether it is flatter than one would expect if studies were powered at only 33%.
Therefore, we compared the observed p-curve with the p-curve that would be expected if
the studies had an average power of 33%. An absence of evidential value can be
concluded if the full p-curve is significantly flatter than the p-curve of 33% power at p <
.05, or if both the half p-curve and binomial tests are significant at p <.10. In this case,
these conditions were not met (half: z = -3.4, p =.9997; full: z = - 1.16, p = .1236;
binomial: p = .3051). Thus, we could not conclusively reject the null that the set of
studies contains a small, but detectable effect. In other words, while the p-curve was not
significantly right-skewed and did not indicate evidential value; p-curve could not reject a
very small effect and could not indicate that evidential value was totally absent.
52
Therefore, the conclusion from this set of findings is inconclusive, the current p-curve is
too noisy to reject the possibility of a small effect and at the same time, does not provide
support for the existence of evidential value. In this case, more p-values are required to
determine whether the studies contain evidential value for a small effect. We must,
however, keep in mind that the average power for this set of studies was low. In the case
of studies investigating a relationship between having had traumatic life experiences and
cortisol hyper-reactivity to a laboratory stressor, the power is 16%, 90% CI [5,42]. Based
on this, if the same studies were done again, only 16% of them would detect a significant
result.
To be thorough and to follow the exact guidelines provided by Simonsohn et al.

(2014a), we also conducted a robustness test for the above p-curve. In the robustness p-
curve analysis, 3 alternative p-values were included in the analysis. The results remained
unaltered with the robustness check; that is, the test for evidential value was
nonsignificant (half: z = -1.23, p = .1086; full: z = -1.2, p = .1147). The follow-up
analysis determined that the p-curve was not significantly flatter than if the studies were,
on average, powered at 33% (half: z = 3.26, p = .9994; full: z = -1.5; p = .0668; binomial
p = .0817), and the estimate of average power was 12%, 90% CI [5,36]. Finally, because
the experience of traumatic events is related to the diagnosis of certain
psychopathologies, we aimed to ensure that these results were not driven by comorbid
psychopathology. Therefore, we removed 6 studies that included patients with
PTSD/MDD from the original 22 studies. The results of the p-curve analysis did not
change. Neither of the conditions of p-curve were met, and the p-curve was not
significantly right-skewed (half: z = -0.25, p = .3996; full: z = -1.48, p = .0692; See
Appendix A: Supplementary Materials for further details).
2.7.2 Study 1.2: Traumatic experiences and cortisol hypo-

reactivity p-curve
Next, we determined whether the literature reporting a relationship between the
experience of traumatic events and cortisol hypo-reactivity to a laboratory stressor
contained evidential value. Our literature search yielded 48 results that were included in
the cortisol hypo-reactivity p-curve analysis. Again, a set of results contains evidential
53
value if either the half p-curve is significantly right-skewed at p < .05, or if both the half
and full p-curves are significantly right-skewed at p < .10. The p-curve for the 48 results
was significantly right-skewed (half: z = -5.37, p < .0001; full: z = -5.85, p < .0001). In
this case, both conditions were met, and the p-curve demonstrates that these results
contain evidential value (See Figure 2.13). Once again, as illustrated in Figure 2.14, these
results do not hinge upon extreme observations and were not altered by removing
extreme observations of p-values from the original analysis. The studies investigating the
association between traumatic life experiences and cortisol hypo-reactivity also have
higher power than those of the hyper-cortisol reactivity set; in this case, the power is
46%, 90% CI [29,63].
Figure 2.13: Trauma-Cortisol Hypo-reactivity p-curve. The blue line shows the
distribution of p-values from the data (48 statistically significant results). For
example, 40% of the significant p values were below .01. The red dotted line shows
the expected distribution o of p values if there were no effect. The green dashed line
shows the distribution of p values if an effect existed with studies powered at 33%. N
= 48 p values, p-curve is significantly right-skewed.
54
We also conducted a robustness test for this p-curve in accordance with the
guidelines provided by Simonsohn et al. (2014a; 2015). Six alternative p-values were
included in the p-curve and the results remained unaltered with the robustness check; the
test for evidential value remained significant (half: z = -3.9, p < .0001; full: z = -4.63, p <
.0001), and the estimate of average power was 35%, CI 90% [19,53].
Once again, to ensure that these results were not driven by underlying
psychopathology within the participants of the studies, we removed 4 studies that
included participants with PTSD/MDD. Once again, the two conditions that determine
whether a p-curve is significantly right-skewed were met; therefore, the p-curve for this
set of results was significantly right-skewed (half: z = -5.5, p < .00001; full: z = -6.53, p <
.00001; See Appendix A: Supplementary Materials for further details).
55
2.7.3 Study 2.1: Adverse experiences and cortisol hyper-reactivity

p-curve
To determine whether evidential value exists in the literature reporting associations
between moderate adversity and heightened cortisol reactivity, we conducted a p-curve
analysis of 15 results from our search. The p-curve for the 15 results was significantly
right-skewed (half: z = -2, p = .0227; full: z = -1.56, p = .0598). The p-curve demonstrates
that these results contain evidential value (See Figure 2.15). Once again, as demonstrated
in Figure 2.16, these results do not hinge upon extreme observations and are not altered
by removing extreme observations of p-values from the original analysis. However, the
estimate of power for tests included in this p-curve was only 19%, 90% CI [5,50].
Figure 2.15: Adversity-Cortisol Hyper-reactivity p-curve. The blue line shows the
distribution of p-values from the data (15 statistically significant results). For
example, 53% of the significant p values were below .01. The red dotted line shows
the expected distribution of p values if there were no effect. The green dashed line
shows the distribution of p values if an effect existed with studies powered at 33%. N
= 15 p values, p-curve is significantly right-skewed.
56
For the robustness test for this p-curve, we entered six alternative p-values. The
test for evidential value remained significant (half: z = -3.31, p = .0005; full: z = -2.84, p
= .0023), and the estimate of average power was 39%, CI 90% [13,68].
2.7.4 Study 2.2: Adverse experiences and cortisol hypo-reactivity

p-curve
Based on our literature search, we found 16 studies that reported results showing that
a history of moderately adverse life experiences was related to cortisol hypo-reactivity.
We entered these results into p-curve to determine whether this set of literature contains
evidential value. In this case, the two conditions that determine whether a p-curve is
significantly right-skewed were not met; that is, the half p-curve was not significantly
right-skewed at p < .05, and both the half and full p-curves were not significantly right-
57
skewed at p <.10. Therefore, as shown in Figure 2.17, the p-curve for this set of results
was not significantly right-skewed (half: z = -1.58, p = .0572; full: z = -0.4, p = .3453).
Figure 2.18 shows that these results do not change if we remove the largest p-values and
the conclusion drawn from the p-curve would remain unaltered.
Figure 2.17: Cortisol hypo-reactivity p-curve. The blue line shows the distribution of p-
values from the data (16 statistically significant results). For example, 31% of the
significant p values were between .01 and .02. The red dotted line shows the expected
distribution of p values if there were no effect. The green dashed line shows the
distribution of p values if an effect existed with studies powered at 33%. N = 16 p
values, p-curve is not significantly right-skewed.
58
Next, we conducted the follow-up test required when p-curve is not significantly
right-skewed (Simonsohn et al., 2014a; 2015). That is, we tested whether the p-curve is
flatter than one would expect if studies entered into the analysis were powered at 33%.
An absence of evidential value can only be concluded if the full p-curve is significantly
flatter than the p-curve of 33% power at p < .05. In this case, the full p-curve was
significantly flatter than the p-curve of 33% (full: z = -1.86, p = .0313), indicating that
evidential value is inadequate or absent. Here, we reject the null of a small effect against
an alternative of an even smaller effect and we conclude that this set of studies lacks
evidential value, meaning that either the effect does not exist or it is too small for us to
rule out selective reporting as an alternative explanation. In other words, these results
demonstrate that direct replications of the submitted studies are not expected to succeed;
the estimated power of these studies was 7%, 90% CI[5,29].
59
In the robustness p-curve analysis, 4 alternative p-values were included in the

analysis. The results remain unaltered, the test for evidential value was nonsignificant
(half: z = -1.6, p = .0548; full: z = -0.45, p = .3254); the full p-curve was significantly
flatter than if the studies were powered at 33% (full: z = -1.8; p = .0358), and the estimate
of average power was 7%, 90% CI [5,30]. Finally, for all of the p-curve analyses
conducted, we calculated effect sizes based on the p-values entered and ran a meta-
analysis using a random effects model in the metafor package in R. The results of the
meta-analyses for all p-curves can be found in Appendix A: Supplementary Materials.
2.8 Discussion
Empirical reports of how life experiences are related to HPA axis reactivity remain
contradictory, with some studies reporting that adverse experiences are related to HPA
axis hyper-reactivity and others reporting HPA axis hypo-reactivity (See Tables 2.1-2.4).
Here, we conducted a systematic review of the literature and used p-curve meta-analytic
techniques to resolve some of the apparent inconsistencies in the reported findings.
Critically, in the current meta-analyses, we operationalized our criteria for life
experiences very carefully. In the literature, the term “adversity” is defined broadly with
adverse events ranging from mild adversities (e.g., changing schools) to extreme trauma
(e.g., sexual abuse); therefore, one factor contributing to the mixed findings may be the
lack of consensus on appropriate definitions of adverse life events versus a traumatic life
events (Krupnik, 2019). One strength of the analyses conducted here is that the terms
trauma and adversity were clearly delineated. Traumatic life events were defined in
accordance with DSM-5 criteria for qualifying events for the development of PTSD. In
contrast, adverse life events were defined as stressors experienced throughout
development that were not traumatic (e.g., low SES, parental conflict). Therefore, by
including studies that were comparable in each meta-analysis, we were able to test
specific hypotheses about the relationship between the severity of adverse life
experiences and HPA axis reactivity.
In Study 1, we used p-curve to examine the literature reporting a relationship between

traumatic life experiences and HPA axis reactivity. The literature here was mixed in that
some studies reported HPA axis hyper-reactivity after exposure to trauma, while others
60
reported HPA axis hypo-reactivity. The p-curve analyses here provided support for the
evidence linking trauma with blunted HPA-axis reactivity. The p-curve of the HPA
hyper-reactivity literature was inconclusive in that while it did not show a significantly
right-skewed p-curve, it also did not demonstrate a curve flatter than one would expect if
the studies were powered at only 33%. Therefore, in this case, the p-curve was too noisy
to determine whether a very small effect exists or not. This limitation of the current meta-
analysis may be reflective of lower power among studies reporting HPA hyper-reactivity
compared to those reporting blunted HPA reactivity. Overall then, we can be more
confident in the literature linking traumatic life events to a blunting of cortisol reactivity
given that the p-curve of these studies was significantly right-skewed and these studies
overall had greater statistical power.
In Study 2, we examined the literature reporting relations between moderately

adverse life events and HPA axis reactivity. In contrast to Study 1, we found that the
literature reporting an association between adverse life events and HPA axis hyper-
reactivity contained evidential value, whereas the literature reporting HPA axis hypo-
reactivity after exposure to moderately adverse life events did not. Although we cannot
be certain whether this latter result reflects the absence of a true association or selective
reporting, suffice to say we can have more confidence in the literature reporting HPA
axis hyper-reactivity after exposure to moderately adverse life events than the literature
reporting HPA hypo-reactivity.
The p-curve results of the current analyses support the notion that traumatic
experiences are associated with a blunted reaction of the HPA-system to laboratory
stressors, whereas moderately adverse experiences are associated with heightened
reactivity of the HPA axis. These results are consistent with the predictions made by
Danese and McEwen (2012), in that they reveal that the literature reporting blunted
cortisol reactivity after exposure to traumatic events contains evidential value. However,
Danese and McEwen (2012) predicted that underlying psychopathology moderates the
association between adversity and HPA axis reactivity. Specifically, they expected that
after exposure to adversity, cortisol reactivity would be heightened in individuals with
underlying psychopathology, and blunted in healthy individuals. While we could not
61
completely control for the confound of underlying PTSD, when we removed studies that
explicitly mentioned underlying mental illnesses in their study sample, the results
remained unaltered. Consistent with the predictions of Danese and McEwen (2012), this
demonstrates that in otherwise healthy individuals, exposure to trauma is related to
blunted cortisol reactivity. Similarly, these findings are congruent with predictions that
can be made using the G-A-S model; according to this model, if exposure to an acute
stressor becomes chronic, the HPA axis will move from a hyper-responsive system to a
hypo-responsive one. While the G-A-S model does not precisely describe associations
between HPA axis reactivity and severity of adversity, the assumption can be made that
the body only reaches the “exhaustion stage” reflected by blunted cortisol reactivity if
stressors are either chronic or severe.
In contrast, the p-curve analyses conducted here do not provide direct support for the
BSC model. According to this model, we would have expected HPA hyper-reactivity in
individuals who had experienced traumatic life events, and HPA hypo-reactivity in
individuals who had experienced moderately adverse life events. Our p-curve analyses,
however, provides evidence for the opposite effect. The results lend support to an
inverted U-shaped relationship between severity of adversity and cortisol reactivity,
while the BSC model predicts a U-shaped function. One potential explanation is that HPA
axis reactivity is only one aspect of “biological reactivity”, it could be the case that
biological reactivity as measured by heart rate variability, respiratory sinus arrhythmia
(RSA), or preejection period (PEP), could provide evidence in support of the BSC model.
Furthermore, it could also be the case that the range of experiences that the BSC makes
predictions about are all within “normative” experiences; that is, the BSC may be making
predictions about variability in experiences ranging from highly protected environments
to moderately stressful, but not traumatic environments. Careful examination of the
empirical study used to provide evidence for the BSC model lends support to this idea, as
the measures of adversity used were moderately adverse (e.g., family discord, SES, and
family stress; Ellis et al., 2005). Consistent with the findings of the p-curve analyses, the
authors reported that adversity was associated with heightened cortisol reactivity (Ellis et
al., 2005). According to the BSC paradigm, the paradox of biologically reactive
phenotypes being present under both protected and stressful environmental conditions is
62
reflective of a heightened “sensitivity to context”. This heightened sensitivity is said to be

adaptive to organisms as it results in increased vigilance to threats under stressful
environments, and increased susceptibility to resources under protected environments
(Boyce & Ellis, 2005). While individual differences in genotype may result in differential
sensitivity to the environment, this differential sensitivity does not translate to heightened
cortisol reactivity under extremely stressful environments as defined by exposure to
traumatic events. It could be the case that individuals who are highly sensitive to their
environments experience a form of “helplessness” when faced with extremely traumatic
situations and exhibit blunted HPA axis reactivity.
Based on the literature and the results of these p-curve meta-analyses, there appears to
be an inverted-U shaped relationship between severity of adversity and cortisol reactivity.
Exposure to traumatic life experiences is related to blunted HPA axis reactivity, while
exposure to moderate adversities is related to heightened HPA axis reactivity. The role
that glucocorticoid release plays in glucose metabolism and the body’s expenditure of
resources may partially explain these relationships. When exposed to stressful
experiences, glucocorticoids are produced and result in increased availability of energy
resources to facilitate “fight or flight” responses (Del Rey et al, 2008). Under chronically
stressful or traumatic conditions, “fight or flight” may not be feasible, and thus, the
organism’s physiology may adapt in a manner that does not waste energy resources. If
the organism perceives the situation as completely inescapable, there is little utility in
releasing glucocorticoids and mobilizing energy resources unnecessarily. On the other
hand, a history of exposure to moderate levels of adversity may signal that the likelihood
of success in overcoming the stressors is high. In this case, the organism’s physiology
may adjust in a manner to release energy resources in the face of acute stressors.
The inverted U-shaped relation between severity of adversity and stress reactivity
corresponds closely to Hans Selye’s G-A-S model, which suggests that chronic stressors
will result in the HPA axis moving from a hyper-reactive system to one that becomes
under-responsive (Selye, 1946). The G-A-S model views hypocortisolism as a harmful
response with significant clinical implications (Selye, 1946). However, it may be the case
that the hypocortisolism in response to traumatic experiences reflects the body’s tendency
63
to use biological resources only under appropriate environmental circumstances. The

evidence obtained from the current p-curve analyses provides partial support for the
inverted U-shaped function. Indeed, we found support for the hypotheses that after
experiencing traumatic life events, the HPA axis exhibits blunted reactivity; while after
experiencing moderate levels of adversity, the HPA axis exhibits heightened reactivity.
What remains to be tested however, is how the HPA axis responds under conditions of
little to no adverse experiences.
The finding that the HPA axis becomes differentially reactive according to the
severity of adversity experienced highlights the importance of having clear definitions for
traumatic versus adverse events. Indeed, the lack of consensus regarding definitions of
adversity and trauma may have contributed to the mixed findings reported to date (Bunea
et al., 2017; Fogelman & Canli, 2018; Krupnik, 2019). If clear boundaries are set for
what is objectively considered traumatic versus adverse, we may be able to find more
reliable patterns in the findings reported in the literature. That said, using DSM-5 criteria
for determining traumatic life events is not the only method available. One limitation of
using DSM-5 criteria to define traumatic life events is that it introduces a confound of
stress-related psychopathology into the analysis. In fact, many individuals that experience
traumatic life events (as defined using DSM-5 criteria) will also fit the criteria for the
development of PTSD and other stress-related disorders. We were able to partially
control for this confound in our analyses by excluding studies that explicitly reported
psychopathology in their study samples (See Appendix A: Supplementary Analyses).
However, it is possible that other studies included in the analyses did not directly
measure or report psychopathology within their samples; as such, we cannot be certain
that the participants in those studies did not also have stress-related psychopathology.
Nevertheless, not all individuals who experience traumatic life events will develop PTSD
or other stress-related disorders. Therefore, using DSM-5 criteria to define traumatic
events may differentiate trauma from adversity, but it may or may not introduce the
confound of psychopathology into the analysis.
A recent model suggests that any stressor can be traumatic if its severity overwhelms
self-regulatory processes (Krupnik, 2019). The model proposes that individuals with
64
compromised or under-developed self-regulation (e.g., children) are most vulnerable to

becoming traumatized; this view of trauma may help explain sensitive periods in HPA
axis calibration (Krupnik, 2019). Future research should investigate whether there are
age-related differences in the HPA axis reactivity as a function of exposure to traumatic
versus adverse life events. If the hypothesis that individuals with under-developed self-
regulation are more vulnerable to the effects of life-experiences is correct, children who
have experienced moderate adversity may demonstrate blunted cortisol reactivity. In
other words, only moderate levels of adversity may be sufficient to produce a biological
profile in children akin to what adults demonstrate after traumatic events. Finally, the
analyses here investigated literature reporting associations between traumatic versus
adverse life experiences and HPA axis reactivity. Basal HPA axis function as measured
by CAR, however, could provide different results. It could be the case, for example, that
traumatic life experiences result in heightened basal cortisol levels, but blunted cortisol
reactivity (Danese & McEwen, 2012). Future research should investigate the literature
assessing associations between life experiences and basal cortisol measures as it may help
further clarify the contradictory results in the field.
2.8.1 Conclusion
Understanding how life-experiences become biologically embedded (Hertzman,
1999) is of great interest to researchers. The HPA axis is one biological system that
exhibits plasticity to environmental conditions and has been shown to become
differentially calibrated as a function of life experiences. Various theoretical models
make different predictions about how life experiences are related to HPA axis reactivity.
Empirical evidence investigating the relationship between life-experiences and HPA axis
reactivity, however, is mixed. P-curve is the distribution of statistically significant p-
values for a set of findings and can be used as a powerful tool when literature on a given
topic provides mixed results. P-curve can determine which subset of contradictory results
contains evidential value. Here, we used p-curve as a tool to determine how traumatic and
adverse life experiences are related to HPA-axis function. We found that the literature
reporting an association between traumatic experiences and cortisol hyper-reactivity was
inconclusive, while the literature reporting an association between traumatic experiences
65
and cortisol hypo-reactivity contained evidential value. Moreover, when we assessed the
literature reporting associations between adverse life experiences and cortisol reactivity;
we found the opposite pattern of results. That is, we found that the literature reporting an
association between moderately adverse life experiences and cortisol hyper-reactivity
contained evidential value, while the literature reporting cortisol hypo-reactivity did not.
These results help clarify the mixed findings in the literature and they provide support for
an inverted U-shaped relationship between severity of adversity and HPA axis reactivity.
Moreover, the results of these analyses highlight the importance of having clear
definitions and conceptual boundaries for defining the terms “adversity” and “trauma”.
66
Table 2.1: P-curve Disclosure Table of Trauma-Cortisol Hyper-reactivity Literature

Original Quoted text from Sample Type of Trauma Stress Induction Cortisol Data Analysis Plan Quoted text from Results P- Conclusion quoted
Paper original paper Characteristics Task Measurement original paper value from original text
indicating with statistical
prediction of results
interest
Kern & Laurent We expected that Participants were the CTQ: The CTQ uses 25 The Strange Salivary Cortisol: Each Childhood Table 2 - CTQ total r(45) = .034 As predicted, early life
2019 women’s reported 47 women from a items scored on a 5- Situation (SS) woman contributed four maltreatment correlation with T3 .31 adversity appeared to
history of childhood larger longitudinal point Likert scale involves seven 3- saliva samples via passive reported at 3 months Cortisol Sample 1 contribute to ongoing
maltreatment at 3 study of (from 1 = "never" to 5 minute episodes in drool: the first soon after postnatal—the total symptom elevations
months postnatal mother-infant stress = "very often") to which the infant is arrival, the second score and, if a via heightened HPA
would predict regulation who assess the occurrence twice separated immediately following the significant overall reactivity to stress
greater cortisol completed all during childhood of from the mother— completion of the SS effect was detected,
reactivity to measures involved in five forms of first with an procedure, the third 20 min individual subtypes—
mother-infant stress the current study. maltreatment: unfamiliar female after the start of the second was used to predict
at 12 months physical abuse, “stranger” in the separation from the mother, women’s cortisol
postnatal, which emotional abuse, room, then with no and the fourth 30 min after response trajectories
would predict sexual abuse, one in the room— the preceding sample. during the mother
higher affective physical neglect, and before being infant stress task at
symptoms emotional neglect. reunited; during 12 months postnatal
(controlling for 12- separations, using multilevel
month symptoms) women were able modeling in HLM.
at 18 months to watch their
postnatal. infant on a remote
computer monitor.
This task has been
shown to elicit
behavioral and
physiological stress
responses in infants
and their mothers.
Shalev et al We hypothesize A total of 427 Bereavement: The TSST: Bereaved and Salivary Cortisol: Salivary We examined the Table 3 - t-test for t(221) = .003 As hypothesized, we
2019 that bereavement bereaved and non- deceased parents nonbereaved cortisol samples were relationship of Bereaved vs. Non- 3.05 found that parental
and suicidal bereaved youth were (probands) died offspring obtained immediately cortisol measures bereaved and bereavement was
ideation will each, enrolled in the study, within 24 h of either participated in a after the social stress task with demographic Cortisol Reactivity associated with greater
independently be of whom 223 (52%) suicide (n = 29), modified version of (15 min after baseline), and and clinical variables cortisol reactivity.
related to baseline completed the acute accidental death (n = the Trier Social 5, 10, 20 min after the social to identify significant
cortisol, total laboratory social 17), or sudden Stress Test (TSST), a stress task (20, 25, 35 min covariates to include
cortisol output, and stress task and were natural death (n = procedure designed after baseline). in our regression
cortisol reactivity in available for follow- 33). to induce a models (Table 1) and
response to the up. The 223 youths moderate stress controlled for
TSST, and that included 114 offspring response. The TSST covariates known to
cortisol measures from 79 parentally- was conducted on potentially affect
bereaved families and average 6 years (SD cortisol levels. We
67
will predict future 109 offspring from 65 = 1.3, range: 2.8– used Bonferroni
suicidal ideation. nonbereaved families. 9.2) after correction to account
bereavement in for multiple
bereaved youth. comparisons.
Chen et al 2018 Our main goal was Participants were 80 Olweus Bully/Victim TSST: The TSST is an Salivary Cortisol: Six saliva Because there were A significant main F(1,78) .009 The current study of
to examine whether children (4-5th grades, Questionnaire: We empirically- samples were collected from no sub-type effect of bullying = 7.07 Chinese children also
cortisol levels 45% girls) with a mean used six items supported protocol each participant, to measure differences, and there victimization was provided evidence for
during baseline, age of 10.83 (± 0.70 involving physical, to elicit the cortisol response to the were significant found, F(1, 78) = a link between bullying
stressful, and SD) years, selected verbal, and relational psychosocial stress, TSST across a 70 min period. correlations in 7.07, p = 0.009, ŋ2 = victimization and
recovery periods of from 970 students in victimization from developed by The first sample (T1) was subtype membership 0.08, Φ = 0.75, cortisol levels, in the
the TSST differed 18 classes in two the Olweus Kirschbaum et al. collected at 5 min after (r- s = 0.291~0.430, ps which indicated that context of the TSST.
for bullied and elementary schools. Bully/Victim (1993). It involves arrival in the room next to < 0.001), we cortisol levels during Children with a history
nonbullied youth, in Sixty-five victims Questionnaire. Each public speaking and the experiment room, combined them into TSST were higher for of victimization had
a sample of Chinese actively agreed to of the three mental arithmetic, during the 10 min rest one larger group of bullied than higher cortisol than
children. We participate in the subscales consisted as stressors. period. The second sample bullied youth, and in nonbullied children. those without a
hypothesized stress test; they were of two items. Sample (T2) was collected when our final analysis we history.
finding a significant selected from 120 items include: "being participants were about to compared them to
association identified victims. hit, kicked, or slapped finish the preparation of non-bullied youth.
between greater Meanwhile, a by others" (physical), their speeches. The third Repeated-measures
bullying nonbullied group of Being laughed at by sample (T3) was collected analyses of variance
victimization and 25 children was others(verbal), and immediately at the end of (ANOVAs) was used
stronger cortisol randomly selected Being excluded by the mental arithmetic task. with the six
reactivity during the from the pool of others The other three samples (T4, measurements of
TSST, such that nonbullied children (relational/social). T5, and T6) were collected cortisol representing
victims would have who had no separately at 10 min, 25 time as a within-
higher cortisol victimization in the min, and 40 min after the subject factor, and
reactivity than non- previous semester; end of the arithmetic task. bullying victimization
victims during they matched the status (bullied versus
stressful and bullied group in age nonbullied) as a
recovery periods. and gender between-subject
distributions. variable
Coppens et al Studies on the Twenty-seven female CTQ: Participants TSST: The TSST is Salivary Cortisol: Salivary The group-by-time Significant main F(1,43) .028 Cortisol responses
2018 association FM patients and 24 completed the the gold standard cortisol was assessed at interaction effect in effects of group = 5.18 were higher in
between ECA and age-matched female Childhood Trauma laboratory baseline (T0), immediately these models was [F(1,43)=7.04, participants with ECA.
HPA axis reactivity controls were Questionnaire – short procedure to before the TSST (after a followed up by a p=0.011, lower in
in FM are lacking, recruited in a tertiary form (CTQ-SF), a well- induce psychosocial relaxation/waiting period of priori planned FM] and ECA
but in other FSS, an care center and validated 25-item stress; it consists of 30 minutes) (T1), contrasts using [F(1,43)=5.18,
association was through self-report a public speech (5 immediately after the TSST independent samples p=0.028, higher in
found between ECA advertisements, instrument, to minutes) and an (T2), and 10 (T3), 20 (T4), 30 Student’s t-tests, with participants with
and HPA axis respectively. measure exposure to arithmetic task (5 (T5), and 75 minutes after step-down Bonferroni ECA] were found for
responses during ECA. minutes) performed the TSST (T6). correction for cortisol responses.
the TSST; however, in front of an multiple testing.
both decreased and evaluating audience
increased responses (two PhDs or MSc
were found in students wearing
individuals with white laboratory
ECA, again stressing coats).
the need for further
research.
68
Mielock et al We hypothesized The sample included The Childhood TSST: involved a 2- Salivary Cortisol: Pre-stress To examine within- The maltreatment X t(24) = .003 Maltreated women
2017 that cortisol 26 women with a Adversity Interview hour pre-stress saliva samples were and between- MDD X time 3.33 showed greater
responses would be history of (CAI): is a semi- period, a 5-minute collected at 30-minute individual changes interaction (i.e., anticipatory cortisol
higher in maltreated maltreatment before structured interview public-speaking task intervals for 2 hours prior to simultaneously, we anticipatory reactivity during the
women with current the age of 10 (15 with that was (following a 5- the stress task, with the final specified a multilevel reactivity) was TSST protocol
MDD and blunted in current MDD) and 26 administered by minute preparation sample provided model (MLM) using significant (b = compared to non-
maltreated women women with no trained raters. Seven period) followed by immediately prior to the Hierarchical Linear −.017, SE = .009, p = maltreated women.
without current history of subtypes of a 5-minute mental stress tasks (five samples). Models (HLM v. 6) .046). Simple slope Maltreated women
MDD compared to maltreatment (17 childhood adversity arithmetic task Post-stress saliva samples consisting of a within- analyses revealed also showed rapid
non-maltreated with current MDD). occurring before age performed in front were collected immediately person (i.e., level 1) stronger anticipatory deceleration in cortisol
women. 10 were assessed: of an audience, and after the tasks and at 10- sub-model describing cortisol reactivity for levels. Whereas non-
separation/loss; life- a 60-minute post- minute intervals for 60 cortisol and alpha- maltreated women maltreated women
threatening illness or stress period. minutes (seven samples). amylase responses to without current showed initial declines
injury; physical each TSST and MDD (b = .020, SE = . in alpha-amylase levels
neglect; emotional between-person (i.e., 006, p < .001) but rapidly increasing
abuse; physical abuse level 2) sub-model compared to those alpha-amylase levels
or assault; witnessing describing how with current MDD (b during the TSST
domestic violence; cortisol/alpha- = .014, SE = .005, p = protocol, maltreated
and sexual abuse or amylase responses to .005). In contrast, women did not exhibit
assault. each TSST varied non-maltreated changes in alpha-
between women with (b = amylase levels during
maltreated/non- .003, SE = .005, p = the TSST protocol.
maltreated and .513) and without Contrary to
current MDD/non- current MDD (b = expectation, MDD did
MDD groups. −.008, SE = .005, p = not impact cortisol or
.110) did not exhibit alpha-amylase
anticipatory cortisol responses.
reactivity.
Flanagan et al This exploratory Our sample was The Adverse TSST: Participants Salivary Cortisol: Enzyme- To estimate the effect Results indicate that t(115) = .0138 Results indicate that
2015 study addressed comprised of 31 Childhood performed the Trier linked immunosorbent of administration of there is a significant 2.5 ACE scores and cortisol
this gap in the cocaine-dependent Experiences (ACE) Social Stress Task assays were used to oxytocin or placebo positive association reactivity to the TSST
literature by individuals who Survey: is a 10-item (TSST)) and were measure salivary cortisol on the response to between baseline were positively
examining whether responded to local self-report survey told that at 2:00 and DHEA levels. the social stress task; ACE scores and correlated in our
ACE severity media used to assess 10 (s)he would soon linear mixed effects cortisol response sample.
influenced the advertisements. domains including perform to an models that use all levels in the
effects of intranasal Inclusion criteria childhood abuse audience a speech serially measured untreated placebo
oxytocin on included current (emotional, sexual and arithmetic task. time points following group (β = 0.24 7
measures of HPA cocaine dependence and physical), neglect administration of the 0.09, t115 = 2.5, p =
axis function (e.g. consistent with DSM- (emotional and study drug tested the 0.014).
salivary cortisol and IV diagnostic criteria. physical) and neuroendocrine
Dehydroepiandrost household responses to
erone [DHEA]) to a dysfunction intranasal oxytocin
social stress (domestic violence, compared to placebo
laboratory paradigm substance abuse, as well as the
among cocaine- mental illness, modification of the
dependent criminal household oxytocin effect by
individuals. increased ACE scores
69
member and parental (treatment ACE

divorce/separation). interaction).
Jaffee et al We hypothesized The sample included Recent Traumatic Social Provocation Salivary Cortisol: Saliva was An ordinary least Tests revealed the t(53) = .01205 The more traumatic
2015 that with consistent 400 children (51% Events: Recent Task (SP): The social collected at 20 and 10 min squares regression predicted effect of 2.6 events children had
exposure to harsh, male) who traumatic events (at provocation (SP) prior to the start of the SP analysis was recent trauma in the recently experienced,
rejecting, or participated in the follow-up) were task was designed task, at the end of the SP conducted in which group that the greater their
dangerous Children's Experiences measured with the to elicit feelings of task, and at approximately cortisol reactivity experienced lower cortisol reactivity to
relationships and and Development Traumatic Events frustration and 20, 45, and 65 min following values were regressed levels of harsh, the social provocation
environments Study (CEDS), which Screening Inventory. anger in children the conclusion of the SP on harsh, nonresponsive task, but only if
across early and was conducted from Events were coded as and was modified task. Because cortisol nonresponsive parenting at age 3 children had
middle childhood, 2009 through 2011 in having happened from a task concentrations at time t parenting, traumatic (θ = .13, SE = .05, p < experienced lower
the cortisol England. Participants only if both the child developed by van reflect cortisol activity t – 15 events, and .01); the more levels of harsh,
response to ranged from 8 to 11 and caregiver Goozen et al. to t – 30 min earlier, cortisol covariates traumatic events nonresponsive
experimentally years (M = 9.99, SD = reported the event (1998). reactivity to the SP task was (medication status, youth had recently parenting in early
induced stressors 0.74). We sampled a (M = 0.61, SD = 0.85). assessed by subtracting the the previous night's experienced, the childhood.
would be relatively ‘higher risk’ group In the past year, 58% lowest cortisol value bedtime, and time of higher their cortisol
low. In contrast, we that included children of children collected by the end of the initial saliva sampling) reactivity.
hypothesized that in who had experienced experienced no SP task (which reflected at the first step and
the absence of an relatively high levels traumatic events, cortisol activity prior to or in on the interaction
early childhood of harsh, 27% one event, and the very early stages of the between harsh,
history of harsh, nonresponsive 15% 2–4 events. SP task) from the cortisol nonresponsive
nonresponsive parenting at age 3 value collected 20 min parenting and
parenting, recent years. A ‘lower risk’ following the SP task (which traumatic events at
exposure to group included reflected cortisol activity at the second step.
traumatic events children who had the conclusion of the SP
would be associated experienced relatively task).
with increased low levels of harsh,
cortisol reactivity. nonresponsive
parenting at age 3
years.
Kuhlman et al Here, we aimed to Participants were 121 Early Trauma Socially -Evaluated Salivary Cortisol: Seven 4 x (adversity In unadjusted t(439.5) .01636 In the reactivity task,
2015 identify the youth (51% male), Inventory: The Cold Pressor Task: samples were collected via subgroup) x 7 (time). models, more = 2.41 more exposure to
contribution of ages 9–16 (M parent indicated The stress task used passive drool at 30 min To determine the physical abuse was physical abuse was
physical abuse, age = 12.8; SD whether their child in this study was before the task (−30), association between related to greater associated with a
emotional abuse, age = 2.3) from a had been exposed to the Socially- immediately before the task childhood trauma acceleration as steeper activation
and non-intentional study aimed to 50 potentially Evaluated Cold (0) and at 25, 35, 45, 55, and subtypes and acute individuals slope, which is
trauma to multiple characterize the traumatic events. Pressor Task, which 65 min after the start of the stress reactivity, we approached their consistent with
indices of mechanisms that Four subtypes: non- was specifically task. used a modified peaks, previous studies
neuroendocrine underlie adolescent intentional traumatic designed and version of Growth PA × Time2b = .0001, showing that childhood
functioning (CAR, anxiety and events (witnessing an validated for the Curve Analysis using t(439.5) = 2.41, exposure to violence is
diurnal regulation, depression. accident); physical activation of the landmark registration p = .02. associated with
acute reactivity) abuse (being hit to HPA-axis in (Lopez-Duran et al., hyperreactivity to
hypothesizing that the point of bruising laboratory settings 2014), where we acute stress.
each trauma or injury); sexual by combining modeled slope of
subtype would be abuse (being forced thermal stress and cortisol activation to
associated with to engage in sexual social evaluation. the stress task, peak
distinct anomalies acts); or emotional cortisol (intercept),
in neuroendocrine abuse such as and the slope of
functioning. persistently being recovery from the
70
ridiculed or insulted stress task

by a caregiver). simultaneously.
Dietz et al 2013 It was hypothesized The participants were Parental Loss: The TSST: Offspring Salivary Cortisol: Salivary Group (2) x Time (4) The total cortisol F(1,179) .01407 Bereaved youth
that parentally 181 youth, aged 10 to deceased parents watched a 10- cortisol samples were ANOVA. ANOVA was secretion, as = 6.15; showed overall higher
bereaved children 29, from 62 parentally (probands) died minute travel video obtained immediately after used to assess group measured by AUC, Robust- total cortisol output
would evidence a bereaved families and within 24 hours of during which the social stress task (15 differences in total was higher in the ness but showed no
significantly 53 nonbereaved definite verdicts of baseline cortisol minutes after baseline), and cortisol output. bereaved than in the t(40) = significant increase in
different trajectory families; 66.3% of suicide (n = 21), sample was 5, 10, and 20 minutes after nonbereaved groups 2.32 cortisol levels over
of cortisol response participants had at accidental death (n = obtained (0 the social stress task (20, 25, [F(1,179) = 6.15, p = time in response to
compared with least 1 sibling in the 13), or sudden minutes). For the 35 minutes after baseline). .01]. social stress. Within
nonbereaved study. natural death (n = next 15 minutes, Cortisol response was the bereaved group,
controls, marked by 28). Nonbereaved participants were operationalized as changes offspring bereaved due
greater reactivity to offspring had two asked to prepare in cortisol levels across the to sudden natural
acute stress. In this living biological and deliver a brief five time points during the death showed a
article, psychiatric parents, lived in the speech and then laboratory procedure. significant increase in
disorders associated home of at least one perform a mental cortisol relative to
with cortisol of them, and had no arithmetic task baseline.
response in first-degree relatives while being
offspring were who had died within observed by
considered possible the 2 years before research staff.
confounds and were recruitment.
controlled to isolate
the influence of
bereavement on the
acute stress
responses of
participants.
Bosch et al We examined Complete HPA-axis Pre/postnatal Social Stress Test: Salivary Cortisol: HPA-axis Repeated-measures Figure 1 Subjects F(1,441) .0046 The transition in
2012 adversities during and adversity adversity, adversities The experimental responses towards the social general linear models with the highest = 8.08 cortisol activity from
five age periods measurements were during ages 0—5 session during T3 stress test were assessed by (GLM, Greenhouse— number of late hypersecretion after
(pre/postnatal, 0—5 available for 640 (early childhood) and consisted of a four cortisol samples, before Geisser corrected) childhood adversities before age
years, 6—11 years, participants. 6—11 (middle number of different the introduction of the were used to examine adversities (ages 6— 11 and hyposecretion
12—13 years, 14— Adolescents were childhood), challenges social stress test (‘pretest’), (1) the main effects of 11) displayed the afterwards emphasizes
15 years), to excluded if stressful adversities during (orthostatic stress, directly after the test adversities per period highest cortisol puberty as a major
investigate their life events were ages 12—13 (early spatial-orienting (‘during test’), 20 min after and (2) the inter- overall level (mean developmental period
specific effects on assessed more than 3 adolescence), and task, gambling task, the test (‘end of test’), and action of level of cortisol of the HPA-axis.
the HPA-axis. We months before (n=60) adversities during startle-reflex task, 40 min after the test (‘post pre/postnatal risk across the four Puberty may be
anticipated that or after the ages 14—15 (middle and social stress test’). with each later cortisol marked by a transition
effects of experiment (n = 60). adolescence) were test). The social adversities on the measurements, in how adversities
adversities on the We excluded measured via semi- stress test HPA-axis. F(1,441) = 8.08, p = affect the HPA-axis,
cortisol responses adolescents with structured (Groninger Social .005). with cortisol
would depend on cortisol outliers (>3 interviews. Duration Stress Test) hypersecretion before
their timing, and SD, n = 11), and who of the adversities was involved a age 11 and
that pre/postnatal used corticosteroids measured in months. standardized hyposecretion after
adversity would (n = 38), leaving a protocol inspired by age 11.
render individuals total of 471 (32.3% the Trier Social
more sensitive to low- risk) participants Stress Test
persistent effects of for analysis. (Kirschbaum et al.,
1993).
71
later adversities on
the HPA-axis.
Gola et al 2012 Here we investigate Cortisol reactivity was War and rape Traumatic Plasma and Salivary Salivary and plasma Differential cortisol t(27) = .02634 Trauma groups had
whether the type of assessed in 30 trauma: To Reminders: Cortisol: Plasma and salivary cortisol reactivity responses were 2.35 comparable basal
trauma (rape vs. individuals with investigate the Participants were cortisol levels of PTSD were analyzed using observed in PTSD salivary or plasma
other trauma types) current PTSD influence of trauma interviewed in a patients were measured at mixed model ANOVAs patients who had cortisol levels,
is associated with according to DSM-IV. type on cortisol standardized three time points during a with trauma type been raped however, our results
differential stress All PTSD patients were reactivity, patients manner about their detailed and standardized (raped vs. non-raped) compared to those demonstrate
responses when refugees (4 from were divided into two individual child and clinical interview, starting at or trauma type and who had not been differential salivary
confronted with Africa, 5 Balkan, 21 groups: raped (n = adulthood 10:00 a.m., once before and gender or current raped [t2, t(27) = cortisol responses in
trauma-related Middle East and 10; 6 female, 4 male; traumatic twice after assessing depression as fixed 2.35, p < .05]. PTSD PTSD patients who had
material, using Afghanistan) with mean age = 34.0 experiences using individual rape, war and effect and time (t1, patients who had been raped compared
cortisol as a multiple highly years, SD = 10.2) vs. the event checklist torture experiences. t2, and t3) as been raped showed to those who had not
physiological stress stressful war and non-raped (n = 20; 8 of the CAPS and repeated measure a significant cortisol been raped.
marker. torture experiences, female, 12 male, the vivo checklist of factor. increase when Consistent with a
including or not mean age = 33.7 war, detention and reminded of their typical diurnal decline,
including rape. To years, SD = 8.7). Rape torture event which traumatic events (t1 salivary cortisol levels
compare basal cortisol was defined as any assesses common vs. t2, p < .001), and decreased in the
levels of the PTSD traumatic stressor traumatic returned to baseline course of the interview
group with a non- involving unwanted experiences in levels at t3 (t2 vs. t3, for the group with no
PTSD group, we vaginal or anal conflict regions and p = .11; t1 vs. t3, p = past experience of
further examined a penetration obtained during torture. .92). rape, whereas those
third healthy control by the use of force or PTSD patients who had
group (n = 28; 17 threat of force. been raped showed a
female, 11 male; significant salivary
mean age = 26.9 cortisol increase when
years, SD = 7.0). reminded of their
traumatic events.
Harkness et al This study examined Participants were 71 CECA: Participants TSST: We followed Salivary Cortisol: 5 (time) x 3 A significant 3-way F(2, 63) .03997 Consistent with
2011 the hypothesis that individuals (48 were interviewed precisely the Participants were asked to (depression group: interaction of time = 3.39; hypotheses, a history
depressed females) ages 12—21 using the Childhood procedure outlined refrain from eating and minimal depression, by depressed group Robust- of childhood
adolescents with a years (M = 15.39, SD = Experience of Care by Krischbaum, drinking for 1 h prior to their mild/moderate by childhood ness maltreatment was
history of childhood 2.11). The depressed and Abuse contextual Pirke, and arrival at the lab. All saliva depression, maltreatment in the F(2,63) associated with
maltreatment will adolescents all met semi-structured Hellhammer (1993) was collected in previously moderate/severe quadratic trend, F(2, = 3.41 significantly higher
show greater Diagnostic and interview and rating in defining the time labeled 5 ml polypropylene depression) x 2 63) = 3.39, p < .05, cortisol reactivity and
cortisol reactivity to Statistical Manual of system (CECA). points for salivary vials by passive drool (childhood h2 = .10. total cortisol exposure
psychological stress Mental Disorders sample collection between the hours of 3 pm maltreatment: (i.e., AUC) to a
challenge than (DSM-IV) criteria for a during the TSST. and 5 pm because this is a presence versus psychosocial stress
those without, and current non-bipolar, period of low cortisol absence) repeated challenge.
this relation will be non- psychotic mood relative to the morning. measures analysis of
moderated by level disorder. covariance
of depression (RMANCOVA)
severity. controlling for age
and parental
occupation status.
72
Hibel et al 2011 Here we extend the Data for this analysis IPV: To assess Challenge Task: Salivary Cortisol: To assess The main analyses Children that were F(2, .00119 The analyses presented
previous analyses came from the larger violence between Lab-TAB- Challenge changes in cortisol indicative employ a series of exposed to IPV 1089) = examined the
by examining the project’s home partners, the Conflict tasks designed to of the child’s adrenocortical mixed-model exhibited a 4.45 contributions of IPV on
influence of IPV on interview and Tactics Scale—Couple elicit emotional response to the emotion repeated-measures significantly Robust- adrenocortical levels,
cortisol levels, assessments at three Form Revised (CTS- reactivity were challenge tasks, three saliva ANCOVAs different response ness reactivity, and recovery
reactivity, and time points across CF-R) was administered to the samples were collected: a determining the to the tasks than F(1, 99) to a set of emotion
regulation across early childhood: when administered. The child. At the early pretask sample collected relationship between those not exposed, F = 4.45 eliciting challenge tasks
early infancy, later the child was CTS-CF-R consists of infancy visit, three prior to administration of violence exposure (2, 1,089) = 4.45, p < when the child was 7,
infancy, and approximately 7 19 items exploring tasks were the challenge tasks, a and physiology across .01. Children 15, and 24 months of
toddlerhood. We months, 15 months, conflict and violence presented to the sample 20 min after the final early infancy, later exposed to higher age. Adrenocortical
determine if and 24 months of age. between the partner child: the barrier task was completed (or infancy, and levels of physical dysregulations first
children exposed to From the total of and the mother in task, the mask task, prior, if the child reached toddlerhood. violence at 24 appeared at 24
IPV exhibit 1,292 dyads eligible, the past 12 months. and the arm “peak emotional arousal”), months had higher months, revealing
divergent patterns 1,102 participated in The CTS-CF-R restraint task. At and then a final sample 20 cortisol reactivity toddlers exposed to
of cortisol reactivity all three home visits contains three the later infancy min later. The child was from pretask to high accumulated
and recovery across and are thus included subscales: verbal and toddlerhood considered to have reached posttask. levels of IPV across
early childhood. in the analyses. discussion, verbal visits, two tasks peak arousal if he or she their first 2 years had
aggression, and were presented to produce. higher cortisol
physical aggression. the child: the toy reactivity than those
removal task and with little to no
then the mask task. violence exposure.
Ivanov et al This protocol tested We identified 30 To assess traumatic Violent film: Since Salivary Cortisol: Cortisol 2 (group: adversity vs The main effect for F(1,23) .04368 The main effect of
2011 the hypothesis that children ages 8–12, stress exposure, the TSST presents samples, pulse and blood no adversity) x 2 (film group was registered = 4.556 group (Trauma vs.
traumatized and who have exhibited caretaker filled out: i) stressful stimuli pressure were collected type: violent vs. non- during the Violent Non-Trauma) showed a
aggressive youths patterns of aggressive the Codington Life unrelated to at baseline just before the violent) x 4 (time) film at T3 (F= 4.556, significantly
will exhibit higher behavior, indicated by Events Scale for the personal trauma start of each film (T1), 15 ANOVA. d.f.= 1/23, p= 0.044), heightened HPA
levels of school suspension due purpose of inquiring history we min into the viewing (T2), at indicating response pattern for
responsiveness to to engaging in about the occurrence introduced the end of the viewing (T3), significantly the Trauma children
the active physical fighting. Of of traumatic emotionally laden and 15 min post viewing heightened that was most
psychological these 25 children experiences within stimuli that we (T4). cortisol response in pronounced during the
challenge (Violent completed the the past 12 months; considered relevant Trauma youth Non-Violent film.
film) than experimental protocol and ii) the to some aspects of compared to Non- Although the Violent
aggressive youth and provided data Posttraumatic Stress these children's Trauma film elicited higher
with no trauma included in the Reaction Index experiences (e.g. counterparts. cortisol response than
whereas the analyses. (PTSRI) Parent physical aggression) the Non-Violent film in
responsiveness to version, which and used two types the Non-Trauma
the neutral queries the parent of films as aggressive children,
psychological about the occurrence psychological these responses were
challenge (Non- of traumatic probes. The Violent overall lower than the
Violent film) will not experiences in the film (VF) involved responses to either
differ between the child and the nature scenes depicting film in the Trauma
two groups of the reactions to preparation for and group, pointing to an
them, following the enactment of a overall heightened HPA
DSM criteria for boxing match. The responsiveness in the
PTSD. Children also Non-Violent film Trauma children.
completed the PTSRI, (NVF) depicted
Child version. everyday family
73
interactions in a
situational comedy
format.
Elzinga et al The aim of the Four groups: (1) Social Traumatic TSST: The Salivary Cortisol: measures 4 x 10 repeated A main effect of F(3,46) .01258 We found that a
2010 present Anxiety Disorder + Experiences Checklist psychological were obtained at 10 measures ANOVA. To Group was found = 4.03 history of childhood
retrospective study Childhood Abuse (TEC): A reliable and challenge test assessment points over a compare the four (F(3,46) = 4.03, p < abuse (CA) was
is to investigate HPA [n=9]; (2) Social valid self-report consisted of a free 130-min period, at -65, -35, - groups on cortisol 0.05, η2 = 0.21), associated with greatly
reactivity to a Anxiety Disorder - inventory that speech (5 min) and 25, -10, 0, +15, +25, +40, reactivity to the TSST, which was specified enhanced cortisol
psychosocial Childhood Abuse assesses emotional a mental arithmetic +50, and +65 min with ANOVA for repeated by a significant reactivity to a
stressor (TSST) in [n=9]; (3) PTSD [n=16]; abuse and neglect, task (5 min), reference to th estart of the measures were Group × Time psychosocial stress task
patients with SAD (4) Healthy controls physical abuse, performed in front stressor. All assessments conducted with interaction in patients with social
who report a history [n=16]. sexual harassment, of an audience of were performed between 1 Group as between- (F(27,414) = 2.62, p anxiety disorder (SAD)
of (emotional, and sexual abuse, as three individuals, PM to 4 PM. subject factor and < 0.05, η2 = 0.15). compared to patients
physical, or sexual) well as general which was Time as within with SAD who did not
childhood abuse vs traumatic events, preceded by an subjects factor. 4 report childhood
patients with SAD including loss of anticipation phase (Group: SAD + CA vs abuse, and compared
who do not report significant others, life (5 min) in which SAD - CA vs PTSD vs to patients with PTSD
experiences of threat by disease, participants Controls) x 10 (Time: related to CA and
abuse during parental divorce, and prepared the 10 assessment points) healthy controls
childhood. psychopathology of speech. repeated measures without a history of CA.
parents (i.e., alcohol ANOVA
or drug abuse).
Pesonen et al Aims: The Finnish The participants came Childhood TSST: Briefly, after Salivary and Blood Cortisol: To illustrate HPA axis Analyses using the t(152) = .029 We have shown that
2010 experience in World from the Helsinki Birth Separation standing for 5 min We obtained saliva (by response with mixed model (Table 2.2 ELS is associated with
War II created a Cohort Study (N = Experiences: In the to register the Salivette1) and blood at conventional 2) show that, altered responsiveness
natural experiment 8760). In years 2001— current sample, 68 cardiovascular baseline (prior to entering summary measures, compared with the of the HPA axis more
to test whether 2004, 2003 randomly participants were in baseline, subjects the laboratory, we used baseline, non-separated, than 60 years after
separation from a selected participants the ‘‘separated from were given 3 min to approximately 18 min post-stress peak those who were childhood separation.
father serving in the attended a clinical both parents’’ group. prepare a 5-min before the TSST stressor value, increment separated from both In comparison to non-
armed forces or examination. rom the The ‘‘father- speech. After the ended and at 0, 10, 20, 30, (post-stress peak parents had higher separated participants,
from both parents subset who attended separated’’ group speech, they were 45, 60, and 90 min after the minus baseline value, salivary cortisol and individuals separated
due to war the 2001—2004 included 129 non- asked to perform end of the stressor). an indicator of HPA plasma ACTH from both parents at a
evacuation are examination, we evacuated a series of serial axis responsiveness), concentrations mean of 3 years of age
associated with invited 407 participants who subtractions for time- weighted across all time points displayed 20—25%
alterations in HPA participants selected reported that their another 5 min. The AUCg), and time- during the TSST. In higher salivary cortisol
axis response to by random-number fathers served in the committee, blind to weighted AUC addition, and plasma ACTH
psychosocial stress tables to participate in Finnish armed forces participants’ increment (time- participants levels across the time
in late adulthood. the Trier Social Stress during these wars. separation weighted AUCg minus separated from both points during the TSST,
Our primary Test (TSST), The ‘‘non-separated’’ experiences, the baseline value, an parents had greater and higher salivary
hypothesis was that performed during group included 85 minimized all verbal indicator of excess salivary cortisol cortisol reactivity in
individuals who 2004—2005. Of these, participants who did and non-verbal HPA axis output reactivity to the response to the TSST,
have undergone 287 (144 men and 143 not experience communication because of stress).. TSST. more than 60 years
separation as child women) agreed to separation from their with the subject. We tested whether later.
evacuees have participate, parents because of these indicators are
different HPA axis corresponding to either war. associated with
responses as adults 70.5% of those separation
to psychosocial invited. We had experiences by
stress. separation data
74
available for 282 of multiple linear

those participants. regression.
Rao et al 2008 The current study 30 adolescents with The Child Adversity TSST: A modified Salivary Cortisol: Baseline 2 (group) x 7 (time) Among experiential t(53) = .00038 HPA response to the
extends findings by depression and 25 Interview: Seven version of a saliva samples were Repeated measures factors, early-life 3.8 stressor was highest in
systematic controls were types of adversity standard collected at 30-minute ANOVA. adversity and those who had a
evaluation of early- recruited. The including psychosocial intervals for two hours (5 chronic stress during combination of early-
life and recent depressed separation/loss, life- stressor used in samples). Saliva samples adolescence were life adversity and high
stress on HPA adolescents met threatening adults, the Trier were obtained immediately significant levels of chronic stress
response to a criteria for major illness/injury, Social Stress Test after the task and at 10- predictors. Table 2 - during adolescence.
psychosocial depressive disorder physical neglect, (TSST) was minute intervals for 60 Predictors of net
stressor in (MDD), with a emotional administered to 30 minutes (7 samples). The peak salivary cortisol
depressed and non- minimum duration of abuse/assault, adolescents with time of HPA assessment was concentration in
depressed four weeks and a physical major depressive in the late afternoon/early response to the TSST
adolescents. score of ≥15 on the abuse/assault, disorder and 25 evening, and was based on - early-life adversity
first 17-items of the witnessing violence, healthy adolescent prior data suggesting a
Hamilton Depression and sexual volunteers. delayed circadian phase in
Rating Scale (HDRS). abuse/assault were adolescents.
assessed.
van der Hal- We expected the Participants were 203 Child Survivors of the Stressful Task: Salivary Cortisol: Three Analysis of covariance The interaction F(1,97) .049 We noticed
Van Raalte et youngest Holocaust child Holocaust Holocaust and PTSD Through the saliva samples were between the two between PTSD = 3.97 neuroendocrine effects
al. 2008 child survivors, who survivors, who were Functional questionnaire collected at 20 min intervals PTSD groups with functional in stress reactivity
lacked the prewar born between 1935 Impairment: PTSD participants were during the procedure, with gender as a second impairment and through elevated
experience of a and 1944 in countries functional confronted with the first sample taken 20 factor and loss of gender was cortisol levels in the
relatively protected occupied by the Nazi impairment was questions about min after the start. After a parents, depression significant for youngest male age
family life, to show regime, and assessed by means of their Holocaust resting period of 40min, a and physical health as reactivity at 20 min group, and in male
the most elevated immigrated to Israel the Post-traumatic survival experiences fourth sample was taken to covariates from the beginning respondents suffering
cortisol responses after 1945. or the stress diagnostic and exposure to assess the post-stress of the session, F(1, from PTSD-related
to a stressful purpose of analysis, scale (PDS). The 49- other shocking life cortisol level. 97) = 3.97, p = 0.049, functional impairment
challenge, with men the sample was item self-report scale events, e.g. sexual, see Figure 3. Males even when controlling
showing stronger divided into three age assesses DMS-IV physical or with functional PTSD for differences in loss
responses than groups: born between symptoms of PTSD. It emotional abuse, impairment showed of parents, depression
women. 1935 and 1937 (n = provides a categorical traumatic the strongest and physical health.
43), between 1938 diagnosis of PTSD, as experiences during reactivity
and 1940 (n = 43) and well as an overall the wars and the
between 1941 and measurement of terrorist attacks in
1944 (n = 47). symptom severity. Israel, combat
trauma, death of
close relatives after
the Holocaust, life-
threatening
illnesses and traffic
accidents.
75
Luecken & The current study Participants included CTQ: The Childhood Speech task: Salivary Cortisol: Five saliva Evaluation of A significant F(1, 79) .03317 Overall, parental loss
Appelhans evaluates cortisol 88 undergraduate Trauma Participants were samples were collected for differences between between-subjects = 4.7 was associated with
2006 responses to a students (ages 18–28 Questionnaire Q given 4 min to determination of cortisol the two groups in the interaction of family elevated cortisol in
stressful speech years, mean age = assesses five prepare and 4 min before, during, and after a pattern of cortisol group and reported response to a novel
task in young adults 19.6 years). The dimensions, including to deliver a speech moderately stressful speech responses to the abuse on cortisol stressor relative to
from bereaved and sample included 55 emotional abuse, to defend task. Samples were collected speech task were across all time participants from intact
maritally intact, females and 33 males, emotional neglect, themselves from a at pretask, immediately conducted with periods, F (1, 79) = families. As
control families. We representing a variety physical abuse, false accusation of following the task, and at repeated-measures 4.7, p = .02. hypothesized, the
expected that of ethnic physical neglect, and shoplifting. The 15, 30, and 45 min following general linear models, combination of early
parental loss in the backgrounds. sexual abuse. A total speech was the task. with five repeated parental loss and
context of higher abuse score was videotaped and was measures of cortisol reports of abusive
reports of abuse or calculated by given in front of an as dependent treatment was
conflict would be summing scores on “audience”. variables, and family associated with
associated with the five subscales. Participants were group, abuse, and the elevated cortisol
elevated cortisol The conflict subscale told that speeches group by abuse relative to those who
across the task and (nine items) from the would be evaluated interaction as experienced parental
greater cortisol Moos Family for clarity and the independent loss in the absence of
reactivity to the task Environment Scale strength of their variables. abusive treatment, and
relative to those provided a self report arguments, and relative to those from
from intact families. of the amount of would be rated in intact early family
conflict within the comparison to environments.
family. speeches given by
other participants.
Bremner et al The purpose of this Subjects included 41 Early Trauma Cognitive Challenge Salivary Cortisol: There was 2 (group: PTSD vs Salivary cortisol F(1,38) .01302 PTSD patients in this
2003 study was to assess men and women, 18 Inventory: PTSD Test: The cognitive a baseline cortisol measure healthy controls) x 10 levels were elevated = 6.79 study had increased
cortisol responsivity years of age or older patients were tests were based on before the cognitive (time) repeated in PTSD patients cortisol levels in
to a stressful who underwent a included with a a protocol challenge. Salivary cortisol measures ANOVA. relative to controls anticipation of and
cognitive challenge stressful cognitive history of childhood previously used in was measured at 0, 10, 15, during the hour during a stressful
in patients with challenge in physical or sexual studies of aging and 20, 30, 35, 45, 60, 75, and 90 preceding the cognitive challenge
PTSD related to conjunction with abuse, defined as included min after the initiation of cognitive stressor compared to healthy
childhood abuse. measurement of rape, attempted challenging the challenge. and during the subjects. Both PTSD
Based on findings in cortisol, heart rate rape, molestation, arithmetic course of the stress patients and controls
animal studies, we and blood pressure, physical assault or (multiplication, challenge (i.e. from - showed a similar 1.5-
hypothesized an and behavioral attack with injury, division, addition 60 to +20 min) (F = fold increase in cortisol
increased cortisol responses. Subjects before the age of 18, and subtraction), 6.79; df = 1,38; p < levels during the
and sympathetic included men and as measured by the cognitive tasks 0.0001). stressor relative to
response to women with civilian Early Trauma (stroop task, e.g. their own baseline.
stressful cognitive PTSD (N = 23) and Inventory (ETI), and looking at the word
challenge in PTSD healthy men and the diagnosis of PTSD red spelled in the
patients relative to women without based on the color green and
controls. trauma or PTSD (N = Structured Clinical naming the color
18). Interview for DSMIV green), problem
(SCID). solving, matching
figures to numbers
and memory for
figure-number
pairings, and
unscrambling
words.
76
Luecken 2000 A three-way Participants included Parental Loss: Prior Speech Task: For Salivary Cortisol: Saliva 2 (adversity: parental A significant three- F(2,54) .01598 Specifically, the current
interaction was 61 Duke University to age 16, the speech task, samples were collected from loss vs no loss) x 2 way interaction of = 4.47 results suggest that
hypothesized, in and North Carolina experiencing the participants were each subject for use in the (parental care: high Loss by Care by participants who lost a
which participants University students, death of one seated in front of a determination of baseline caring vs low caring) x Period was found parent and reported
who lost a parent age 18-27. biological parent, video camera. They cortisol levels and cortisol 3 (time) repeated (F(2,54) = 4.45, p = low caring from the
and reported low Participants were versus controls who were given reactivity to a stressful measures ANCOVA .01). surviving parent had
levels of parental divided between loss had not experienced instructions by a lab speech task; one with participant sex higher cortisol levels
caring from their (n=30) and no-loss parental loss. assistant reading a immediately prior to the and time of day as following a stressful
surviving parent (n=31) groups. Criteria script, and were task, one 5 minutes covariates. task relative to other
would show for participation were told to give a 3- following the completion of participants, who
increased cortisol either: (1) prior to age minute speech that the task, and one 20 showed stable cortisol
reactvity to a 16, the participant would be minutes after the levels during and after
stressful task experienced the death videotaped and completion of the task. the task.
relative to other of one biological evaluated.
participants. parent, or (2) the
participant was raised
by both biological
parents, and both
parents were still
living and had never
divorced.
Luecken 1998 This study proposes Subjects included 30 Parental Loss: Before Speech stressor Salivary Cortisol: Six saliva Statistical analyses A loss by period F(2,52) .01870 Cortisol elevation was
to test the university students age 16, the subject and Movie stressor: samples were collected from involved repeated- interaction was = 4.3 associated with
hypothesis that who lost one parent experienced the For the speech each subject for use in the measures analysis of found for cortisol parental loss during
early loss of a before age 16, and 31 death of one stressor, subjects determination of baseline covariance with BP, reactivity to the the speech task only,
parent, coupled control subjects. biological parent. were seated in cortisol levels and cortisol heart rate HR, and speech, (F(2,52) = and with family
with poor quality Criteria for front of a video reactivity to both tasks; one cortisol levels as the 4.3, p < .03), such relationships only
family relationships, participation were camera. They were immediately before each dependent variables, that subjects in the during the movie.
would result in long- either: 1) before age instructed to give a task, one 5 minutes after the parental loss group loss group showed
term increased 16, the subject 3-minute speech, completion of each task, and and FR as between- an average increase
cardiovascular and experienced the death which would be one 20 minutes after the subjects factors, and in cortisol levels
cortisol reactivity to of one biological recorded and completion of each task. period (baseline, task, (from baseline to
stress. parent, or 2) the evaluated. The or recovery) as recovery) as a result
subject was raised by video stressor, within-subjects of the speech,
both biological viewed in the same factors. 2(group) x whereas no-loss
parents, both parents room, was taken 6(time) ANOVA. subjects showed an
were still living and from the movie average decrease
had never divorced. "Terms of
Subjects could not be Endearment." The
experiencing 7-minute clip
significant health depicts two young
problems or taking boys experiencing
medications that the death of their
could affect mother.
cardiovascular or
cortisol measures.
77
Table 2.2: P-curve Disclosure Table of Trauma-Cortisol Hypo-reactivity Literature

Original Paper Quoted text from Sample Type of Trauma Stress Induction Cortisol Data Analysis Plan Quoted text from Results P- Conclusion quoted
original paper Characteristics Task Measurement original paper value from original text
interest
Peckins et al 2020 Given that past The final sample Violence Exposure: Socially Evaluated Salivary Cortisol: We performed GCM- Violence exposure t(218) = .00024 We found that violence
research found an resulted in 222 Violence exposure Cold-Pressor Task Adolescents provided a total LR’s to test whether was associated with a -3.73 exposure experienced
association adolescents (n= 117 was conceptualized (SECPT): The SECPT of 9 saliva samples via violence exposure less steep reactivity during childhood was
between hypo- girls) who identified as as: (1) violence is a laboratory passive drool throughout (with and without slope, b= -0.01, SE = associated with overall
reactivity of the Black or African directed at the child stress paradigm the SECPT protocol: controlling for social 0.001,p< .001, blunting of the HPA-
HPA-axis and American/non by a primary caregiver that contains both immediately prior to the deprivation) and decreased peak axis, above and beyond
interpersonal Hispanic (n= 167), (physical and physiological and onset of the SECPT (Sample social deprivation activation, b= -0.28, the effects of social
violence exposure, Caucasian/non- emotional abuse), (2) social-evaluative 1) and at 3 min, 15 min, 25 (with and without SE = 0.07,p< .001, deprivation and a
but not poverty, we Hispanic (n= 29), intimate partner components. min, 30 min, 35 min, 40 min, controlling for and less steep myriad of relevant
hypothesized Hispanic (n= 9), violence exposure, During the SECPT, 45 min, and 60 min post- violence exposure) recovery slope, b= covariates.
exposure to biracial (n= 14), and and (3) exposure to subjects immerse SECPT onset (Samples 2–9). were associated with 0.003, SE = 0.001,p<
violence would be other (n= 3). community violence. their hand in ice the cortisol response .01. Violence
associated with a water while being to the SECPT. exposure was
blunted cortisol watched by a associated with a
response over and research assistant blunted cortisol
above the effects of and videotaped. response to the
social deprivation. SECPT even when
controlling for the
effects of social
deprivation and all
covariates.
DePasquale et al The present study 280 children ages 7 Adoption from TSST-C: Salivary Cortisol: Saliva Group differences (PI Adoption status t(278) = .011 In sum, early-pubertal
2019 investigated through 14 years; 122 orphanages: Inclusion Participants samples were collected vs. NA) in family significantly 2.56 post institutionalized
whether (a) at children were criteria for the PI engaged in a seven times during the environment were predicted cortisol children show blunted
earlier stages of adopted from youth - international modified version of session ( 20, 0, +5, +20, +40, tested via a t-test. levels such that PI HPA axis reactivity
puberty, PI children institutions in 14 adoption after having the TSST-C, +60, and +80 min, where 0 children exhibited compared to
show a smaller countries between the spent 50% of involving a socially represents the beginning of lower cortisol levels nonadopted peers. At
cortisol stress ages of 6 months and preadoption life in evaluative public 5-min preparatory period than NA children (t = later pubertal stages,
response compared 5 years, after institutional care. speaking task and before the TSST-C. 2.56, p = .01). this pattern seems to
to nonadopted spending an average verbal math normalize, providing
children earlier in of 95% of their lives in performance. support for the
puberty while (b) at institutional care, and pubertal recalibration
later stages of 158 children of hypothesis following
puberty, there similarly high early
would be no socioeconomic status institutionalization.
difference between in their biological
PI and nonadopted families served as the
children’s cortisol nonadopted
responses, comparison group.
78
suggesting pubertal
recalibration of the
stress response.
Joos et al 2019 We also Data were collected Community Violence: TSST-M: a Salivary Cortisol: Cortisol Data analysis was Table 4: Chi-squared Chi2(2) .016 In accordance with the
hypothesized that for a study of early Community stress (α = standardized reactivity was measured in completed in two value for community = 8.26 toxic stress model,
adolescents who adolescents from .83) included 13 items protocol previously response to the TSST-M via steps. First, latent stress row. exposure to more
reported urban low-income reflecting witnessing demonstrated to saliva, collected by passive profiles of cortisol community violence
experiencing 59% female or experiencing elicit a cortisol drool into a 5-mililiter tube. reactivity were and less family support
greater number of households (N = 101; community violence response in Immediately prior to identified and were associated with
stressors would be 10–12 years old; 59% or crime (e.g., “You typically beginning the TSST-M, described using blunted cortisol
more likely to female). saw someone developing early participants provided the latent profile analysis reactivity, and Reactive
present with threatened with a adolescents. Youth first saliva samples (S1). The (LPA). After the profile membership
blunted reactivity knife or gun”). were instructed by second saliva sample (S2) optimal profile was associated with
Adolescents indicated a two-person panel was taken immediately solution was selected fewer trauma
whether each event of “experts” to following completion of the based on theoretical symptoms.
happened or not in prepare and TSST-M, and four additional rationale and fit
the last six months. deliver a 5-minute samples were taken at 10- indices, we examined
speech introducing minute intervals (S3-S6) whether profile
themselves to a thereafter. membership was
new class, then associated with
completed oral proposed variables of
serial subtraction interest, expressing
for 5 min. these as pairwise
differences.
79
Lovallo et al 2019 The contribution of The present sample ELA: ELA was assessed Stress Protocol: Salivary Cortisol: Saliva Dependent variables Cortisol responses F(1,689) .03289 We previously
ELA to blunted includes 709 persons as follows: Physical or The protocol (75 samples were collected were cortisol and are shown in the top = 4.57 reported that exposure
reactivity and risk (398 females) Sexual Adversity min) included a using the Salivette device heart rate responses panels of Figure 1. to ELA during
for alcoholism gains recruited through (“Have you ever prestress baseline and were taken at 9 periods to stress along with FH+ subjects had childhood and
significance in light community been mugged or (30 min), during across both days: self-reports. smaller adolescence
of disrupted family advertisement. threatened with a which the subject awakening, arrival at the lab, Cortisol data were responses than FH−, contributes in a dose-
relationships often weapon or ever sat quietly and min 10 and 20 of the log transformed to F(1, 707) = 4.15, response manner to
occurring in experienced a break- read general baseline period, min 15, 30, normalize the p<.043, and ELA led blunting of stress axis
alcoholic in or robbery?” “Have interest magazines, and 45 of the stress protocol distribution. Cortisol to progressively reactivity in early
households. We you ever been raped followed by mental or continued rest, and at 30 reactivity was diminished cortisol adulthood. The present
examined cortisol or sexually assaulted stress (45 min) min post stress or rest, and measured as the responses to stress, findings are based on
and heart rate by a relative?” consisting of at bedtime. average change from F(1, 689) = 4.57, an expanded study
responses to mental yielding a possible 3 simulated public rest day to stress day p<.011, in agreement sample and show that
stress in healthy points) and speaking (30 min) in saliva samples with our previous FH+ persons are
FH+ and FH− young Separation from and mental taken at report on a smaller vulnerable to ELA while
adults and Parents (“Before you arithmetic (15 minutes 30 and 45 of sample their FH− counterparts
predicted that were 15, was there a min). the stress period and appear to be resilient.
exposure to ELA time when you did not at the same two
would have a live with your times during the
greater impact on biological mother for extended rest
stress reactivity in at least 6 months?” protocol. The
FH+ persons. “Before you were 15, analysis used a
was there a time general linear model
when you did not live including ELA (scaled
with your biological as 0, 1, or >1 adverse
father for at least 6 experiences), FH
months?”, for a (FH+, FH−), and the
possible 2 points). ELA x FH interaction.
Timothy et al 2019 We wanted to The Children of Childhood adversity, TSST-C: The Trier Salivary Cortisol: Cortisol Data was tested for Table 4: Pearson r(49) = - 02 In our study, the
evaluate the effect alcoholics (COA) collected from the Social Stress Test estimation in the saliva normality. correlation between 0.32 cortisol curve after the
of early adversity on group, comprised 50 subjects using a for children (TSST- samples was done by ELISA Differences between ACE score in the COA Trier stress test was
the HPA axis male children and modified version of C) (public speaking according to the protocol the demographic group and AUCG. found to be flatter in
function (Cortisol adolescents, (age 8– WHO Adverse an arithmetic) and provided (kits from IBL Int variables, adversity the COA compared to
reactivity to acute 16 years), ascertained Childhood modified for Indian GMBH). The cortisol indices and cortisol indices controls. The relative
stress), and its from offspring of Experiences Scale children was generated were cortisol between the COA blunting of cortisol
association with treatment seeking (International administered to concentrations at each time and control groups response to stress
DNA methylation in alcohol-dependent Questionnaire) (ACE each child, and point (t_B at baseline and were tested using (AUCG and AUCI) was
the SLC6A4 gene (5- patients (from (IQ)). The domains of saliva samples t0, t10, t20, t30, t45 t60 at the Student's t-test. associated with higher
HTTLPR). It was multiplex alcoholism the adverse collected (once subsequent time points). Repeated measures externalizing symptom
hypothesized that families) at the Centre experiences were before the test, ANOVA was used to scores, especially in the
early adversity for Addiction physical, emotional and at 0, 10, 20, test the change in COA, who had
would influence Medicine, National and sexual abuse 30, 45 and 60 min cortisol experienced higher
both DNA Institute of Mental along with neglect, after the test) and concentration with ELS.
methylation and Health and peer violence and stored at −80 °C time.
HPA axis reactivity Neurosciences collective violence. until analysis.
and possibly affect (NIMHANS), The severity of abuse
behaviour. Bangalore, India. Fifty was graded on a Likert
healthy children and Scale and all the
adolescents, matched scores were summed
80
for age, socio- up to give an ACE

economic status and score.
ethnicity, were
recruited from the
community to form
the control group.
Hengesch et al 2018 The aim of the The ELA group -- 22 Institutional Rearing: Two previously Salivary Cortisol: Saliva Group differences Peak salivary cortisol F(1,43) .033 Although ELA
current study was adults (14 females; Adopted by validated stressful samples were taken at were explored by responses were = 4.85 participants showed
to compare age ranging from 19 Luxembourgish elements (bilateral minutes −30, −20, −5, +3, two factorial ANOVA different between higher stress-induced
neuroendocrine, to 30 yrs.) who were families after being feet CPT and the +15, +25, +40, and +55. examining the ELA groups (p = .03), changes in arousal and
cardiovascular/auto adopted by institutionalized in the Paced Auditory influence of two with lower cortisol anxiety ratings, as well
nomous, and Luxembourgish 1990s in international Serial Addition independent reactivity in ELA as similar
subjective stress families after being orphanages during Task (PASAT)) were variables (ELA and participants cardiovascular stress
reactivity of ELA vs. institutionalized in the their early childhood. combined in an GENDER) on one compared to reactivity, their HPA
control participants, 1990s in international extended Cold dependent variable controls, but no main stress axis
and in addition, to orphanages during Pressor Test (CPT). (MAP; HR; cortisol effect of GENDER, responsiveness was
address potential their early childhood. responses; subjective and no significant blunted as compared
gender differences. The mean age at ratings; performance interaction effect of to controls.
adoption was 6.4 measures). ELA x GENDER (see
months. Twenty-two Fig. 1 and Table 2).
healthy participants
(11 females; age
ranging from 19 to 28
yrs.) without
institutionalization/
adoption history
served as control
group.
O’Connor et al 2018 Therefore, using One hundred and Childhood Trauma Maastricht acute Salivary Cortisol: To Hierarchical linear At step 2, again as F(1,92) .02 Second, higher levels
data from the sixty participants (100 Questionnaire: A brief stress test (MAST): measure cortisol reactivity regression was predicted, childhood = 5.32 of childhood trauma
recent O’Connor et females) were 28-item self-report The MAST is a to stress, saliva samples utilised, following the trauma (β = −0.23, p were found to be
al. (2017) study, the recruited to a inventory was used to recently developed were taken using Salivettes procedures outlined = 0.02) significantly associated with lower
primary aim of the previous attempt (n = assess for a history of stress protocol (Sarstedt, Germany) during by Kenny et al. entered the resting cortisol and
current paper was 49), a suicidal ideation abuse or neglect in designed to be the resting cortisol phase (1998), to test equation, F (1, 92) = blunted cortisol
to investigate but no attempt (n = childhood or both and then at the beginning of whether childhood 5.32, p = 0.02, and reactivity to stress in
whether childhood 55) and a control adolescence. The CTQ physiologically and the stress task (T00), at +10, trauma was explained an adulthood (while
trauma was group (n = 48) based has five subscales psychologically +20, +30 and +40 min post- associated with additional 5% in the controlling for age,
associated with low upon established relating to types of challenging by task. lower levels of variability in AUCg gender, BMI,
levels of cortisol measures of suicidal maltreatment: combining an cortisol reactivity to levels, indicating medication usage, time
reactivity to a behavior (see below). emotional, physical uncontrollable a laboratory stressor higher levels of of day and smoking
laboratory stressor Participants were and sexual abuse and physical stressor (AUCg and AUCi) and trauma were status).
and lower resting aged between 18 and emotional and (i.e., a cold pressor lower resting cortisol associated with lower
cortisol levels (not 62 years (M = 26.84 physical neglect, with challenge) with a levels in suicide cortisol levels.
previously reported) years, SD = 9.32) with five items for each social-evaluative attempters and
in individuals 73.8% identified as subscale (1 = never (i.e., mental ideators.
vulnerable to Caucasian. true, 5 = very often arithmetic)
suicide. true). In the current component. In
study, we were addition, it has
interested in exposure been shown to
81
to any type of yield similar

childhood trauma, subjective and
therefore, we created cortisol stress
a total childhood responses to the
trauma score. Trier Social Stress
Test, however, it
does not require
the presence of a
panel.
Monteleone et al On the basis of our Twenty-four AN Childhood Trauma TSST: The TSST was Salivary Cortisol: Saliva A two-way analyses Two-way ANOVA F(10,19 .002 The first finding of our
2018 previous findings, women and 17 Questionnaire: The performed in the sample were collected in of variance (ANOVA) with repeated 0) = study was a
we hypothesize that healthy women were CTQ asks respondents afternoon, Salivette tubes (Sarstedt; and covariance with measure yielded a 2.82 significantly decreased
AN patients with enrolled in the study. to report on between 2.30 p.m. Rommelsdorft, Germany) at repeated measures significant main saliva cortisol response
early traumatic Patients were childhood experience and 4.30 p.m., the end of the resting period were employed to effect for time (F5,16 to TSST in AN women
experiences display classified as mal- across five types of when the HPA axis (T=-20); immediately before test differences in = 4.08, p = 0.001) and reporting childhood
a reduced cortisol treated (Mal) or non- childhood activity is relatively starting the TSST (T = 0), at saliva cortisol levels a significant group × maltreatment
response to TSST maltreated (noMal) maltreatment: stable and at a low the end of the TSST (T=10) and anxiety scores time interaction compared to both AN
together with an according to their emotional neglect level, so changes and 10 (T=20), 30 (T=40) and among the groups. (F10,190 = 2.82, p = women and healthy
altered negative Childhood Trauma (EN), emotional abuse induced by 50min (T=60) after the end Between- group 0.002), indicating women without history
emotional response. Questionnaire scores. (EA), sexual abuse experimental of the test. Saliva was differences at each that saliva cortisol of childhood trauma
(SA), physical neglect challenges are separated by centrifugation time point were levels changed exposure. Indeed, Mal
(PN) and physical clearly evident. and stored at −20 °C until evaluated by the significantly after the AN women exhibited a
abuse (PA). Scoring of assayed for cortisol levels. post-hoc Tukey's TSST and that flattened cortisol
the CTQ produces 5 test. significant response to TSST
subscale scores for differences among whereas noMal AN
each subtype of the groups emerged patients had a TSST-
maltreatment and a in the timing of induced clear-cut in-
composite CTQ total cortisol changes. crease in saliva cortisol
score. levels that did not
differ statistically from
healthy controls.
Trautmann et al We hypothesized 96 subjects agreed to Trauma History The Trauma Film Salivary Cortisol: As a To test for (Fig. 2) with a F(2,95) .00048 Although ELA
2018 that (a) acute participate and were Questionnaire: Other Paradigm: We marker of endocrine stress differences between significant three-way = 8.3 participants showed
trauma exposure randomized to either than the broader used the trauma reactivity, we assessed participants of the interaction sex × higher stress-induced
would lead to an the trauma or a construct of childhood film paradigm (TFP) salivary cortisol levels. Saliva trauma and the childhood traumas × changes in arousal and
increase in reported neutral condition (also maltreatment, which as a laboratory samples were collected 1 neutral film film condition (b = anxiety ratings, as well
alcohol craving see description of the has been the trauma-analogue min before the film as well condition in 8.3 [4.4– 12.1], p < as similar
compared to a trauma film paradigm predominant stressor. In the as 1, 10, 20, and 30 min demographics and 0.001). In males, but cardiovascular stress
neutral condition, subsequently) with an construct in previous TFP, non-clinical after the film using baseline variables, F not in females (p’s > reactivity, their HPA
(b) trauma-related equal group size (n = studies on effects of participants watch Salivettes devices. tests from robust .459), the number of stress axis
increase in craving 48). One participant in early adversities, the films containing regression models childhood traumas responsiveness was
would be positively the trauma condition THQ measures the scenes showing (see following text) was positively blunted as compared
related to dropped out during history of exposure to traumatic events for dimensional and associated with to controls.
experienced the study resulting in actual traumatic such as accidents, chi-square tests for increase in HRR (b =
childhood traumas, a group size of n = 47 events according to physical assault, or categorical out- 3.0 [1.2–4.7], p =
and (c) the for this condition. the DSM-IV A1 rape. comes are reported. 0.001) and SCR (b =
association criterion of 0.3 [0.0– 0.7], p =
between childhood posttraumatic stress 0.050), whereas
traumas and disorder (American CortR in the trauma
trauma- induced Psychiatric condition decreased
82
craving could be Association 2000). The with increasing

explained by THQ consists of numbers of
increased peri- questions on 24 childhood traumas (b
traumatic stress traumatic events = − 62.4 [−107.9 to −
reactivity, indicated including information 16.9], p = 0.008) (Fig.
by (c1) an about the age of 3a–c).
association trauma exposure. All
between childhood reported traumatic
traumas and stress events experienced
reactivity and (c2) until the age of 18
an association were summarized.
between stress
reactivity and
craving.
Cordero et al 2017 Based on the For the current study, PTSD: Participant- Modified Crowell Salivary Cortisol: The Effect of IPV-PTSD on ANOVA repeated F(3,129) .015 The current findings
literature reviewed 45 mothers (HC, n=18; mothers were Procedure: A 30- “laboratory stressor” cortisol levels during measures of = 3.63 showed that mothers
above, we expected IPV-PTSD, n=27) and interviewed using the min mother-child consisted of two 3-min the laboratory mothers’ cortisol with IPV-PTSD as
that IPV-PTSD their young children Clinician Administered interaction separations from the session and diurnal levels during the compared to HC had
mothers and their (12-42 months of age) PTSD Scale to assess procedure known mother with 8 min interval release were laboratory stressor lower cortisol levels in
toddlers would have were included in the PTSD with as the “Modified between them, and analyzed using showed a significant response to the
a low diurnal analyses. confirmation of the Crowell exposure to two novel ANOVAs with effect for time laboratory stressor,
cortisol release severity of the current Procedure", which stimuli. Four salivary repeated (F3,129=29.03, which included child-
pattern, and PTSD symptoms via consisted of free samples for the assessment measurements. In p<0.001, ηp2=0.41) parent separation; this
blunted cortisol the Posttraumatic and structured of cortisol were collected case of a significant and group x time being the case even
reactivity to stress. Symptom Checklist - play, separation- from the mother and the group by condition interaction though distress
short version. reunion, clean-up child: baseline (before the interaction effect, (F3,129=3.63, behavior of the
and exposure to procedure begin), post-hoc t tests were p<0.015, ηp2=0.08) children and mothers'
novelty. immediately after the end of performed. rating of toddler’s
the stress situation (T0), and anxiety was similar
at 30 (T30) and 60 (T60) min between groups.
England-Mason et The purpose of the Participants included CTQ: The CTQ is a 28- The Emotional Salivary Cortisol: All visits The main analytical Of all of the entered t(134) = .035 The significant
al 2017 current study was a community sample item self-report Stroop: At the 7- commenced between 1000 approach was to predictor variables, -2.13 interaction effect
to investigate of 140 new mothers questionnaire that month visit, h and 1430 h (M = 1138 h, determine whether only maternal history revealed that only
cortisol reactivity who were recruited evaluates types and participants SD = 0135.7 h), and and how latent of maltreatment (β = women who reported
and difficulties with from the maternity frequency of abuse completed a 17- immediately following difficulties with −0.18, t = −2.13, p = higher levels of child
emotion regulation ward at St. Joseph's and neglect using a 5- min computerized consent procedures the first emotion regulation 0.03), bedtime (β = maltreatment and
in postpartum Healthcare Hamilton, point Likert scale (1 = Emotional Stroop salivary sample was would interact with 0.16, t = 1.98, p = greater difficulties with
women with a Ontario. never; 5 = very often). task involving collected. The start time of maternal exposure to 0.048), and visible emotion regulation
history of childhood In the present sample emotional and salivary sample collection childhood minority status (β = exhibited decreased
maltreatment. using moderate- neutral stimuli. The was included in analyses as a maltreatment to 0.21, t = 2.71,1 p = cortisol reactivity to
severe cutoffs, 15.7% Emotional Stroop covariate. Immediately predict cortisol 0.01) significantly psychosocial challenge.
of women reported paradigm is a following the baseline reactivity. contributed to the
experiencing clinically-relevant sample collection, variance in cortisol
emotional abuse, adaptation of the participants completed the reactivity.
12.9% reported sexual original Stroop computerized Emotional
abuse, 7.9% reported procedure, where Stroop task. Then two more
physical abuse, 4.3% the emotional samples were collected at
reported physical significance of 20- and 40- min following
neglect, and 0.7% word stimuli is completion of the task.
83
reported emotional varied so as to

neglect. Overall, study the
22.1% of women interference of
reported experiencing emotionally salient
at least one form of words on colour
moderate-severe naming. The
childhood trauma, Emotional Stroop
and 11.4% of women has been
reported experiencing previously used to
two or more forms of investigate the
moderate-severe relationship
childhood trauma. between selective
attention to cues
of social and
emotional threat
and engagement of
the HPA axis.
Ewing-Cobbs et al The primary aim of Participants were 8-15 Injury: The external TSST-C: All Salivary Cortisol: Saliva Variable The healthy Chi2(1)= .015 Compared to a healthy
2017 the study was to years of age and cause of injury in a participants samples were collected distributions, comparison group 5.86 non-injured group,
characterize salivary hospitalized for vehicle accident was completed The using 1 × 4 CM polyolefin comparability of showed significantly injured children (8 to
cortisol and sAA traumatic brain injury selected to meet the Trier Social Stress swabs placed under the demographic and greater reactivity 12 years), but not
stress-related (TBI; n=55; M age Diagnostic and Test for Children participants tongue for 2 psychosocial than the injured adolescents (13 to 15
reactivity and =13.9 yrs; 40 males) Statistical Manual of (TSST-C), an mins at 20 minutes before variables for the TBI, groups, Χ2 (1)=5.86, years), had elevated
regulation in or extracranial injury Mental Disorders established and right before the TSST-C. EI, and healthy p=.015; no cortisol, and injured
children and (EI; n=29; M age 12.3 (American Psychiatric laboratory Post-stressor samples were comparison groups differences in post- participants
adolescents yrs, 20 males) Association, 2013) procedure collected immediately after were examined using stressor regulation of (regardless of age)
sustaining traumatic following vehicular criterion A for a PTSD involving public (sAA only), as well as 20 and ANOVA or chi square. cortisol were showed dampened
injury in comparison accidents, and a diagnosis that speaking (social 40 (cortisol only) minutes GLM examined the obtained cortisol reactivity to
to a healthy non- healthy non-injured specifies exposure to evaluative stress) after to capture the distinct effects of group, social evaluative
injury control group. comparison group a potentially life- and mental profiles of the analytes. saliva collection time, threat.
(n=33; M age =12.5 threatening situation. subtraction age, and their
yrs, 16 males). (cognitive stress). interactions on
salivary analytes
Lovallo et al 2017 In the Oklahoma Participants were 252 Early Life Adversity: Stress Procedure: Salivary Cortisol: Saliva Cortisol responses to In contrast, persons F(4, .028 Exposure to ELA may
Family Health healthy young adults ELA was derived from Stress testing samples for cortisol stress were then with greater ELA 244) = lead to diminished
Patterns cohort, participating in the C-DIS-IV items as lasted 105 min, determination were calculated as the exposure had 2.78 stress responses in
exposure to ELA Oklahoma Family follows: Physical or including a resting collected using Salivettes at average of the three progressively smaller otherwise healthy
leads to Health Patterns Sexual Adversity baseline in a 9 times across each day samples taken during cortisol responses, F young adults, although
progressively (OFHP) project who (“Have you ever been seated position (30 including: 2 at home (upon the stress period on (df = 2,244) = 4.56, p carriers of specific
diminished cortisol had been genotyped mugged or threatened min) followed by awakening and at bedtime) the stress day (STR1, = .011, partial eta2 = genotypes may
and heart rate for the COMT with a weapon or simulated public and 7 in the lab (3 prestress, STR2, STR3) minus .037 and we respond differently to
reactivity to Val158Met experienced a break- speaking (30 min) 3 during the stress protocol, the same three observed a significant ELA history.
psychological stress polymorphism along in or robbery?” “Have and mental and 1 during recovery). samples taken on the G × ELA interaction, F
in adulthood. In the with ancestry you ever been raped arithmetic (15 min) Cortisol responses to the rest day. Cortisol (df = 4, 244) = 2.78, p
present analysis, informative markers or sexually assaulted and a 30-min stressor were computed as data were analyzed = .028, partial eta2 =
we examined (AIMS) and had by a relative?” and resting recovery the difference between the by a general linear .045, suggesting that
whether the impact sufficient background Emotional Adversity period as described mean of the 3 values model including: effect of ELA on adult
of ELA on cortisol data to compute ELA (“Before you were 15, elsewhere. The obtained during the stress Genotype (Met/Met, cortisol responses
and heart rate scores was there a time resting control day protocol and the Val/Met, Val/Val), depends in part on
reactivity would when you did not live involved sitting for ELA (0, 1, ≥ 2), and
84
differ as a function with your biological 105 min during the corresponding values on the the G × ELA the person’s
of COMT mother for at least 6 same time of day resting control day interaction term. genotype.
Val158Met months?” Each person while reading
genotypes. was assigned to an general interest
ELA group based on 0, magazines and
1, or ≥ 2 reported watching nature
adverse events. videos.
Troller-Renfree et We predicted that The sample included Institutional Care: Peer Evaluation Salivary Cortisol: First, following an Table 1: Cortisol t(95) = .004 At age 12, among
al 2017 at age 12 positive 136 children, Children raised in Stress Paradigm Participants provided a “intent to treat levels following peer -2.94 children who received
biases would abandoned in infancy institutions that (details available in saliva sample during a approach” analyses task (difference the foster care
predict less blunted and placed into experienced original article) . baseline resting period prior were conducted with between CAUG and intervention, positive
(similar to never institutions in psychosocial to beginning the peer the two randomized NIG) biases were related to
institutionalized Bucharest, Romania, deprivation evaluation task and a second groups (CAUG, FCG) less blunted (or more
children's reactivity) and who were part of sample approximately 15 examining the size reactive) cortisol
physiological the Bucharest Early minutes after the task was and direction of reactivity during a
reactivity to stress Intervention Project. finished attention biases at social stressor, which
in children who Children were 12 years of age. may indicate stress
received the foster randomized to either Attention biases at responses more similar
care intervention receive care as usual 12 were related to to the NIG.
(CAUG) or a high social outcomes.
quality foster care Third, attention
intervention (FCG). A biases were related
sample of 48 (26 to physiological
female) age-matched reactivity.
community-reared Regressions were
children who had conducted for the
never been entire previously
institutionalized (NIG), institutionalized
were recruited for sample and
comparison separately for each
group.
Kliewer et al 2016 It was expected that Youth and their Victimization: Survey The Social Salivary Cortisol: Saliva Descriptive As hypothesized, r(235) = .02088 Structural equation
as youth’s maternal caregivers of Children’s Exposure Competence samples were collected from information on and victimization was -.15 modeling revealed that
victimization (N= 358 families) were to Violence. Youth Interview (SCI): the adolescents directly correlations among negatively associated youth’s victimization
experiences recruited for a 4-wave indicated on a scale During the hot before the start of the SCI (2 the study variables is with total cortisol experiences were
increased their longitudinal study of from 0 (never) to 4 phase, the samples), at the end of the presented first. Next, output (b= −.15, p< negatively associated
cortisol responses the consequences of (almost every day) interviewer asks “hot” phase, and 10, 10, and structural equation .05), after accounting with total cortisol
would decrease, exposure to violence. how often they had the adolescent to 20 min later, yielding 6 modeling was for time of day of output in response to a
and their salivary The analytic sample experienced 10 types re-experience a samples. utilized to test the data collection, stress task. However,
alpha amylase for the present study of victimization stressful event and primary study adolescent sex, and no such association
responses would included 242 urban ranging from being asks questions hypotheses using pubertal status at was observed between
increase. adolescents who had slapped, hit, or about the Mplus version 7.31, baseline, each of victimization and
biological data at punched by someone participant’s which allowed which were salivary alpha amylase.
Wave 4. Youth were to being shot. Items thoughts/feelings missing data to be significant, and Our findings contribute
either in the 5th or began with the stem, during the event. handled with full adolescent age, to a growing literature
8th grade at baseline “How many times in The cool phase information which was not supporting the
(M= 11.98 years, SD= the past year have asks the maximum likelihood significant. attenuation
1.56, range = 9–16 you yourself been …” participant to (FIML). hypothesis, which is
years) All items used in the describe how the particularly applicable
85
study are listed on situation would for youth experiencing

Table 2. have ideally ended. chronic stress.
Koss et al 2016 We sought to Participants included International Each laboratory Salivary Cortisol: Three Analyses were Figure 3: Association r(167) = <.0000 Blunted cortisol
examine differences 167 children taking Adoption: Children session was saliva samples were conducted in four between early -.56 1 reactivity to laboratory
in children’s cortisol part in a larger adopted approximately 2 collected throughout the parts. First, adversity and stressors was observed
reactivity to a longitudinal study of internationally from hours and laboratory session at each of descriptive statistics hypocortisolism among post-
laboratory setting the transition to institutions or consisted of a the four assessments (12 were examined and institutionalized and
among children family care following orphanages number of total laboratory samples). ANOVAs were post-foster care
with varying international challenging tasks, The samples were not conducted to children compared to
degrees of early life adoption. Sixty-five including brief collected in response to any examine group non-adopted peers.
experiences. This children (38 female, separations, one task but rather reflect differences in
included 27 male) were exposure to novel, children’s reactivity to the children’s
investigations of the adopted from arousing stimuli, demands of the laboratory kindergarten
impact of early life orphanages or interactions with session as a whole. behavior problems.
adversity on cortisol institutions (post- strangers, Second, latent
reactivity as well as institutionalized; PI), transitions growth models were
examinations of 49 children (19 between tasks and fit examining cortisol
longitudinal change, female, 30 male) were rooms, and reactivity in the
indicative of adopted from electrophysiologica laboratory using
recovery of the HPA international foster l assessments (see multilevel structural
axis, in children’s care (post-foster care; supplemental equation modeling
cortisol reactivity PFC) and 53 non- materials for a (MSEM) in MPLUS.
following adoption. adopted (NA) same- detailed Next, regression
We examined the aged children (27 description of tasks analyses were
extent to which female, 26 male) who and timing of conducted to
indices of diurnal were born and raised cortisol sampling. examine whether
cortisol and stress in similar types of home diurnal
reactivity together families hypoactivity
reflect (education/income) predicted laboratory
hypocortisolism in that adopt hypoactivity.
the context of internationally-born
chronic stress. children.
Martinson et al HPA axis (i.e., Participants were 50 Sexual Trauma and Participants came Salivary Cortisol: Salivary A 2 (group: sexual Results of this F(2.26,1 .04 As hypothesized, the
2016 cortisol) reactivity women, 18 years of PTSD: a list of 16 into the lab and cortisol samples were trauma, controls) × 5 analysis revealed a 08.3) = primary finding of this
to an emotional age or older (MAge = behaviorally-specific participated in a 45 collected using the (time: baseline, 15 significant effect for 3.09 study was that women
closeness induction 19.70, SD = 4.19), who sexual trauma min emotional Salivette® Cortisol with a min into task, 30 Time, F(2.26,108.3) = with a history of sexual
task is expected to were enrolled in questions was used as closeness exercise synthetic swab from min, 45 min into 2.95, p < .001. There trauma exhibited a
be suppressed (i.e., undergraduate a pre- screening with a male Sarstedt, Inc. (Newton, NC). task, after 15 min was also a main blunted cortisol
blunted) in women psychology courses at measure to evaluate confederate and post-task relaxation effect for Group, response and greater
with a history of a large public potential eligibility for completed self- period) mixed F(1,48) = 5.31, p = anxious mood in
sexual trauma university in Maine. this study. PTSD report factorial design. The .03. These results reaction to the
compared to The majority of checklist—civilian questionnaires of primary dependent were qualified by a intimacy induction task
women without women were single version (PCL-C) is a closeness, state variables were significant time × compared to women in
such histories. PTSD (43.1%) and Caucasian 17-item self-report anxiety/depression cortisol levels and group interaction, the control group.
symptom severity is (88.2%). measure for civilians , and cortisol psychological F(2.26,108.3) = 3.09,
expected to be reflecting the 17 DSM- assays at the reactivity. Post-hoc p = .04
inversely correlated IV symptoms of PTSD. aforementioned analyses for ANOVA
with cortisol A total severity score time points. analyses were
reactivity. was obtained by conducted with
86
summing the scores follow up tests of

from each of the 17 simple main effects.
items
Hostinar et al 2015 We aimed to test We recruited 81 Institutional Rearing: TSST-C: This Salivary Cortisol: Hierarchical Linear Results (Table 3) F(1, .025 Cortisol reactivity for PI
whether parent children aged Children raised in paradigm consisted Participants expelled saliva Modeling was used indicated a significant 194) = children in both
support would be approximately 9–10 institutions that of a public through a straw into pre- to analyze the interaction of Group 5.08 conditions was lower
more effective than years (M = 9.85, SD = experienced speaking task labeled vials. Time was cortisol data because x Condition on than that of non-
stranger support in .55, range: 8.87 – psychosocial (introducing coded as 0, 1, 2, and 3 given it is ideal for auto- cortisol reactivity adopted children in the
lowering cortisol 11.09), half of whom deprivation. oneself to a that the samples were correlated samples (linear term: F(1, stranger support
stress responses for were internationally hypothetical new equally spaced at 20-minute collected from the 194) = 5.08, p = .025; condition.
NA children, but not adopted PI children (N classroom of intervals. same individual and quadratic term: F(1,
PI children, as a = 41, M age = 9.7 students) and a it allows for greater 147) = 3.76, p = .055),
reflection of the years, SD = .56, range mental arithmetic statistical power with no main effects
enduring effects of 8.87 – 10.99; 51.2% task (subtracting than traditional RM of Group or
severe early social females) and half who out loud by 3s from ANOVA models Condition on
deprivation. were non-adopted – 307). reactivity (p’s > .09).
i.e., born and raised
by their birth families
in a large urban
Midwestern area (N =
40, M age = 9.97
years, SD = .52, range
9.09 –11.09; 50%
females).
Jaffee et al 2015 We hypothesized The sample included Recent Traumatic Social Provocation Salivary Cortisol: Saliva was An ordinary least The main effect of t(54) = .007 The lowest levels of
that with consistent 400 children (51% Events: Recent Task (SP): The collected at 20 and 10 min squares regression recent traumatic -2.8 cortisol reactivity were
exposure to harsh, male) who traumatic events (at social provocation prior to the start of the SP analysis was events on cortisol observed among
rejecting, or participated in the follow-up) were (SP) task was task, at the end of the SP conducted in which reactivity was children who had
dangerous Children's Experiences measured with the designed to elicit task, and at approximately cortisol reactivity qualified by a experienced higher
relationships and and Development Traumatic Events feelings of 20, 45, and 65 min following values were significant interaction levels of harsh,
environments Study (CEDS). Screening Inventory. frustration and the conclusion of the SP regressed on harsh, involving harsh, nonresponsive
across early and Participants ranged Because agreement anger in children task. Nineteen percent of nonresponsive nonresponsive parenting in early
middle childhood, from 8 to 11 years (M between caregiver and was modified children refused to parenting, traumatic parenting (Table 1 childhood and two or
the cortisol = 9.99, SD = 0.74). We and child reports was from a task participate in round 2 of the events, and and Figure 1). Results more recent traumatic
response to sampled a ‘higher risk’ modest (kappas developed by van SP task (often because they covariates at the first were unchanged, events.
experimentally group that included ranged from 0 for low Goozen et al. were very distressed). step and on the controlling for
induced stressors children who had base rate events like (1998). Because cortisol interaction between participation in round
would be relatively experienced relatively ‘child was kidnapped’ concentrations at time t harsh, nonresponsive 2 of the SP task
low. In contrast, we high levels of harsh, to .51 for ‘family reflect cortisol activity t – 15 parenting and (interaction: b = .26,
hypothesized that in nonresponsive member was in to t – 30 min earlier, cortisol traumatic events at SE = .10, p < .05) or
the absence of an parenting at age 3 trouble with the reactivity to the SP task was the second step. All baseline cortisol
early childhood years. A ‘lower risk’ police or in prison’), assessed by subtracting the variables were mean- values (interaction:
history of harsh, group included events were coded as lowest cortisol value centered prior to b = .21, SE = .09, p <
nonresponsive children who had having happened only collected by the end of the analysis. The main .05).
parenting, recent experienced relatively if both the child and SP task (which reflected effect of recent
exposure to low levels of harsh, caregiver reported the cortisol activity prior to or in traumatic events on
traumatic events nonresponsive event (M = 0.61, SD = the very early stages of the cortisol reactivity
would be associated 0.85). SP task) from the cortisol was qualified by a
87
with increased parenting at age 3 value collected 20 min significant

cortisol reactivity. years. following the SP task (which interaction involving
reflected cortisol activity at harsh, nonresponsive
the conclusion of the SP parenting
task).
Kempke et al 2015 We expected, in line Participants were 43 Childhood Trauma TSST: he TSST is a Salivary Cortisol: Salivary Hierarchical Results of the TSST t(38) = .02090 Results revealed that
with extant theories consecutively (CTQ): The Childhood widely used cortisol samples ere regression analyses showed that - 2.41 emotional neglect was
and findings admitted female Trauma Questionnaire standardized collected at arrival, after a were performed with emotional neglect clearly negatively
concerning the patients diagnosed — Short Form (CTQ- psychosocial stress 30-min rest period/2 min stress indices (AUC) (EN) was inversely related to cortisol
long-term effects of with CFS according to SF) was used to test consisting of a before the onset of the TSST as dependent associated with reactivity in the TSST.
early childhood the Center for Disease measure exposure to role-play (5 min) (baseline), immediately after variables and the measures of increase Taken together, our
trauma on the HPA Control and early childhood and mental the TSST (0), and at +10, different trauma in cortisol. findings seem to
axis, that higher Prevention criteria for maltreatment. The arithmetic (5 min) +20, +30, +45 and +90 min domains as Specifically, suggest that child-
levels of self- CFS. Among the 43 CTQ-SF measures five task in front of an relative to the end of the independent emotional neglect hood emotional
reported early enrolled patients, two dimensions of evaluative TSST. The stress test was variables to evaluate (EN) was strongly neglect is associated
childhood trauma patients did not childhood trauma: audience performed between 2:45pm the association significantly with a reduced ability
would be negatively complete the study emotional abuse (EA), instructed to and 3:15pm in a lab between cortisol negatively related to to activate the
associated with for personal reasons, sexual abuse (SA), maintain a neutral specifically set up for the reactivity and early AUCI (β= −.535, p < neuroendocrine stress
both cortisol activity and one patient physical abuse (PA), facial expression. TSST, with a separate childhood trauma. .05) and delta cortisol system when con-
and cortisol refused to be exposed emotional neglect waiting and testing room. (β = −.508, p < .05). fronted with acute
responses to to experimental (EN), and physical psychosocial stress, but
experimentally- stress. Thus, the final neglect (PN). The CTQ not with a fundamental
induced study sample scores range from 5 to impairment in daily
psychosocial stress. consisted of 40 25. cortisol activity.
patients.
McLaughlin et al We present A sample of 136 Institutional rearing: TSST: Participants Salivary Cortisol: 3 (Group: FCG, From Table S2: Trier F(2,137) .00751 Children who remained
2015 comprehensive data children (aged 6–30 children raised from completed three Participants expectorated CAUG, NIG ) x 6 social stress task F(2, = 5.07; in institutional care
on autonomic months) was recruited early infancy in laboratory-based ∼1.5 mL of saliva into a (Time: Peer 137) = 5.07, p = .008 Robust- exhibited significantly
nervous system from Bucharest. An institutions in procedures cryovial with a plastic straw. evaluation, Speech ness blunted SNS and HPA
(ANS) and HPA axis age-matched sample Bucharest, Romania designed to elicit a Saliva samples were stored preparation, Speech, F(2, axis responses to
reactivity from the of 72 community- versus a randomized physiological immediately at −20 °C until Math, Frustration) 137) = psychosocial stress
Bucharest Early reared children was controlled trial of response: (i) a they were shipped on dry ice ANCOVA controlling 4.54 compared with
Intervention Project recruited from foster care as an peer-evaluation to a laboratory in Boston. for sex. children randomized to
(BEIP), the only pediatric clinics in alternative to task hat was Samples were assayed for foster care, whose
randomized Bucharest and institutional rearing passive in nature cortisol and DHEA-S by using stress responses
controlled trial of comprised the NIG. for abandoned (ii) an evaluated commercially available approximated those of
foster care as an Half of children in the children. social performance luminescence immunoassay typically developing
alternative to institutionalized group task requiring kits (CLIA; IBL) children.
institutional rearing were randomized to a instrumental
for abandoned foster care cognitive
children. We intervention, resulting responses—the
provide evidence in two groups: the TSST, a widely used
for a causal link foster care group stress induction
between the early (FCG) and the group procedure that has
caregiving who received care as been used with
environment and usual [prolonged children and
stress response institutional care adolescents; and
88
system reactivity in (CAUG). A total of 138 (iii) a nonsocial

humans with effects children participated task designed to
that differ markedly in the 12-y elicit frustration
from those physiological that required
observed in rodent reactivity assessment active responses.
models. (CAUG, n = 43; FCG, n
= 48; NIG, n = 47).
Negriff et al 2015 We hypothesized At Time 1 (T1), the Maltreatment: The TSST-C: Salivary Cortisol: Six saliva Structural equation For boys, t(240) = .00004 For boys, maltreatment
that maltreatment sample was composed maltreatment group Participants were samples were obtained over models were tested maltreatment was - 4.16 was associated with
would be linked of 454 adolescents was recruited from read the beginning 90 min: the first two were using Mplus with the associated with attenuated cortisol
with dampened aged 9–13 years (241 active cases in the of a story, given 5 collected 45 min before the MLR estimator. attenuated cortisol, patterns, supporting
cortisol, which in males and 213 Children and Family min to develop the stressor and 10 min before Latent variables were and more pubertal the attenuation
turn predicts females). Time 2 (T2), Services (CFS) of a next part of the the stressor (immediately constructed for change predicted hypothesis. For girls,
accelerated Time 3 (T3), and Time large west coast city. story, and then after a 5-min relaxation delinquency (three subsequent maltreatment did not
pubertal 4 (T4) occurred on The comparison group spent 4 min protocol that included soft manifest indicators) delinquency. Table 2: predict attenuated
development, which average 1, 2.7, and 7.2 was recruited using presenting that music and a still slide of a and substance use Maltreatment --> T1 cortisol. This is
then predicts riskier years after baseline. names from school story to an beach scene). The third (two manifest cortisol (b = -.25, inconsistent with other
psychosocial The retention rate lists of children aged interviewer and a saliva sample was obtained indicators). literature showing the
functioning (higher between T1 and T2 9–12 years residing in panel of two immediately after the Maltreatment status,
.
S.E.=.06, p < .01
effects of
rates of depressive was 86.1% (n = 391), the same 10 zip codes judges who stressor was complete, and cortisol excretion maltreatment on
symptoms, between T1 and T3 as the maltreated maintained neutral the fourth, fifth, and sixth (AUCg), pubertal cortisol response for
delinquency, and was 70.9% (n = 322), sample. . facial expressions samples occurred 10, 20, stage, and depressive girls.
substance use). and between T1 and throughout the and 30 min after the end of symptoms were
T4 was 77.5% (n = task. Next, the the stressor, respectively. included as manifest
352). youth performed a variables.
challenging 4-min
serial subtraction
task before the
judges.
Peckins et al 2015 Based on previous The present study Maltreatment: TSST-C: The TSST-C Salivary Cortisol: At each Only data from T2, Frequencies and t(388) = .00029 A history of
research, it was used data from the Neglect, physical is a social and wave of data collection, T3, and T4 were used percentages for all -3.66 maltreatment was
hypothesized that second, third, and abuse, or sexual cognitive stressor participants provided six in the analyses in categorical variables associated with an
maltreated youth fourth time points of a abuse. consisting of two saliva samples by spitting order to be able to are provided in Table increased likelihood of
would be more longitudinal study of tasks: a story into a 5-ml tube to measure control for the 1. Descriptive exhibiting the blunted
likely than the effects of completion task cortisol reactivity in effects of recent analyses were cortisol profile in
nonmaltreated maltreatment on the and a mental, response to the TSST-C. Two adversity (e.g., performed and are comparison to the
youth to present health and well-being serial subtraction baseline samples of saliva recent exposure to presented in Table 2. moderate and elevated
with the of adolescents. At ask, each lasting 4 were collected, the first violence) on HPA axis Maltreated profiles, and this
unemotional (i.e., Time 1 (T1), the min. Participants immediately upon arrival activity. Descriptive adolescents pattern was consistent
blunted) pattern of sample was composed were told prior to after consent/assent forms analyses were presented with lower for both
stress reactivity at of 454 adolescents the story had been signed (at T −45 performed for all levels of cortisol than sexually/physically
each occasion of (n = 212 females) ages completion task min TSST-C) and the second variables. their nonmaltreated abused youth and
measurement. A 9–12 years (M = 10.95 that their immediately following a 5- Independent t tests peers across all emotionally
blunted pattern was years). Time 2 (T2), performance min relaxation period (at T were performed to samples at T2 and T3. abused/neglected
expected to be Time 3 (T3), and Time would be −10 min TSST-C). Four test for mean Refer to Table 2 for youth, but only at T2
present in 4 (T4) occurred compared to the poststressor saliva samples differences between Cortisol Time 2 t-test. and T3
maltreated youth approximately 1, 2.5, performances of were collected; one sample maltreated and
and 4.5 years after their peers. immediately following the nonmaltreated youth
89
regardless of baseline, respectively. stressor (+0 TSST-C) and in recent exposure to

maltreatment type. The mean age of three samples at 10-min violence, recent
youth at T2, T3, and increments (+10 min, +20 exposure to trauma,
T4 was 12.11, 13.69, min, and +30 min TSST-C). and cortisol
and 18.23 years. concentrations.
Schalinski et al To allow a Patients with stress- Early Trauma Idographic Trauma Salivary Cortisol: Salivary 2 (Group: with and Salivary cortisol F(2.29, .045 In addition to previous
2015 differentiation of related disorders (N = Inventory (ETI): The Interview: he cortisol, heart rate, systolic without sexual levels differed for the 71) = reports, we found
phasic and tonic 43) were recruited Childhood adversities interview about and diastolic blood pressure abuse) x 4 (time) group with and 3.08 distinct activity
levels of cortisol, through the up to age 18 were traumatic were recorded at 4 different repeated measures without childhood patterns of cortisol
the current study outpatient clinic for recorded using the experiences time points during a ANOVA. sexual abuse (F(1, 31) (measured by
measured acute refugees at University Early Trauma included (1) the standardized interview. = 6.25, p = .018, η2 = responder rates facing
cortisol stress of Konstanz. To Inventory (ETI); event checklist, (2) .17). Interestingly, the trauma
reactivity and long- compare results of German version. In an followed by the the ANOVA showed a assessment, mean
term hair cortisol hair cortisol levels, we interview 55 items interview of significant time x salivary cortisol and
levels. Specifically, further recruited 12 that score on four childhood group interaction cortisol accumulated in
we investigated the healthy control domains of adversities. (F(2.29, 71) = 3.08, p hair) among patients
impact of childhood subjects from the maltreatment and = .045, η2 = .09). displaying hints of
sexual abuse (and general community. abuse were assessed Whereas the group neurobiological
other adverse Forty-three female (general traumatic with childhood sexual subtypes.
experiences) on survivors of violence experiences, abuse responded
saliva cortisol participated in the emotional abuse and with a down
reactivity and study. The age of the neglect, physical regulation of cortisol,
autonomic participants ranged maltreatment and the group without
parameters towards from 16 to 56 years sexual abuse). childhood sexual
an idiographic (M = 34.9, SD = 10.5). abuse did not differ
assessment of from baseline.
traumatic
experiences in
female patients
with stress-related
psychopathology.
Voellmin et al 2015 Therefore, the aim The sample included Adverse Childhood The MIST was used Salivary Cortisol: Saliva was 4 x 6 repeated Repeated measures F(2.33, .00192 Our results are in line
of the present study 104 young and Experiences: ACEs to induce a collected at seven measures ANOVA. analysis of cortisol 221.60) with previous reports
was to replicate the healthy females in the before the age of 18 psychosocial stress measurement points, GLM for repeated response to stress = 5.89 of attenuated
findings of age of 18 to 25 years. years were assessed response. The whereof two took place measures served to showed a significant endocrine (as well as
attenuated retrospectively using a MIST is a before the stress test (-10 determine the interaction of time x cardiovascular (stress
endocrine and German translation of standardized and -1 min) and five after effects of ACEs on ACE total sum score responses to a
cardiovascular the "Early Trauma psychosocial stress the stress test (+1, +10, +25, endocrine and [F(2.33, 221.60) = psychosocial stress test
stress reactivity in Inventory-Self Report" test during which +40, and +55 min) using cardiovascular 5.89, p < .001; ηp2 = in healthy adults with a
association with a (ETI-SR), which participants have salivettes. responses. In these .06] as well as a history of adverse
history of ACEs in a includes general to solve arithmetic models, the ACE total significant main childhood experiences.
young, healthy, trauma, physical, tasks displayed on sum core as well as effect of ACE total Importantly, blunted
female adult emotional, and sexual a computer screen the different scores sum score [F(1,95) = cortisol and heart rate
sample. abuse items. under time for duration and age 7.52, p < .01; ηp2 = responses were
pressure and social of occurrence were .07] independent of
evaluation. used as continuous emotional responses,
variables to examine suggesting that the
effects of time, ACE diminished endocirne
90
scores, and the and cardiovascular

interaction of time stress reactivity cannot
by ACE scores. be explained by a
reduced emotional
reaction to stress.
Aiyer et al 2014 Based on previous The present study Cumulative Exposure Interview with Salivary Cortisol: We utilized We conducted a Introducing r(246) = .03407 We found evidence for
findings, we included 266 to Violence (ETV): sensitive salivary cortisol as our hierarchical cumulative exposure -.13 the influence of
hypothesized that participants who had Cumulative ETV was information: The measure of the stress regression analysis to to violence in step 2 cumulative exposure to
cumulative cortisol data and data calculated by saliva was response, as it reflects the examine the unique resulted in improved violence during
exposure to on our study variables standardizing each collected during free portion of cortisol in contributions of model fit. Further, adolescence and early
violence during across waves 1–7. The subscale (witnessing wave 7 at three plasma recommended by cumulative exposure cumulative exposure adulthood on cortisol
adolescence would cortisol collection violence and violence points during an in- many researchers. The saliva to violence, mother to violence was responsivity in early
be associated with a procedure is victimization) across person paper-and- was collected during wave 7 support and father negatively adulthood. Focusing on
hyporesponsive described in detail each wave (waves 1– pencil interview at three points during an in- support to association with a uniquely high-risk
cortisol pattern in below. Participants 7),and adding scores concerning person paper-and-pencil cumulative exposure cortisol area under sample, we found that
early adulthood. were between ages to represent sensitive interview concerning to violence. At each the curve with participants exposed to
Specifically, 13.9 and 16.9 years at cumulative exposure information (e.g., sensitive information. step in the analysis, respect to increase chronic violence
participants wave 1 (mean = 14.9; to violence across substance use, Cortisol values were log we also examined (See Table 3). demonstrate signs of
exposed to violence SD = .64) and between adolescence and early sexual risk transformed to reduce the the contribution of an attenuated stress
were predicted to ages 21.9 and 24.3 adulthood. behavior, violent skewness of the distribution. participant race, sex, response.
exhibit signs of an years at wave 7 (mean behavior, and To make the repeated and interview start
attenuated stress = 22.1; SD = .66). The discrimination). measurements of salivary time to reduce
response. analytic sample was Such questions cortisol more useful, we potential
80 % African were thought to be assessed the cortisol confounding.
American, 17 % somewhat stressful response by calculating area
Caucasian, and 3 % for participants, under the curve with respect
Mixed Race. The based on their to increase [AUC (I)]. This
analytic sample was sensitive nature. allowed us to assess changes
43 % Male and 57 % in cortisol secretion over
Female. time.
Grimm et al 2014 In the present Thirty-two healthy Childhood Trauma Montreal Imaging Salivary Cortisol: Nine saliva Behavioral data were There was a main F(1,27) .018 Subjects who
study, we sought to male subjects (age Questionnaire (CTQ): Stress Task (MIST): samples were collected with analysed using t-tests effect of ELS (F(1,27) = 6.24 experienced ELS
clarify the impact of 28.4 +/- 4.5; range The CTQ consists of 28 Psychological the Salivette sampling for independent = 6.24, P = 0.021) on showed both blunted
ELS on stress 21–37 years; IQ 112 items that are stress was induced device (Sarstedt Inc.) samples. cortisol stress reactivity and
responsivity and +/- 15) were recruited assigned to the using the Montreal throughout the experiment Physiological data concentrations at T3 limbic deactivation
associated neural out of a preexisting following 5 subscales: Imaging Stress Task at the following time-points: were analysed using (immediately after during stress. Our
activity as well as to psychologically and emotional neglect, (MIST). The MIST Baseline (saliva 1= T0), 45 repeated measures the MIST) and of findings demonstrate
determine OXT somatically healthy emotional abuse, uses a block min after OXT/placebo analyses of variance substance on cortisol that healthy subjects
effects on these community-dwelling physical neglect, design, and administration (saliva 2 / 6 = (ANOVAs) with the concentrations at T2 who experienced
parameters in sample (N = 541). physical abuse and consists of mental T1), immediately before within-subjects (immediately before moderate to severe
healthy subjects sexual abuse. arithmetic (saliva 3 / 7 = T2 (20min factor substance the MIST; (F(1,27) = emotional neglect and
with and without a Participants were challenges that after T1)) and after the MIST (OXT, placebo) and 4.89, P = 0.039) with abuse in their
history of ELS. We recruited for the ELS must be answered (saliva 4 / 8 = T3 (20min the between- lower concentrations childhood not only
therefore, firstly, group, if their CTQ- under time after T2)) as well as 20min subjects factor ELS. in the ELS group and show attenuated
aimed to investigate score was no lower pressure. after the completion of the Greenhouse–Geisser the placebo hormonal and limbic
the effects of OXT than 3 points below MIST (saliva 5 / 9 = T4). corrections were condition, reactivity during
on deactivation of the 90th percentile, applied where respectively. psychosocial stress, but
limbic regions whereas control appropriate. Further might also have an
91
implicated in subjects had scores no statistical analysis altered OXT sensitivity,

psychosocial stress higher than 3 points was conducted using since the previously
response. Secondly, above the 50th t-test comparisons. reported stress-
we aimed to percentile. All tests were two- buffering effect of OXT
examine whether tailed and the was only found in
these OXT effects significant threshold subjects who did not
might differ in was set at P < 0.05. experience ELS.
subjects
with/without ELS.
Pagliaccio et al We tested the Data were analyzed Stressful and Behavioural Tasks: Salivary Cortisol: Salivary We used hierarchical AUC cortisol levels t(118) = .003 Congruous with the
2014 hypotheses that from 120 children (58 Traumatic Life Events: The second saliva cortisol was collected three linear regressions in were positively -3.00 literature suggesting
genetic profile females; 57.5% White, Analyses examined collection occurred times. SPSS to predict predicted by genetic that life adversity
scores and early life 30.0% African- stressors experienced approximately cortisol and regional profile scores (b = results in blunted HPA
stress exposure American, and 12.5% from birth through 30min after brain volumes. 0.32, t = 3.04, p = axis responses to acute
would predict of other or mixed the baseline separation from Predictors were 0.003) and negatively stress (Carpenter et al,
increased cortisol race) enrolled in the assessment when the parent, during added in steps to predicted by life 2007; Lovallo et al,
responses to Preschool Depression participants were 3–5 which time the understand their events (b = - 0.28, t = 2012; Ouellet- Morin et
laboratory stressors Study (PDS), a years old (current child performed a effects alone and -3.00, p = 0.003), al, 2011), we found
in preschool-age prospective subsample mean age variety of controlling for even when that the number of
children. longitudinal study of at baseline behavioral tasks. covariates. controlling for all stressful and traumatic
preschool-age assessment = 4.45 Following this Regressions first covariates. Together life events experienced
children (N = 306) years, SD 1⁄4 0.77), at collection, the child controlled for these factors by preschool age
conducted at the which time cortisol performed more ethnicity and sex. accounted for 8% of negatively predicted
Washington was collected. We behavioral tasks Next, centered, the variance in cortisol levels.
University School of summed the instances and several stress- continuous variables cortisol beyond
Medicine Early of stressful and /frustration- for genetic profile ethnicity and sex.
Emotional traumatic life events inducing episodes scores and life events
Development reported by parents from the were entered.
Program (WUSM during the PAPA. Laboratory Interactions between
EEDP) in St Louis. Temperament these factors and
Assessment interactions with sex
Battery. were then entered.
92
Shenk et al 2014 Adopting a multiple One hundred ten Child Maltreatment: Stressor Paradigm: Salivary Cortisol: The first A multiple mediator r(104) = -.21 (Table 2) r(104) = .03073 The maltreated group
levels of analysis adolescent females Child maltreatment The performance sample, Cortisol-Rest, was model was employed -.21 did display a
approach, the between the ages of was determined by a aspect of the collected approximately 25 to identify the total hypocortisol reaction
current study 14–19 years of age CPS investigation that stressor paradigm minutes (M= 25.30, SD= .24) and specific indirect during the stressor
simultaneously participated in this resulted in a involved each after participants began effects of RSA-Stress, condition once
tested the indirect study. wo distinct substantiated/indicate participant their study appointment to Cortisol-Stress and relevant demographic
effects of groups of adolescent d designation of child completing a series give them time to acclimate AAQ scores when variables and health-
experiential females participated. maltreatment. Of the of affect to the research environment explaining the related behaviors were
avoidance, RSA, and A child maltreatment 51 participants in the recognition tasks. and procedures. Samples relationship between controlled. This finding
cortisol in mediating group (n= 51) was maltreatment group, Participant two through five were child maltreatment is consistent with
the relationship recruited from Child 49% experienced responses were collected 5-, 10-, 20- and 30- and subsequent previous research
between child Protective Service sexual abuse, 45% timed and each minutes post stressor to PTSD symptoms. The showing a relationship
maltreatment and (CPS) agencies experienced physical participant was detect the maximum cortisol multiple mediator between child
subsequent PTSD investigating abuse, and 16% asked to identify response to the combined model was maltreatment and
symptoms using a allegations of physical experienced physical the emotion as stressor, Cortisol-Stress. performed with hypocortisol reactions
sample of neglect or contact neglect with 10% quickly as they Saliva was obtained through Mplus v.6 using the during stress
maltreated and physical or sexual experiencing more could while not passive drooling into 20 mL maximum likelihood paradigms.
non-maltreated abuse..A non- than one form of making any polypropylene vials and estimator to account
adolescents. maltreated, abuse. Comparison mistakes before stored at −80°C until for missing data.
comparison group (n= females were the time elapsed. assayed.
59) was recruited screened and The interpersonal
using posted flyers in excluded if they had a stressor involved
a primary care substantiated case of participants
outpatient clinic child maltreatment viewing a series of
servicing the general within the twelve video-clips of
medical complaints of months prior to study parent-adolescent
at-risk, adolescent participation. conflict.
females.
Sumner et al 2014 We hypothesized A community-based Childhood Trauma TSST: Participants Salivary Cortisol: Saliva 3(time) x 2(childhood There was a F(1.36, .016 In this community
that child sample of 168 Questionnaire: The completed the samples were taken after maltreatment) significant Time x 205.90) sample of adolescents,
maltreatment adolescents aged 13— CTQ is a 28-item scale Trier Social Stress the initial baseline period, ANOVA. Child Maltreatment = 5.05 a history of child
would be associated 17 was recruited for that assesses the Test (TSST; 15 min following the interaction, F (1.36, maltreatment was
with blunted participation at frequency of Kirschbaum et al., beginning of the speech 205.90) = 5.05, p = associated with
cortisol reactivity to schools, after-school maltreatment 1993), a widely- portion of the TSST .02, h2 = .03. As blunted cortisol
psychosocial stress programs, medical exposure during used stress (reactivity period), and 15 shown in Fig. 2, reactivity to a
as a function of clinics, and the childhood and induction min following completion of adolescents with a psychosocial stressor.
rs110402 genotype general community in adolescence. Three procedure that has the recovery period. history of child
(i.e., primarily Boston and types of abuse are been used with maltreatment
among G allele Cambridge, MA. assessed: physical, children and exhibited lower
carriers). sexual, and emotional. adolescents cortisol levels at the
Childhood (Buske-Kirschbaum second [t(156) = 1.83,
Experiences of Care et al., 1997; Stroud p = .07] and third
and Abuse (CECA) et al., 2009). [t(146.23) = 2.69, p =
Interview: The CECA .01] cortisol
assesses multiple assessments
aspects of caregiving compared to those
experiences, including without a history of
physical and sexual maltreatment.
abuse.
93
Suzuki et al 2014 The current study A total of 80 Childhood Trauma Stress inducing Salivary Cortisol: Salivary 2 (depression) x 2 A two-way ANCOVA F(1,68) .039 We found a significant
aimed to examine participants (28 males Questionnaire (CTQ) images from cortisol was measured to (trauma) x 3 (time) (childhood trauma × = 4.4; interactive effect of
stress reactivity in and 52 females) were was used to assess International investigate the effects of ANOVA. depression) showed Robust- depression and
individuals with and recruited into four experiences of Affective Picture image exposure on HPA axis a significant ness childhood trauma on
without a history of groups comprising: 17 childhood trauma. . System (IAPS): reactivity. Saliva samples interaction of F(1,66) stress reactivity. The
childhood trauma healthy participants Those who scored at Each participant were obtained by passive depression and = 4.6 first main finding is
by measuring with a history of or above ‘moderate- viewed 48 stress- drool into sterile plastic childhood trauma that, irrespective of
cortisol responses childhood trauma severe’ level in at inducing images as tubes. Over the course of (F(68, 1) = 4.4, p < individuals’ depressive
to the passive (Healthy Abused: H/A) least one of the five well as 48 non- the sessions, two saliva .05). The main finding states, those with a
viewing of stressful and 24 without subscales (emotional stress inducing samples were collected 5 is that stress history of childhood
images, specifically (Healthy Non-abused: abuse (EA), physical images. Stress min apart on three reactivity was higher trauma showed a
including images H/NA); and 21 abuse (PA), sexual inducing images occasions: before session 1 in non-abused reduced cortisol
relevant to depressed patients abuse (SA), emotional comprised 24 (S1 and S2), between depressed patients reactivity to stress.
childhood trauma. with a history of neglect (EN) and negative affect session 1 and 2 (S3 and S4), than in all of the Although there were
In addition, childhood trauma physical neglect (PN)) images chosen and after session 2 (S5 and other groups. no differences in
participants with (Depressed Abused: of the CTQ were then from IAP) and 24 S6). Another two-way relation to admission
and without a D/A) and 18 without assigned to the images chosen ANCOVA (childhood status, childhood
diagnosis of current (Depressed relevant experimental from a public trauma × depression trauma was related to
depression were Nonabused: D/NA). groups (H/A or D/A). photo archive to severity) showed a overall lower cortisol
studied to Those who scored represent each of main effect of responses.
investigate whether below the same level four types of childhood trauma
cortisol stress were assigned to non- childhood trauma with overall larger
reactivity may abused groups (H/NA (emotional abuse, reactivity in non-
underlie resilience or D/NA). physical abuse, abused group than
or vulnerability to sexual abuse and abused group (F(66,
depression. emotional neglect. 1) = 4.6, p < .05).
Trickett et al 2014 Because mal- The maltreatment Maltreatment: The TSST-C: This Salivary Cortisol: We 6 (time) x 2 (group: A significant F(5, .00004 Findings indicated that
treated youth tend group (N = 303) were nature of each modification of a collected six saliva samples, maltreatment vs. maltreatment group 426) = for these young
to have repeated recruited from active maltreated common method two before and four after control) repeated x sampling time 5.78; adolescents,
and chronic cases in a county adolescent’s used to elicit an the TSST-C. The rationale for measures ANOVA interaction effect, Robust- maltreatment was
exposures to Department of experience of abuse HPA stress the number of samples and F(5, 426) = 5.78, p < ness associated with
maltreatment, we Children and Family and neglect was response in a the timing of intervals was .001, was found, F(2, attenuated cortisol
expected the Services (DCFS) of a obtained by laboratory context based on the importance of suggesting the 424) = reactivity overall and
maltreated youth to large western U.S. examining the DCFS includes a social accurately assessing both cortisol response 3.89 that this result was
exhibit an city. The inclusion case record for each challenge baseline levels of cortisol curve differed pronounced in those
attenuated cortisol criteria were (1) a new individual and (composing and and reactivity to the between maltreated youth who
response to the substantiated referral entering this performing g a stressor. samples involved and non-maltreated experienced more
laboratory stressor to DCFS in the information into the story in front of a measures selected for adolescents. severe maltreatment
as compared with preceding month for Maltreatment Case panel of judges) content unrelated to trauma as indicated by
the non-maltreated any type of Record Abstraction and a cognitive and stress. Data collection maltreatment that
group. maltreatment; (2) Instrument (MCRAI). challenge (oral occurred primarily in the included physical
child age of 9–12 serial subtraction). afternoon, with 96% of and/or sexual abuse as
years; The comparison participants providing the well as, frequently,
group (N = 151) was baseline saliva sample other forms of
recruited via mail between 1200 and 1700 hr. maltreatment, and as
using names from The earliest baseline sample noted in the
school lists of children was collected at 1223 hr, description of the
aged 9–12 years and the latest was collected sample, by more
reports to DCFS.
94
residing in the same at 1735 hr. We control for

10 zip codes. time of day in all analyses.
Feldman et al 2013 We hypothesized Participants included War Exposure: To elicit the stress Salivary Cortisol: Baseline, A repeated-measure The results indicated F(2, .00209 Children with PTSD
that children with 232 children from 232 Citizens of Sderot response, we used following challenge, and multivariate analysis significantly greater 197) = exhibited consistently
PTSD would exhibit families aged 1.5 to 5 were exposed a detailed recovery. of covariance stress reactivity 6.37; low CT and sAA,
hypocortisolism, years (M = 33.08 repeatedly to rocket evocation of (MANCOVA) with among controls as Robust- exposed-no-PTSD
expressed in low months, SD = 10.89) attacks from Gaza traumatic group (controls, compared to the ness displayed consistently
baseline CT and and their mothers (M over a period of memories, exposed children exposed children, F F(2, high CT and sAA, and
minimal reactivity, = 31.27 years, SD = several years, had combined with a with PTSD, and (2, 197) = 6.37, p < 197) = controls showed
and would select 5.55, range = 22.3– only 15 s to enter fear paradigm that exposed-no-PTSD) .01. 5.82 increase in CT
withdrawal 47.4). The sample protected spaces after involved the and child gender as following challenge
strategies to included 47.6% males hearing the alert presentation of the between-subject and decrease at
regulate fear as and 47.1% firstborns. sirens, and were increasingly fearful factors and child age recovery and low sAA.
compared to exposed to frequent novel objects as a covariate was
exposed children mortar shelling. adapted from the computed for three
without PTSD. Lab-TAB. measurements of CT.
Bosch et al 2012 This study extends The data were Life Adversities: Social Stress Test: Salivary Cortisol: HPA-axis Repeated-measures Figure 1: Subjects F(1,441) .028 The transition in
previous findings by collected in a focus Including traumatic The experimental responses towards the social general linear models with the highest = 4.8 cortisol activity from
exploring the sample of TRAILS experiences measured session during T3 stress test were assessed by (GLM, Greenhouse— number of hypersecretion after
effects of (Tracking Adolescents’ via interview and self- consisted of a four cortisol samples, before Geisser corrected) adversities during adversities before age
adversities from the Individual Lives report. number of the introduction of the were used to early adolescence 11 and hyposecretion
prenatal period up Survey), a large different social stress test (‘pretest’), examine (1) the main (ages 12— 13) had afterwards emphasizes
to adolescence on prospective challenges directly after the test effects of adversities the lowest cortisol puberty as a major
adolescents’ cortisol population study of (orthostatic stress, (‘during test’), 20 min after per period and (2) overall level (F1,441 developmental period
stress responses. Dutch adolescents spatial-orienting the test (‘end of test’), and the interaction of = 4.80, p = .03). of the HPA-axis. Results
We examined from age 11 to 21. task, gambling 40 min after the test (‘post pre/postnatal risk highlight the
adversities during Complete HPA-axis task, startle-reflex test’). with each later importance to take the
five age periods and adversity task, and social adversities on the timing of stress
(pre/postnatal, 0—5 measurements were stress test). The HPA-axis. Adversity exposure into account.
years, 6—11 years, available for 640 social stress test variables were tested In addition to
12—13 years, 14— participants. (Groninger Social simultaneously and programming effects,
15 years), to Adolescents were Stress Test) hence their effects pre/postnatal adversity
investigate their excluded if stressful involved a were adjusted for interacts with
specific effects on life events were standardized each other. childhood adversity in
the HPA-axis. We assessed more than 3 protocol inspired Investigated producing deviant
anticipated that months before (n=60) by the Trier Social outcome variables cortisol levels. Puberty
effects of or more than six Stress Test were cortisol may be marked by a
adversities on the months after the (Kirschbaum et al., reactivity and cortisol transition in how
cortisol responses experiment (n = 60). 1993). overall level. Cortisol adversities affect the
would depend on In addition, we reactivity reflects the HPA-axis, with cortisol
their timing, and excluded adolescents pattern of response, hypersecretion before
that pre/postnatal with cortisol outliers while cortisol overall age 11 and
adversity would (>3 SD, n = 11), as well level represents the hyposecretion after
render individuals as adolescents who average level across age 11.
more sensitive to used corticosteroids all four samples
persistent effects of (n = 38), leaving a
later adversities on total of 471 (32.3%
the HPA-axis.
95
low- risk) participants

for analysis.
Fisher et al 2012 We aimed to follow- Our participants Early Adversity: Early TSST-C: This Salivary Cortisol: Five The children’s AUCI There was a chi2(1) .034 Area under the curve
up Lester et al.’s included 57 children adversity (i.e., abuse standardized salivary cortisol samples scores were then significant main = 4.49 with respect to
findings, focusing placed in foster care and neglect protocol is were collected at 15-min dichotomized to effect of early increase (AUCI)
on whether foster during the preschool experienced prior to designed to induce intervals beginning compare children adversity, physical analyses revealed that
children in a similar period. (Results with placement in foster moderate immediately prior to the with increases and abuse in particular, prenatal substance
age range to the 53 children) care during the psychosocial stress TSST-C. decreases in cortisol Pearson χ2(1, N = 53) exposure or physical
children in the prior preschool period) was in laboratory levels over time. = 4.49, p = .03. abuse significantly
study (with varying coded from child settings and has Pearson’s chi-square Among the children increased the
profiles of prenatal welfare system been shown to tests were conducted who experienced likelihood of a negative
substance exposure records using the produce elevations to examine the physical abuse, 77% AUCI (i.e., little or no
and maltreatment) Maltreatment in cortisol in adults relations between (n = 17) showed a HPA reactivity). Among
would show a Classification System and children. early adversity, decreased cortisol children with prenatal
pattern of blunted (MCS). prenatal substance response over time. substance exposure
cortisol responsivity exposure, and and physical abuse,
to a laboratory salivary cortisol 85% exhibited a
psychosocial response over time. negative AUCI.
stressor.
Lovallo et al 2012 The present study Sample includes 354 Adversity: Lifetime TSST: 30 minute Salivary Cortisol: Saliva 3 (adverse events: 0 The size of the F(2,348) .00336 The present study
examines cortisol participants (158 m; adversity was based prestress baseline, samples were collected and vs 1 vs 2 or more) x cortisol response = 5.79 shows that men and
and heart rate 196 f). The study on C-DIS-IV items: when subjects sat taken at: awakening, arrival 2(gender: male diminished as the women who
responses to a population included Physical or Sexual quietly and read at the lab, min 10 and 20 of versus female) number of lifetime experience more
standardized healthy young adults Adversity (Have you general interest baseline period, at min 15, ANOVA adverse events adverse life events
psychological stress free of psychiatric ever been mugged or magazines, 30, and 45 of the stress increased [F(2,348) = before age 15 also
protocol comorbidities but who threatened with a followed by 45 min protocol or continued 5.79, p < .004]. have smaller cortisol
incorporating had experienced a weapon. Have you of behavioral resting protocol, and 15 and and heart rate
simulated public range of physical and ever been raped or stress. Stress 30 min poststress or rest, responses to
speaking and psychological adverse sexually assaulted?) included simulated and at bedtime. Stress psychological stress.
mental arithmetic events in childhood and Emotional public speaking reactivity as reported here These findings seem to
challenges. The and early adolescence Adversity (Before you followed by mental was measured at min 10 and illustrate an impact of
study population were 15, was there a arithmetic. 20 of the baseline period stress exposure in
included healthy time when you did not and at min 30 and 45 of the childhood and
young adults free of live with your stress period contrasted adolescence on the
psychiatric biological mother for with samples taken at the regulation of the stress
comorbidities but at least 6 months? same time during the axis in adulthood.
who had Each person was extended resting protocol.
experienced a range assigned an adversity
pf physical and score ranging from 0
psychological (no adverse events) to
adverse events in a maximum of 5.
childhood and early
adolescence.
96
Peckins et al 2012 Cortisol reactivity Participants were 124 Exposure to violence: TSST-C: The TSST-C Salivary Cortisol: Children Two multiple linear Refer to Table 2: r(60) = .01388 The hypothesis that
(CR) was (n = 64 females) My Exposure to consisted of a story passively drooled into a 5- regression models r(60) = 0.311, p = 0.311 adolescents with
hypothesized to be adolescents from a Violence (MyETV) is a completion task mL tube until 4-mL of saliva were developed and .01388. greater recent ETV
attenuated in small city and rural self-report measure of followed by was collected or 5 minutes Time 2 CR was would have
children exposed to communities. In the ETV occurring within mental, serial passed. The samples were regressed on age, significantly lower CR
violence. first wave of the past 12 months subtraction while collected 20 min and 5 min Time 1 lifetime ETV, was supported for
measurement at Time and over the child’s being monitored pre-TSST-C, and 0 min, 10 Time 1 CR, and males. The findings
1, girls were ages 8 (n lifetime. The MyETV is by two min, and 20 min post-TSST- symptoms of held even after
= 21), 10 (n = 20) or 12 a well-validated confederate C. psychopathology controlling for
(n = 23) years and measure consisting of judges. Different (Model 1), and Time symptoms of
boys were ages 9 (n = six scales (Lifetime story completion 1 covariates and ETV psychiatric disorders,
19), 11 (n = 21) or 13 and past year tasks and judges occurring in the 12 age, and CR and
(n = 20) years. exposure to: were used at Time months before Time lifetime ETV at Time 1.
witnessed violence, 1 and Time 2. 2 (Model 2). Symptoms of MDD and
victimization, and Participants were Symptoms of MDD, GAD were positively
total exposure). told their GAD, and PTSD were associated with
performance included to examine lifetime ETV at Time 1,
would be the association indicating that
compared to between ETV and adolescents with
others of the same depression and greater lifetime ETV
age. anxiety, and to rule report more symptoms
out the effect of of MDD and GAD.
symptoms of
psychopathology on
CR.
Sturge-Apple et al In accordance with Participants included Interparental Simulated Phone Salivary Cortisol: Baseline We examined Analyses revealed r(199) = .03863 Consistent with the
2012 EST 201 two-year-old Violence: First, Argument Task: samples were collected on unconditional latent that interparental .146 hypocortisolism
conceptualizations children and their mothers self-report of Children and their average within the first 15 growth curve models violence was a hypothesis, greater
of differentiation in mothers in a inter-partner violence mothers minutes upon arrival to the (LGC) for children’s significant predictor levels of adversity in
children’s emotional moderately-sized were assessed using participated in the laboratory, after maternal cortisol reactivity of children’s cortisol family relationships
security in family metropolitan area in subscales from the Simulated Phone consent had been obtained. using MPlus 6.0 reactivity in the SPAT were associated with
systems, we the northeastern Revised Conflict Argument Task Cortisol – SPAT—Two post- statistical software. paradigm (β= -.21, p children’s lower
hypothesized that United States. We Tactics Scale (CTS2). (SPAT) to assess task saliva samples were Given the presence = .039). This pathway cortisol reactivity to
interparental administered the The CTS2 Physical children’s reactivity also obtained to assess of significant accounted for 8% of the family stressors
violence would be abbreviated version of Assault subscale to interparental trajectories of cortisol individual differences the variability in
primarily associated the Physical Assault contains 24 items conflict. During this change across three in cortisol cortisol reactivity in
with children’s Scale of the Conflict designed to assess procedure, assessments. The two trajectories in both the SPAT paradigm.
cortisol reactivity to Tactics Scale 2 (CTS2) maternal and partner children witnessed post-task saliva samples the SS and the SPAT The beta for this
a standardized to insure that roughly acts of physical live simulations of were obtained paradigms, our next effect was negative
laboratory paradigm equal proportions of violence toward each their parents approximately 25 and 37 set of analyses which indicates that
in which children participating mothers other in the inter- engaging in a minutes after the end of the examined as interparental
were exposed to a experienced (a) no partner relationship. conflict and a simulated disagreement, the interparental violence increased,
simulated violence (40%), (b) subsequent period in the SPAT to violence and children’s cortisol
interparental mild/moderate resolution over the correspond roughly with the maternal emotional reactivity was lower
conflict. physical violence telephone. Each midpoints of the two peak unavailability as on average. This
(24%), and (c) severe exchange lasted periods of cortisol reactivity predictors of finding supports the
approximately 1 to stressors. hypocortisolism
97
physical violence minute and was children’s cortisol hypothesis for the
(36%) in the last year. interspersed by a functioning. interparental
three-minute free context.
period.
Carpenter et al In contrast to our One hundred and ten The 28-item version of TSST: protocol Salivary Cortisol: To Repeated measures Individual models F(4,109) .00105 In the present study
2011 prior study, cortisol women, ages 18 to 61 the Childhood involves public minimize the effects of General Linear assessing the effects = 4.95 involving a sample of
concentrations years, were enrolled Trauma speaking role- diurnal variation in cortisol Model (GLM) was of moderate– severe 110 women from the
were measured in and completed all Questionnaire (CTQ) - playing, and response, the TSST was used to determine maltreatment community, we found
saliva rather than in assessments in the generates a total mental arithmetic administered at the same the effects of according to a significantly lowered
plasma. We study protocol. score and subscale tasks in front of a time each test day. Subjects maltreatment on maltreatment type salivary cortisol
approached this scores for five types of panel of were instructed not to eat cortisol reactivity to revealed that response to a
study by examining maltreatment confederate or exercise for 1h before the TSST in models that physical abuse (PA) psychosocial stress task
the effect of each of (emotional abuse, judges. distress. test. Saliva samples were included CTQ data was a significant among subjects who
five different physical abuse, sexual collected at time points 0 (summarized as total predictor of lower reported a history of
subtypes of abuse, emotional (1:45 PM; immediately score for 28-item cortisol response to childhood physical
maltreatment neglect, and physical before meeting the judges scale, dichotomized the TSST (effect of PA abuse.
independently and neglect). CTQ subscale and being told about the scores for each of group—F= 4.95,
hypothesized that scores reaching the public speech and arithmetic five subscales, or as p<0.005).
exposure to early threshold for the recitation), 15 min (2:00 PM; the total number of
life adversity would "moderate to severe" the end of the anticipation subscales scored
be associated with range were period during which the positive for
diminished cortisol considered positive subject prepared his or her maltreatment),
response to acute for maltreatment. speech), 30 min (2:15 PM; estrogen use
psychosocial stress Subjects scoring several minutes after (dichotomized), and
challenge. "none or minimal" for completing the subject age. Huynh–
each subscale were speaking/arithmetic test), 45 Feldt adjustment was
considered controls. min, 60 min, 75 min, and 90 made for violations
min of sphericity
assumptions.
Ouellet-Morin et al Based on previous This study sample Bullying TSST: One hour Salivary Cortisol: We 5 (time) x 2 (bullying) Repeated-measures F(2.2, .02113 This study provides
2011 findings suggesting comprised 30 12-year- Victimization: We after arrival, twins collected five samples of repeated-measures ANOVA showed 122.9) evidence for a causal
that early-life stress old MZ twin pairs derived a cumulative took part in an saliva to measure the ANOVA. distinct patterns of = 3.82 effect of early-life
is associated with (mean [SD] = 12.53 index of frequent, adapted version of cortisol response to the PST. cortisol secretion stress on cortisol
abnormal cortisol [0.52]) discordant on chronic, and severe the TSST for Saliva was collected by over time between reactivity in human
secretion, we bullying victimization bullying victimization children, which asking children to use a bullied and beings. Our
hypothesized that and with valid cortisol by summing the includes a social straw to pass through 1 ml nonbullied MZ twins psychosocial stress test
bullied twins would data (43.3% males). scores to the three stressor (speaking of saliva into the cryovials. (time × elicited a cortisol
have impaired Hence, this substudy questions asked to in front of judges) The first samples were bullying: F2.2,122.9= response in nonbullied
cortisol responses sample included one mothers described and a cognitive collected 20 and 2 minutes 3.82, P = .02). children compared
to the PST whereas twin who had been above across age 7, stressor (mental before the PST. A third with a blunted
their nonbullied MZ the victim of bullying 10, and 12 arithmetic). sample was collected response in bullied
co-twins will show (n = 30 children) while assessments. The immediately at the end. A children.
an increase in their co-twin had not victims of frequent fourth and a fifth samples
cortisol secretion (n = 30). bullying who were collected 25 and 35
after exposure to experienced frequent minutes after the start of
this experimental psychological and the tasks.
stress task. physical harm had a
maximum score of 6
98
at each time point (7,

10, and 12 years).
Ouellet-Morin et al We examined From the total E-Risk Childhood Harm: The Psychosocial Salivary Cortisol: Cortisol We examined Comparison children t(188) = .00451 This study provides
2011b whether 12-year- sample, we identified Childhood harm Stress Test (PST) was measured through the cortisol response exhibited the -2.875 support for the
old 190 12-year-old consisted of either a took place at the collection of five saliva using a piecewise expected increase in hypothesis that early
maltreated/bullied children (50.5% history of childhood research samples. We asked children latent growth curve cortisol after the PST, exposure to adverse
children had lower males) eligible to maltreatment or laboratory in the to use a straw to pass model. We then whereas life conditions could
cortisol responses participate in a frequent bullying early afternoon through 1 mL of saliva into tested whether maltreated/bullied result in lower HPA axis
to psychosocial substudy of cortisol. victimization. A total (mean [SD] = the cryovials. The first two maltreated/bullied children did not reactivity to stress in
stress than same- Children were of 64 children from 1:45 pm [22 samples were collected 20 and comparison (Table 2). childhood. Our
aged children. selected if they were this substudy sample minutes]). A video minutes and 2 minutes children showed laboratory stress
identical twins and of 190 participants camera was before the PST. A third distinct patterns of paradigm elicited a
only one twin within were harmed (33.7%); installed in a room sample was collected 2 cortisol response to significant cortisol
the pair was bullied. 23 were maltreated to record the minutes after the task. The stress. We extracted response in
Most twins were by an adult (12.1%), cognitive and fourth and fifth samples the intercept, comparison children
Caucasian (91.6%) and 27 were frequently public speaking were collected 25 minutes baseline, and but not in
one third of the bullied by their peers tasks adapted from and 35 minutes after the response slope maltreated/bullied
families came from a (14.2%), excluding the Trier Social start of the task. estimates derived in children for whom
low socioeconomic those who were only Stress Test for Mplus to test, using measures of harm
background (33.7%). occasionally bullied; Children. linear regression were collected
Children had an IQ and 14 were both models, whether prospectively and
within the normal maltreated and cortisol responses repeatedly in
range when they were frequently bullied were associated with childhood.
5 years of age (from (7.4%). social, emotional,
64 to 142; mean [SD] and behavioral
= 101.38 [14.32]). problems.
Brand et al 2010 The primary goal of A total of 126 mother Abuse history (CTQ): Noise/arm Salivary Cortisol: Mother 2 (Group: high vs low ANCOVAs revealed a F(1,111) .03021 No differences in
this study is to take / infant dyads Mothers completed stressor: The infant and infant salivary cortisol abuse) x 4 (time) statistically = 4.82 baseline cortisol were
this line of research qualified for the the CTQ, a validated was placed in a car measures were: 1) baseline ANCOVA. Hypothesis significant found for mothers with
one step further present study. The measure assessing seat behind an cortisol (T0 – study entry); testing utilized association between a history of child
and investigate the infant sample childhood exposure to occlusion screen, and 2) cortisol change analyses of maternal abuse abuse; however, these
association contained 62 boys and emotional, physical, and the mother calculated as the area under covariance (ANCOVA) history and mother women demonstrated
between maternal 64 girls, with a mean and sexual abuse. was permitted to the curve (AUC; linear with maternal cortisol change greater decreases in
history of child age of 187±17 days at Child trauma was view her infant on trapezoid method) for T1, baseline cortisol, (F(1,111)=4.82, eta cortisol in the context
abuse and maternal the time of the operationalized as a TV monitor T2, and T3 cortisol samples, maternal cortisol squared=.04, p=.03) of the infant stressor
cortisol levels in a laboratory study. mild or higher levels during a noise as measured from baseline change, infant but not maternal paradigm. Our finding
clinical sample of Most (94%) of the of sexual abuse or burst and an arm (T0). baseline cortisol, and cortisol at baseline that maternal cortisol
postpartum women, mother/infant dyads moderate or higher restraint stressor infant cortisol change (F(1,119)=1.37, eta decreases in response
and to explore were Caucasian, the levels of physical task. Saliva as dependent squared=.01, p=.24). to stress is consistent
whether depressive median maternal abuse. We excluded samples were variables.). The maternal post- with those seen in
symptoms and education level was emotional abuse in taken from the Covariates included stressor cortisol healthy women with a
stressful life events, college graduate, order to maintain mother and the demographic and concentrations were history of childhood
as well as comorbid mean maternal age consistency with the infant immediately clinical lower for mothers maltreatment
PTSD moderated was 34 ± 4 years, and previous ELS after the lab characteristics that with a history of (Carpenter et al.,
this relationship. 95% of the mothers literature. stressor tasks were were significantly childhood abuse 2007).
were married or completed (T2 - associated with the compared to mothers
cohabitating at the post-noise/arm dependent variables. with no such history.
stressor I), and All tests were two-
99
time of the infant again 20 minutes tailed, and an alpha

follow-up. later (T3 - post- of 0.05 was used
noise/arm stressor throughout.
II).
Cutuli et al 2010 We expected a This resulted in a Negative Lifetime Cognitive Tasks:. Salivary Cortisol: Children’s The relationships Those with the t(63) = .02795 A history of many
positive relationship sample of 66 children Events: Lifetime These tasks served salivary cortisol was between negative highest lifetime event -2.25 negative life
between lifetime (25 females). An effort Events Questionnaire as the challenge to collected in two contexts: 1) lifetime events, scores have higher experiences in this
event scores and was made to include completed by the elicit a cortisol during the assessment cumulative cortisol levels at the sample of young
cortisol response to only families who had primary caregiver. response. First, session as a measure of HPA socioeconomic risk, start of the session homeless children was
the session tasks, stayed in the shelter This questionnaire children completed reactivity to the tasks, and and cortisol response and, on average, linked to higher
given past evidence for at least three elicits the number of three subscales of 2) in the morning prior to to cognitive tasks decline over the cortisol levels in the
suggesting that nights to allow for stressful events that the Wechsler breakfast to assess diurnal over the session course of the session. morning, higher levels
family difficulties some time to adjust to the child experienced Preschool and cortisol levels following were tested using a In contrast, those at the start of an
that are more the context; yet two during his or her Primary Scale of wake up. During the linear mixed model with the lowest assessment session,
common in low (3%) families lifetime, including Intelligence – Third assessment session, three (LMM), controlling number of lifetime and falling levels across
income contexts participated prior to events and situations Edition. Then a set samples were taken at for age, sex, and time events not only have a session of challenging
(e.g., child their third night in that threaten the child of widely used roughly 30 minute intervals of day the session lower intercept cognitive tasks.
maltreatment, shelter. Nearly all directly as well as tasks were following a 15 minute occurred. Again, values at the start of
exposure to participants had those experienced by employed to assess consent process and an additional models the session, but also
domestic violence) ethnic minority the child that threaten executive function initial 5 minute rapport were run to inform tend to show either
are linked with backgrounds [n = 64; a parent, a family skills, including building period. the impact of lifetime no response or a
differences in the 97%: 53 (80.3%) member, or the Simon Says, Peg events without rising pattern of
stress response to African American; 10 integrity of the family Tapping, the socioeconomic risk cortisol response
psychosocial (15.2%) multiracial; (see Table 2). *Note: Computerized considered, and of over the course of
challenge. and 1 (1.5%) Native items included Pointing Stroop socioeconomic risk the session (See
American]. Two traumatic events task, and the without lifetime Table 6 Fixed Effects,
(3.0%) children were Dimensional events. Slope Estimates).
Caucasian. Change Card Sort.
Gunnar et al 2009 Early life stress (ELS) The participants were Adoption from TSST-C: The TSST- Salivary Cortisol: Eight Covariates with Because the lowest 3 chi2(1) .002 Diminished cortisol
is expected to 124 children ages orphanages or other C was then samples were taken in the missing data were and highest 2 = 9.53 activity was noted for
increase reactivity 10.02–12.21 years institutions: The types administered laboratory. Three were used, so the sample trajectories appeared the EA/FC group
of the Children in the LA/PI of adversity to which (Buske-Kirschbaum collected over the initial size varied slightly by to differ in cortisol (moderate ELS), while
hypothalamic– group (23 girls, 19 these children were et al., 1997). This baseline period [after analysis. N's are intercept, we the LA/PI group (severe
pituitary– boys) were adopted at exposed likely varied, task involved consent/assent (0 min), reported for each examined the ELS) did not differ from
adrenocortical 12 months or older but nearly all providing the child after electrode placement analysis. The percentages of the NA group. These
(HPA) axis; (range 12–64 months) institutions included with a story (+15 min) and after statistical tests we children who were results suggest that
however, several having spent at least the lack of a scenario and then completion of baseline used enable analyses assigned to these moderate ELS is
recent studies have 75% of their pre- consistent, giving him/her 6 autonomic assessment (+30 with unbalanced levels. Of the EA/FC associated with
shown diminished adoption lives in responsive, adult min to develop a min)]. Five were collected designs. In all children, 85% were diminished cortisol
cortisol reactivity institutional care. caregiver. Exposure to story (Speech during the stress and analyses, covariates on the lowest three activity; however,
among adults and Children in the EA/FC pathogens is also Preparation recovery periods were examined and, trajectories, marked individual
children with ELS group (20 girls, 24 greater in group care Period), followed [immediately after the if non-significant, compared to 53% of differences in cortisol
exposure. The goal boys) had been settings such as by 5 min of mental speech preparation (+40 were removed from the NA children, χ2(1) activity among the
of this study was to adopted before 8 orphanages. While arithmetic min), after the speech/math the models. Thus, = 9.53, p < .001. LA/PI children suggest
examine cortisol months (range 0.5–8 many institutions (Speech/Math period (+55 min), and then only significant that child factors
activity in 10–12- months) and had make adequate Period). at 10 min intervals until +75 covariates appear in modify the impact of
year-old spent less than 2 nutrition available, min]. the results. severe ELS.
internationally months in institutional problems in delivery
adopted children to care. Children in the (propped bottles) and
100
determine if NA group (17 girls, 21 health issues

moderate and boys) were raised (parasites) may
severe ELS have continuously by their reduce the nutrition
different impacts on birth parents in the the child actually
the HPA axis. United States. receives.
MacMillan et al The purpose of the Female youths Childhood TSST: The stress Salivary Cortisol: Over the To examine group Differences in slopes t(90) = .0007 We found a blunted
2009 present study was between the ages of maltreatment was protocol for course of the first visit, both differences between the -3.51 cortisol response (but
to examine whether 12 and 16 from three assessed using the children and youth HR and saliva samples were independent t tests maltreated and no difference in heart
resting levels of local child protection Childhood has a brief stress recorded at −45, −25, −5, for continuous control group are rate response) to a
cortisol, heart rate, agencies (CPA) Experiences of anticipation +20, +40, and +60 minutes, variables and chi- evident during the psychosocial stressor
and pattern of qualified for eligibility Violence period, a public with the TSST at 0 minutes. square analyses for reactivity phase. among maltreated
stress response to a review. We recruited Questionnaire (CEVQ), speaking task (5 During the preliminary categorical variables Youth in the control female youth even
psychosocial 67 female youths and the Childhood minutes), and phase of the study, a were conducted. group show an after controlling for
stressor differed aged 12 to 16 with no Trauma Questionnaire mental arithmetic baseline salivary sample at Piecewise multilevel increase in levels of SES and Tanner
between youth prior history of (CTQ) (5 minutes) in front −25 minutes was obtained growth curve cortisol following the pubertal staging, as
exposed to depression from child of a panel of two (n = 16 of 67 participants), modeling examined TSST and a gradual well as current
maltreatment and a protection agencies judges. Peak but initial analyses showed differences between flattening over time, symptoms of MDD and
control group. and a control group of cortisol that participants had maltreated and while maltreated PTSD.
25 youths matched on concentrations are elevated cortisol in this first control youth in youth do not show
age and postal code. reliably found 10 sample, so a sample was resting and post-TSST an increase in levels
minutes after added at −45 minutes to levels of cortisol and of cortisol, nor do
stress termination have two possible true HR. they have a gradual
in this protocol. resting (basal) samples flattening of
before the stress test. response (Table 3).
Pierrehumbert et al We expected Two groups of adult Childhood Sexual TSST: Consists of a Salivary Cortisol: Saliva 3 (Group: Abused We conducted a t(34) = .03961 The group of abused
2009 abused women (i.e. female subjects were Abuse/ The Early public speaking samples were collected for Organized, Abused regression analysis to 2.14 subjects with the
with CSA) to present included in the study: Trauma Inventory: task and mental the measurement of the Unresolved and examine the possible Unresolved
more a clinical group Childhood sexual arithmetic stress hormone at 15, 44, Controls) x 8 time) contribution of classification of trauma
psychopathological (women with an abuse such as performed in front 60, 65 min (i.e. before the ANOVA and a psychopathology in expressed the highest
troubles, to be experience of CSA) provocative touching, of an audience. stress procedure), 75, 85, separate regression the association perceived stress, they
more often and a control having been forced to This standardized 100, and 115 min (i.e. after analysis. between cortisol also showed the most
categorized as group. he inclusion touch another procedure was the stress procedure). responses and the suppressed endocrine
Unresolved at the criteria were to be person's intimate developed so that unresolved reactions, when
AAI, and to show over 18 years old (civil parts, having been to induce a classification. Only compared to control
more altered majority) and to touched in intimate moderate the variable subjects.
cortisol stress report at least one parts, having psychosocial stress Unresolved remained
reactions than episode of CSA (such experienced under laboratory in the model
controls. as provocative attempted or conditions. (Beta = 0.35; t = 2.14,
touching, having been completed genital p = 0.04).
forced to touch rape) during
another person's childhood and/or
intimate parts, having adolescence (up to
been touched in the age of 18 years).
intimate parts, having The Early Trauma
experienced Inventory (ETI) is a
attempted or structured interview
completed genital concerning traumatic
rape) during experiences in
101
childhood and/or childhood and/or

adolescence (up to adolescence.
the age of 18 years).
Elzinga et al 2008 The purpose of the Eighty college Traumatic TSST: This Salivary Cortisol: Salivary 2 (low adversity Using ANOVA rm F(4, .01039 In the present study,
present study was students participated Experiences Checklist psychosocial stress cortisol was measured to group vs. high with Group (high 296) we found that non-
to investigate for financial or course (TEC): The TEC is a test, which mainly investigate the effects of adversity group) x 5 versus low AE group) =3.36; clinical individuals that
cortisol reactivity to credit (52 men and 28 reliable and valid self- consists of a free exposure to early adverse (time) repeated as between subjects Robust- have been exposed to
a psychosocial women, age: mean +/- report inventory that speech and mental events on the reactivity of measures ANOVA, factor and Gender as ness a relatively high
stress task (TSST) in S.D.: 21.6 +/- 3.61 assesses emotional arithmetic task of the HPA axis. All gender was included covariate and (log F(1,78) number of adverse
a sample of healthy years). Participants abuse and neglect, 15min duration has physiological assessments as a covariate. transformed) cortisol = 6.42 events had a
men and women between 18 and 37 physical abuse, sexual repeatedly been and the subjective measure as dependent significantly lower
who had years old were harassment, and found to induce of distress were obtained at variable, a significant cortisol reactivity to a
experienced either recruited from the sexual abuse, as well significant five assessment points over Group x Time psychosocial stress task
very few or University of Leiden as general traumatic endocrine and a 75-min period, at interaction was compared to
relatively many through events. cardiovascular approximately -20, -1, +15, found (F(4,296) = individuals who had
adverse life events. announcements. responses. +40, +55 min with reference 3.36, p < 0.01). experienced none or
to the start of the stressor. one adverse event,
while the two groups
did not differ at
baseline cortisol levels.
102
Matheson et al The present study Participants (N = 90) Traumatic Life Events Reminders of Salivary Cortisol: To 3 (Group: no trauma, Contrary to F(3,123) .00001 Among those
2008 evaluated the were recruited from Questionnaire (TLEQ): stressful evaluate salivary cortisol single trauma, expectation, as seen = 10.27; individuals that had
trauma experiences the Somali community A modified version of experiences: levels in response to multiple traumas) x 4 in Table 4, the Robust- experienced multiple
of Somali refugees (Ottawa, Canada) over the Traumatic Life Participants also reminders of their stressor (time) ANOVA number of traumatic ness traumatic events,
to Canada and their the course of 2003, Events Questionnaire provided saliva experiences, participants events encountered r(82) = - salivary cortisol levels
capacity to cope through bulletins (TLEQ) was samples to assess completed the survey during was negatively 0.22 declined over the
effectively with posted at community administered to cortisol reactivity the l at afternoon (i.e., when related to cortisol course of the
acculturation centers and the assess exposure to a to reminders of basal cortisol levels are levels in response to questionnaire session.
stressors. It was university, in flyers, range of traumatic their stressor relatively low). The timing of reminder cues. In This was contrary to
expected that local and community events. These were experiences. the questionnaire was paced effect, as seen in Fig. expectation, but there
coping difficulties newspapers, and presented in to allow for the collection of 2, those who had not are several possible
would mediate the through referrals from behaviorally saliva at four time points, 15 experienced any processes that might
relations between social service workers. descriptive terms min apart, following (1) traumatic events, F < operate to promote
trauma experiences (consistent with the instructions and the 1, or only one event, such a decline. Trauma
and poorer health DSM-IV stressor completion of the informed F(3,66) = 1.27, ns, did symptoms and multiple
outcomes that may criterion A1). When consent and demographic not show a cortisol traumatic experiences
be evident years participants reported questions (baseline); (2) decline over the were associated with
after the traumatic experiencing an completion of the TLEQ course of the an exaggerated
events. event, they then (excluding familial assault), questionnaire morning cortisol rise,
indicated whether it IES-R, and BDI (along with session, whereas but with a blunted
had caused them to questions regarding cortisol levels response to stressor
experience intense acculturation hassles and dropped steadily reminder cues.
fear, helplessness, or adaptation that were not throughout the
horror (PTSD stressor analyzed in the present session among those
criterion A2 in the study) (30 min); (3) the who had
DSM-IV), and when description of a stressful encountered multiple
the event occurred. event and how they coped traumatic events,
Whether participants (45 min); (4) self-reported F(3,123) = 10.27, P <
had encountered each physical health and trauma .01, ɳ2 = .200.
of five clusters of events involving familial
trauma types was assault (60 min).
then evaluated.
103
Table 2.3: P-curve Disclosure Table of Adversity-Cortisol Hyper-reactivity Literature

Original Quoted text from Sample Type of Adversity Stress Induction Cortisol Data Analysis Plan Quoted text from Results P- Conclusion quoted
Paper original paper Characteristics Task Measurement original paper value from original text
interest
Blankenship et It was hypothesized The present analyses Observed Parenting Stressor Task and Salivary Cortisol: At There were three See Table 2 - r(61) = .02624 Critically, a significant
al 2019 that greater early were performed on a Behavior: At both Time 1 TSST: The stressor Time 1 and Time 2, primary aims of the Bivariate correlations .28 indirect effect was
maladaptive subset of children and Time 2, parents and task required salivary cortisol present analyses: (a) of Time 2 negative observed, with greater
parenting behaviors (n = 63) from a their children completed children to match samples were to assess with time 2 cortisol. negative parenting
(low positive or high longitudinal study (N = parent–child interaction colored chips to an collected associations between during early childhood
negative) would 175) of high‐risk tasks modified from the animal as quickly as at baseline (after 30 early (Time 1) and predicting greater
predict increased offspring of Teaching Tasks Battery. possible while being min of quiet play concurrent (Time 2) cortisol reactivity three
cortisol reactivity depressed mothers Each task was coded on the evaluated. The preceding the parenting and cortisol years later, which, in
and reduced and a nondepressed following: maternal experimenter stressor), and reactivity on Time 2 turn, predicted
hippocampal sub‐ community hostility, maternal manipulated the 20‐, 30‐, 40‐, and 50‐ hippocampal reduced left
region volumes and comparison intrusiveness, maternal timer so that the min post‐stressor subregion volume in hippocampal tail
that the association group. support, maternal positive child failed to using a cotton roll separate models, (b) volume.
between parenting affect, and maternal complete the task dipped determine whether
and off‐spring negative affect. three times before in Kool‐Aid (~0.025 these associations
hippocampal the child was mg) and chewed were timing‐
volume would be informed that the until saturated (~ 1 dependent, and (c)
mediated by timer was broken min). evaluate whether
offspring cortisol and received their cortisol reactivity
reactivity prize. At Time 2, the mediated
stressor included a associations between
modified version of parenting and
the Trier Social hippocampal
Stress Task during subregion volume.
which children were
instructed to tell a
4.5 min story after
30 s of preparation
followed by an
unsolvable puzzle.
Doom et al As there are few A total of 242 children Negative Life Events — The TSST- C: The stress Salivary Cortisol: Three regression The overall model t(140) = .00675 Greater negative life
2018 data examining who participated in 30-item Life Events reactivity protocol The cortisol AUCi models were significantly 2.75 events during
cortisol reactivity in the middle childhood Checklist (LEC) was consisted of was calculated using calculated with predicted T2 cortisol Robust- preschool and greater
low-income children follow-up of a completed by the parent to approximately 10 the trapezoidal rule cortisol intercept, AUCi, F (11, 140) = ness concurrent family
in middle childhood, longitudinal study are report what negative min of academic and reflected the diurnal cortisol slope, 2.32, p = .01. Greater t(140) = conflict were
we tentatively included in analyses events happened to the testing with a strict child’s total free and cortisol AUCi T2 family conflict, t 2.17 associated with
predict that greater (see Table 1 for family in the past year. teacher and 10 min cortisol output (area under the curve (140) = 2.75, p < 0.01, increased free cortisol
current negative life participant Family Conflict —Conflict in of the modified Trier increase from the with respect to and greater negative reactivity in middle
events will be demographics). the family was assessed Social Stress Test for first sample (prior to increase) in response life events at T1, t childhood.
associated with using the 12-item conflict Children (TSST-C). starting the stressor) to stress as (140) = 2.17, p = .03,
heightened cortisol scale of the Self Report to the fourth sample dependent variables. were associated with
reactivity, while Family Inventory. Items of (40 min after The final models were an increased T2
greater preschool the conflict scale measure created by first cortisol AUCi.
104
family conflict will overt conflict, such as conclusion of including preschool

be associated with openly fighting, blaming, challenging tasks). and middle childhood
blunted cortisol arguing, negative tone, and negative life events,
reactivity. unresolved conflicts. chaos, family conflict
as independent
variables.
Busse et al We hypothesized The complete study Discrimination: Assessed TSST: The TSST Salivary Cortisol: Pearson Product Higher levels of r(141) = .0423 Latinos experience
2017 first, that being of sample consisted of with the Experiences of consists of a 5-min After a 15-min rest Moment Correlations discrimination were .17 more discrimination
Latino ethnicity and 158 individuals. Discrimination Scale (EOD). mock job interview period, a first saliva were conducted to associated with a throughout their
scoring high on Fifteen individuals The EOD contains 9 items and a 5- min mental sample (–2 min) was obtain an initial steeper maximum lifetime than non-
experiences of were excluded, four about past incidences of arithmetic task in collected, and impression of cortisol increase (r = Latinos, and that more
discrimination because they reported discrimination in specific front of two neutral, participants were associations between 0.17, p < 0.05) and a discrimination
would be being half Latino and settings, including finding non-supportive ex- escorted to an variables of interest. more pronounced experiences are
independently half non-Latino, and housing, a job, and medical pert evaluators of adjacent room to Univariate ANOVAs maximum cortisol associated with
associated with 11 because of missing care. Each item is rated diverse ethnicities complete the and basic regression recovery (r = −0.19, p heightened cortisol
blunted cortisol or insufficient cortisol based on the frequency of (e.g., Latino, laboratory stressor. models were then < 0.05) in the overall reactivity.
reactivity; and data. The remaining these occurrences over the European, East Upon completion of conducted to test the sample.
second, that the 143 participants (80 lifetime (0=never, 1=once, Asian, mixed the TSST, a second ethnicity,
association female, 63 male) were 2.5=two or three times, background) and saliva sample was discrimination, sex
between Latino between 18 and 29 5=four or more times). both sexes. Including collected (+1 min). variables and their
ethnicity and years old (M = 20.49, Individuals are also asked to the instruction and Participants then re- interactions as
blunted cortisol SD = 2.08). indicate what category preparation period, turned to the predictors of the
reactivity would be (e.g., race, sex) they the TSST lasted 15 waiting room where maximum cortisol
mediated by consider to be the major min. additional saliva increase and
experiences of reason for the experience samples were recovery.
discrimination. of discrimination. collected at +10, 20,
30, 45, 60 and 90
min. (nmol/L).
Fearon et al We explored the In total, 316 Cumulative Contextual TSST: Participants Salivary Cortisol: We modelled the As Figure 4 shows, t(143) = .03089 The results of our
2017 role of cumulative adolescents provided Adversity: Using data were told that they Before TSST cortisol response for females there was 2.18 study, though in need
contextual adversity cortisol samples, 15 of completed by the child’s would be given 3 procedure, profile using a stronger (linear) of replication, provide
on the stress whom were excluded carer and the child at the minutes to prepare a participants (who polynomial functions increase in cortisol further evidence that
response, testing due to asthma steroid age 13 assessment, we 5-minute speech on had not eaten or to describe the response for those early caregiving may
whether this might pump use, which can dichotomized the following anything about drunk anything in change in cortisol with high levels of indeed be implicated in
account for any interfere with cortisol indices, scored as present themselves. After the last hour) were over time, including adversity, compared HPA axis development
observed effects of measurements. A versus absent (zero versus the speech, one of first asked to provide terms for the to those with lower into adolescence and
attachment further 14 cases were one): overcrowding, the ‘examiners’ a saliva sample by intercept (the adversity (Linear B= in buffering the HPA
security or lost due to problems community violence administered a serial directly filling a 2 baseline level), linear, .053, 95% CI[.005, axis from the effects of
sensitivity, and with the labelling and exposure, house has no sevens subtraction milliliter plastic quadratic and cubic .100], p= .030; chronic and extreme
whether higher storage of the running water, house has task. This was sampling device trends. The effects Quadratic B= −.006, adversity.
quality of early care samples. Of these, no toilet, house has no maintained for 4 (SaliCap) or using a were then modelled 95% CI[−.013, .001],
might moderate the 232 had been electricity, the parent minutes without any short plastic straw to as a function of level p= .106).
impact of observed in the endorsed that members of intervention or do so. After the 2 variables (across
cumulative Strange Situation at the family had gone for responses from the TSST stressors individuals), such as
contextual adversity 12 months, 217 had whole day without eating ‘examiners’ research assistant attachment and
on HPA reactivity. completed because of a lack of food, irrespective of how entered the room parental sensitivity.
assessments of primary caregiver is well the participant and led the
general sensitivity and unemployed, caregiver has was performing. participant to the
212 completed the only primary-level next room where a
assessment of education, relationship second saliva sample
105
sensitivity during breakup with partner or was collected. After

feeding husband, partner has been the TSST, an
violent towards unrelated interview
mother/caregiver. was interrupted
every 10 minutes to
collect a saliva
sample.
McFayden- We examined The analyzed sample Economic Stress: Primary Challenge Task: The Salivary Cortisol: — All analyses were For Aim 2, to t(163) = .00573 Greater economic
Ketchum et al whether economic was 165 preschool- caregivers reported on their teacher and child Children’s cortisol conducted with path examine whether 2.8 stress was associated
2016 stress predicted age children ( M age = family’s level of economic were presented with was collected three analysis in Mplus in economic stress was with poorer child self-
physiological 4 years, 2 months, SD stress via the 54- item a standardized set of times across a order to take associated with regulation and
reactivity (cortisol = 7 months; n = 85 Multicultural Events materials for four challenge task advantage of full physiological heightened
levels) across a [52%] female) Schedule for Adolescents successive 5-min presented to the information reactivity across a physiological reactivity
challenge task in the enrolled full time in for Preschoolers (MESA). intervals beginning child and his or her maximum likelihood challenging task, we across a challenging
preschool high-quality child care The MESA used in the with two books as a primary teacher. (FIML) estimation—a tested the model classroom task for the
classroom, centers (as indexed by current study was modified warm-up, followed Samples were theory driven direct estimating the effects sample as a whole
hypothesizing that the Early Childhood from the original for use by novel and collected just before estimation approach of economic stress on
higher levels of Environment Rating with a younger age group interesting puppets, the 20-min challenge that allows for cortisol reactivity (as
economic stress Scale–Revised; M = by removing questions only followed by the pair (baseline), modeling with measured by AUCG),
would predict 5.52, SD = 0.58) in a appropriate for being approached by immediately after partially missing again controlling for
greater midwestern city and adolescents. We used the a robotic toy, and the 20-min challenge cases. All analyses child age and sex.
physiological recruited by a Economic Hassles subscale concluding with a (reactivity), and 20 involving cortisol as The model had
reactivity, as bilingual and in the current analyses as difficult-to open min after completion an outcome excellent fit, χ2(1,
measured by bicultural project an index of economic stress Plexiglas® box of the challenge controlled for cortisol N=165) = 0.04, ns ,
cortisol reactivity manager. rather than income, which containing an (recovery). The area collection times. CFI = 1, RMSEA = .00.
patterns. was confounded by group attractive toy. The under the curve with Economic stress was
and also is difficult to assess items were chosen respect to ground associated with
accurately with immigrant to mimic the types (AUCG) was greater reactivity (β =
samples because of of challenges calculated for the .24, CR = 2.89, p <
unstable income, income children face at child reactivity data. .01).
from multiple sources, and care as viewed from
shared income with family the child’s
not in the United States. perspective (e.g.,
academic content,
novelty,
unpredictability, and
frustration)
Mrug et al 2016 We hypothesized Participants were 84 Low Family Income: TSST-C: Adolescents Salivary Cortisol: Bivariate correlations Family income was r(82) = - .0056 Interestingly, family
that shorter sleep adolescents (M age Reported on a 13-point were given five Whole saliva among all variables associated with lower .30 SES also was related to
duration and more 13.36 years, SD = scale from 1(<$5000/year) minutes to prepare samples were were examined. AUCi (See Table 3 cortisol levels, with
sleep problems will 0.95; 50% male; 95% to 13 (>$90,000/year). the ending to a story collected by passive Parent and child correlation between higher pre-test cortisol
predict higher African American, 4% and then five drool immediately reports of wake and family income and levels among
cortisol levels and Caucasian and 1% minutes to tell that before the TSST bed times, sleep AUCi). adolescents with more
reactivity to stress Hispanic) who took story in front of two began (pre-test), 30 quantity, and sleep highly educated
(i.e., a social part in the Coping judges. Next, the min after the 15-min problems were parents, and lower
evaluative threat). with Violence Study. judges asked the test began (15 min compared with paired stress reactivity in
We also examined The sample was participants to post-test), and 70 samples t-tests, as those with higher
gender differences socioeconomically perform serial min after the test were cortisol levels family incomes.
in the effects of heterogeneous, but subtraction for 5 min began (55 min post- across the three time
sleep on stress- comprised primarily (subtracting 7 test). points. Total cortisol
106
related cortisol low-income families; starting at 758). secretion and its

regulation and average annual family However, if the increase during the
reactivity. *Note: income was $20,000- adolescent made studied period were
The authors did not $25,000 (range five mistakes in a also analyzed with
have any ˂$5000 to $70,000- row, they were given area under the curve
hypotheses $90,000) and average an easier task with respect to
regarding the parental education (subtracting 3 ground and increase
relationship was some college but starting at 307). (AUCg and AUCi).
between economic no degree.
stress and cortisol
reactivity.
Andreotti et al We predicted that Participants for this Family Conflict: All The Noisy Neighbor Salivary Cortisol: Descriptive statistics As shown in Table 2, r(114) = .00482 As hypothesized,
2015 chronic stress study include 116 participants completed the Task: Was used as The five data points were first calculated exposure to family .26 individuals in the
exposure during female undergraduate frequency (6 items) and an acute laboratory allow for analyses of for attentional bias, conflict during Robust- current study reporting
childhood would be students currently intensity (7 items) subscales stressor and as a both reactivity to cortisol, and development as ness a history of higher
related to reduced enrolled at a selective of the Children’s Perception measure of stress (as reflected in questionnaire measured by the r(100) = levels of family conflict
cortisol reactivity to university in the of Interparental Conflict physiological and increases from pre- variables for the CPIC scale was .28 tended to have
an acute laboratory southeastern U.S. The Scale. This scale is designed psychological stress to post-stress task) whole sample as well significantly increased AUC.
stressor. Further, mean age of the to measure various aspects reactivity. During the and total as by randomization positively correlated
we predicted that sample was 18.96 of conflict occurring in the task, each concentration over group. ANCOVA with cortisol AUC in
cortisol reactivity years (SD = 1.13), with home from a child’s participant engaged the assessment procedures were then the sample as a
would account for a range of 18 to 22 perspective. Each item is in a conversation period (as reflected used in this analysis whole (r = .26, p <
variability in years. The sample was rated on a 3-point scale role-play with a in area under the to test study .01) and in the group
attentional bias to 78.6% Caucasian, (true/sort of true/ false). In gender-matched curve of the five hypotheses for that performed the
threat observed 8.5% African order to quantify family research assistant in time points). differences in dot probe first (r =
after acute American, 6.8% Asian conflict exposure prior to which the attentional bias index .28, p < .05).
laboratory stress American, 4.3% 16 years of age in the participant was scores between
exposure. Hispanic/Latina, and current study, the items prompted to present groups (i.e.,
0.9% endorsed more from the Frequency (6 an argument to randomization and
than one ethnicity. items) and Intensity (7 convince a neighbor conflict exposure),
items) subscales of the CPIC to lower the volume accounting for
were summed to create a of her music. potentially
CPIC total score. Participants were cofounding factors
told they would be (i.e., DSM Depression
evaluated on the and Anxiety scores).
persuasiveness of
their argument and
were videotaped
during the
conversation.
Hagan et al The current study One hundred six Childhood Exposure to Role-Play Task: Salivary Cortisol: For the evaluation of Whereas IPC Chi2(1) = .04886 We hypothesized that
2014 tested a curvilinear young adults were Interparental Conflict (IPC): Participants role- After a 20-min reactivity, raw frequency exhibited a 3.88 perceived exposure to
relation between invited to a lab Young adults’ perceptions played a situation in resting period, the cortisol values were curvilinear relation Robust- childhood IPC would be
childhood exposure session 1–3 months of childhood (prior to age which they were first saliva sample log-transformed, and with cortisol ness related to cortisol
to IPC and the after completing the 16) exposure to IPC were trying to study for an (T1) was collected. residualized change reactivity, B=.02, Chi2(1) = reactivity in young
cortisol stress screening survey, at measured with the important exam but Participants then scores were SE=.01, x2=3.88, 8.65 adulthood in a
response in young which time they again Children’s Perception of their neighbor was completed a 10-min computed by p<.049, R2=.04 curvilinear manner. In
adulthood. We completed the FES-C Interparental Conflict Scale playing his/her roleplay task. regressing the 20min similar to the pattern line with this
hypothesized that in reference to their (CPIC). Scores on the six- music too loud. The Immediately after post-task (T3) cortisol observed with AUCg hypothesis, we found
higher and lower family relationships item Frequency (e.g., “I participant was the task, participants sample on the (see Fig. 1), IPC that young adults who
107
self-reported prior to 16. never saw my parents instructed to ask provided a second baseline cortisol intensity did not reported lower and
childhood exposure Demographic argue or disagreeing.” his/her “neighbor” sample (T2). Twenty sample (T1) and exhibit a linear higher exposure to
to IPC, relative to characteristics of the a=.92) and seven-item to turn down the minutes after the identified covariates. (p=.84) or curvilinear childhood IPC exhibited
the mean, would be final sample (n=91) Intensity (e.g., “My parents music. The neighbor end of the task, Because the cortisol (p=.48) association greater cortisol output
associated with a are displayed in Table got really mad when they was played by a participants gave a response peaks with reactivity. during a stressful
greater cortisol 1. argued.” a=.91) subscales gender-matched third sample (T3). approximately 20min interpersonal task
response to and were summed, with higher undergraduate Participants then after exposure to compared to youth
across a stressful scores reflecting higher research assistant responded to all self- stress, the T3 sample who reported
interpersonal task exposure to IPC. who maintained a report was used to best moderate levels of IPC.
neutral facial questionnaires. The capture reactivity
expression and fourth sample (T4) (change from
followed a scripted was collected 40min baseline).
series of responses after the end of the
indicating refusal to task.
cooperate.
Aschbacher et 1) anticipatory This study was Caregivers: Twenty-five TSST: The Trier Salivary Cortisol: All hypotheses were CG had significantly t(46) = .00287 A consistent pattern
al 2013 cortisol reactivity conducted as part of a participants were caregivers Social Stress Task Salivary samples tested at the same greater anticipatory 3.15 emerged whereby
will mediate larger study of of a relative with dementia (TSST) is a gold- were collected via time using a (15-min) but not Robust- chronically stressed
associations caregiving stress and and twenty-three were age- standard laboratory the passive drool bootstrapped peak (30-min) ness caregivers who
between perceived biological aging.. A matched controls with stress task used to method in 2 mL moderated mediation cortisol reactivity r(25) = reported greater
stress and oxidative subset of 48 healthy spouses. Caregivers evoke stress-arousal SaliCaps tubes (IBL model with the aid of than NC (p's=.003 .565 perceived stress,
damage markers, participants from the had provided care for 4.7 responses. In this Hamburg, Germany) the “Process” macro and .16 respectively; negative affect and
and 2) the indirect larger study years on average (range: 8 study, because many at multiple time- for SPSS. All variables Figure 2), even when threat had significantly
path from perceived participated in an months-11.42 years). participants were points during the were normalized controlling for greater cortisol
stress to oxidative acute laboratory Perceived Stress Scale retired, they did not acute stress task, prior to model entry demographic, health increases while
damage via cortisol stress task and had (PSS): The Perceived Stress do a mock job which permitted to facilitate centering behavior and medical anticipating a stressful
reactivity will be data available for Scale-10 (Cohen et al., interview, but quantification of and interpretation of factors, as reported event.
significant among analysis in the current 1983) is a standard 10-item instead discussed anticipatory and coefficients. in a previous analysis.
chronically stressed study. Twenty-five questionnaire that assesses their strengths and peak cortisol Regression analyses As hypothesized,
caregivers, but not participants were subjective perceptions of weaknesses and reactivity: 0 were used to identify among chronically
among age- caregivers of a relative stress over the previous completed a math (baseline), 15 (after coefficients for stressed CGs, higher
matched low-stress with dementia and month. Items reflect task in front of an 5 min instructions, 5 individual paths perceived stress was
controls. twenty-three were uncontrollability, evaluative audience. min preparation, 5 significantly
age-matched controls unpredictability and feeling All participants min speech) and 30 associated with
with healthy spouses. that demands outweigh completed the TSST (to capture peak greater anticipatory
Caregivers had coping resources. Response between 1400h and cortisol). and peak cortisol
provided care for 4.7 options form a 5-point 1500h to control for reactivity (p's ≤.01).
years on average Likert scale ranging from diurnal rhythm
(range: 8 months- 0=never to 4=very often. variation.
11.42 years).
Koss et al 2013 Consistent with the Data for the current Child-Rearing Simulated Phone Salivary Cortisol: A growth mixture Higher levels of r(193) = .00321 The rising pattern,
distressing nature of study were drawn Disagreements: Mothers Anger Task: Children Children provided modeling approach children’s perceived .21 reflecting the
conflict, higher from participants completed an 8-item and mothers were baseline cortisol was utilized to model threat were sensitization
levels of children’s taking part in a larger abbreviated version of the asked to engage in a levels collected after different patterns of associated with a hypothesis, was
perceived threat longitudinal project Child-Rearing play activity when arrival in the cortisol reactivity. higher probability of associated with
and child-related examining the effects Disagreements scale the mother received laboratory setting. being in the rising histories of child-
conflict were of family functioning assessing the frequency of a phone call believed Cortisol levels were cortisol group related conflict,
expected to be on child adjustment. marital conflict about child- to be from the assessed prior to the compared to the perceived threat, and
related to an The original sample rearing issues; Perceived child’s father. simulated phone falling group. children’s emotional
consisted of 235 (129 Threat: Children completed Mothers were anger task to assess Similarly, more insecurity.
108
elevated pattern of girls, 106 boys) the six-item perceived provided a scripted a baseline indication frequent child-
cortisol reactivity. families including, threat subscale of the interaction by an of physiological related conflict was
mother, father, and Children’s Perception of experimenter to functioning. Two associated with a
their kindergarten- Interparental Conflict scale stage a marital saliva samples were higher probability of
aged child; 193 for younger children dispute over the collected after the being in the rising
families were included assessing the degree to phone. simulated phone group compared to
in the current study. which children feel anger task; the first both the falling group
threatened by marital was collected and the flat group.
conflict. Sample items approximately 25 (See Table 1
included “When my parents minutes after the Correlation between
argue, I’m afraid that peak of the Child-Related Conflict
something bad will simulated phone and Post Conflict
happen”. anger task. The Cortisol).
second poststressor
sample was
collected 10 minutes
later.
Wirtz et al 2013 Hypothesis 1: The study sample Role Uncertainty: Role TSST: Participants Salivary Cortisol: For Following previous As depicted in Table F(2.5, .01569 Our study investigated
Cortisol responsivity comprised 43 men uncertainty was measured were exposed to the determination of methods we assessed 3, general linear 89.4) = for the first time
to the Trier Social (mean age +/- SEM: by three Likert-scaled items Trier Social Stress salivary cortisol associations between models with 3.94 whether role
Stress Test will be 44.5 +/-2.0 years, of the scale “uncertainty” Test (TSST), which levels, saliva samples uncertainty and the repeated cortisol Robust- uncertainty at work as
stronger for mean body mass from the Instrument for combines a 5-min were taken cortisol/blood measures and role ness a background stressor
participants high in index +/- SEM: 25.7 Stress-oriented Task preparation phase immediately before pressure stress uncertainty as F(3.9, related to altered
role uncertainty +/- .4) who completed Analysis (ISTA), a well- followed by a 5-min as well as every 10 response by continuous 105.7) = physiological reactivity
compared with the measure of role established instrument in mock job-interview, minutes, at 0, 10, 20, calculating general independent 6.57 to acute psychosocial
participants low in uncertainty (see German-speaking countries. and 5-min mental 30, 40, 50, and 60 linear models with variable, and age, stress in working men.
role uncertainty. below). Our The uncertainty scale arithmetic in front of minutes after repeated measures of BMI, and MAP as The main finding of our
participants reported assesses role ambiguity and a jury that displayed completion of the cortisol/blood covariates, revealed study was that higher
a great variety of role conflict: (a) “How often a stern attitude. The TSST. pressure as that higher role role uncertainty was
different jobs to you get unclear TSST has been used dependent variables uncertainty scores associated with higher
including technicians, instructions?”; (b) “How frequently and it and role uncertainty were associated with cortisol stress
engineers, teachers, often do you get conflicting reliably induces as continuous higher cortisol stress reactivity even when
or craftsmen, and instructions from different increases in independent variable reactivity (interaction controlling for a broad
most (61.4%) worked supervisors?”; (c) “From biochemical while controlling for uncertainty-by-TSST: set of potential
for a private how many people do you parameters, most age, BMI, and MAP as F(2.5, 89.4) = 3.94, p confounders including
employer. receive instructions on a notably cortisol. covariates in the main = .016, partial eta2 = cardiovascular risk
regular basis?”). Items had analysis. .10, f = .33). Further factors and potentially
a 5-point Likert format, controlling for the confounding
ranging from 1 (very second set of psychological
seldom/ never) to 5 (very confounders constructs.
often), with higher scores (overcommitment,
reflecting higher stress appraisal,
uncertainty. perfectionism, time
pressure) did not
change the results;
(interaction
uncertainty-by TSST:
F(3.9, 105.7) = 6.57, p
˂ .001, partial eta2 =
.20, f = .49).
109
Hackman et al To date, studies of Analyses included a Neighborhood TSST-G: Participants Salivary Cortisol: Piecewise hierarchical With male coded as r(77) = .00298 In this analysis we
2012 cortisol reactivity total of 77 Disadvantage: We completed the Trier Salivary cortisol was linear modeling 0, and coefficients for .33 found that
and SES have largely participants (37 employed a measure of Social Stress Test for the primary (HLM) was the disadvantage thus concentrated
omitted female, 48.1%) concentrated neighborhood Groups (TSST-G) a outcome of interest, primary analysis reflecting the effect neighborhood
investigation of between the ages of disadvantage based on the modified version of and saliva samples strategy 1. Piecewise for males, higher disadvantage, but not
neighborhood 13and18(M 2000 United States (U.S.) the TSST that were collected at HLM is a strategy concentrated parental education,
components of =16.4,SD=1.2). census tract for the induces a moderate nine different times: based on longitudinal disadvantage was associated with
disadvantage. This participant’s home address level of stress. In at baseline, after growth modeling that predicted increased cortisol reactivity and
study is uniquely when the stress reactivity order to increase speech preparation, allows distinct cortisol reactivity recovery. However,
positioned to protocol was completed. social evaluation at the end of the modeling of the (B=0.006, this relationship was
determine if The 77 participants lived in participants stressor tasks, and different phases of p=0.004,reffect moderated by gender,
parental education 59 census tracts. Six underwent the 10, 20, 30, 45, 60, change over time, =0.33) and a steeper such that higher
and neighborhood measures were employed: protocol in groups of and 75 min after the permits the separate decline during concentrated
disadvantage, as percentage of individuals 2(n=27, 35.1%) or stressor. modeling of reactivity recovery (B=− 0.003, disadvantage was
moderated by below the poverty line, 3(n=50, 64.9%), and recovery phases p=0.003,r effect associated with higher
gender, are unemployed, and receiving matched for gender. following =0.34). cortisol reactivity and
associated with public assistance, as well as administration of a steeper recovery in
cortisol reactivity to the density of African– stressor, and offers boys alone.
stress. Americans, children under advantages over the
the age of 18, and female- use of ANOVA or
headed households. change scores.
Ouellet-Morin The main goals of Participants were Familial Adversity: Familial Exposure to Salivary Cortisol: Differences in mean Adverse Chi2(2) = .00723 Familial adversity was
et al 2008 the present study families of twins from adversity was assessed to unfamiliar Saliva was collected cortisol values environments have 9.86 associated with higher
were to analyze the the Quebec Newborn reflect the putative situations: Each twin before and after the according to zygosity been associated with Robust- cortisol reactivity to
genetic and Twin Study recruited cumulative adverse effects was exposed to 2 unfamiliar situations and sex were both lower ness stress.
environmental between April 1, of environmental risk unfamiliar situations and stored at −80°C investigated through and higher cortisol Chi2(2) =
contributions to 1995, and December factors on children’s known to be until analysis. The analyses of variance. reactivity. We 11.01
cortisol reactivity to 31, 1998, in the adjustment. In the present moderately stressful posttest sample was The association partitioned the
an unfamiliar greater Montreal study, an FA index was at that age. In the obtained 20 minutes between FA and sample
context in 19- area. A total of 989 created by combining first situation, 1 twin after the end of the cortisol reactivity was into quartiles of
month-old twins families were information about 7 risk and the mother procedure to tested using a 2-tailed cortisol reactivity,
and to examine contacted, factors: maternal smoking were alone when a capture the peak χ2 test, and allowing for different
variations in genetic of which 672 agreed during pregnancy, low birth woman dressed as a cortisol response. multinomial logistic associations to
and environmental to participate (68.0%). weight, low family income, clown entered the regressions were emerge at the lower
contributions to low maternal educational room and invited the performed on the (first quartile) and
cortisol reactivity level, single parenthood, child to approach by discrete patterns of higher
according to early young motherhood, and offering a set of cortisol reactivity. All (fourth quartile) ends
FA. maternal hostile or reactive familiar toys. In the statistical tests, of the distribution.
behaviors. second situation, a except for the genetic Figure 1 presents the
noisy, odd-looking, modeling, were distribution in
moving toy robot conducted on a quartiles of cortisol
was placed on a subsample composed reactivity for high
platform in the of only 1 twin, and low FA (χ2
opposite corner of selected at random, 2,211=9.86;P=.007).
the room. per family.
110
Zwolinksi et al The ultimate goal of A total of 33 female Relational Victimization: The Yale Salivary Cortisol: 2(relational To examine changes F(1, 23) .04901 This study found some
2008 this study was to undergraduate The Revised Self-Report of Interpersonal Stress reactivity was victimization group: in cortisol over two = 4.32 support for the tend-
examine whether students were Aggression and Social Stressor Task: Used measured using high, low) by 2 time points, a 2 befriend model
stress reactivity was recruited to Behavior Measure was used to induce the salivary cortisol. Two (menstrual phase: (relational as indicated by
heightened participate in the to assess history of experience of social salivary samples (1.0 follicular, luteal) by 2 victimization group: biological reactivity as
following social current study. Initial relational victimization by exclusion. A number ml each) were taken (time: baseline, post- high, low) by 2 well as distinct
exclusion for recruitment was peers at the current time of tactics were used from each stressor) repeated (menstrual cycle psychological
relationally based on two factors. and in the previous year to exclude the participant using a measures ANCOVA, phase: follicular, responses to social
victimized females Three participants did (e.g., ‘‘A friend of mine has participant. screw tap container controlling for state luteal) by 2 (time: exclusion in women.
in their luteal phase. not meet all of the gone ‘behind my back’ and Exclusion began for the collection of anxiety was baseline, post- This biobehavioral
Because of their eligibility criteria shared private information slowly and gradually sputum: once at the conducted. stressor) response was
interpersonal noted above, resulting about me with other included verbal (e.g., end of the baseline repeated measures influenced by history of
sensitivity, in a participant pool people’’). Participants rated disagreeing, adjustment period ANCOVA, controlling relational victimization;
relationally of 30 females. each item on a seven-point ignoring, criticizing and another for state anxiety, was low victimization
victimized women scale, ‘‘not at all true’’ to participant’s following the second conducted which tended to be
are expected to be ‘‘very true’’. Using the same comments) and non- stressor session. revealed a significant associated with unique
at greatest risk for cutoff score in prior verbal tactics (e.g., three-way psychological response
cortisol reactivity to research to determine confederates shift interaction(1, 23) = whereas high
social evaluative victimization group, their bodies and 4.32,p<.05, partial η2 victimization predicted
stressors, namely individuals who received an chairs = . 16, power = .51. neuroendocrine
peer exclusion. average score or above away from the To investigate the reactivity to peer
(e.g., > 16) on the four-item participants and lack rejection.
Relational Victimization of eye contact or
subscale were considered smiling).
to be targets of relational
aggression.
Ellis et al 2005 High stress A sample of 122 Family Stress: Family stress Challenge Tasks: An Salivary Cortisol: In the present High anticipatory Chi2(1) = .0164 In Study 2, a relatively
reactivity in the children and their scores were computed unfamiliar female There were four analyses, scores adrenocortical 5.76 high proportion of
context of acute, parents and teachers separately for mothers and researcher collection times: one reflecting children’s activation to the in- children in very
severe childhood was recruited from fathers based on their administered the immediately social environments, home laboratory stressful environments
stressors may one of two cohorts self-reports of five domains: challenges utilized in following the arrival along with their procedures was showed evidence of
function to increase (total N=500) in the (a) depression symptoms, the reactivity of the study team, a respective ranges of associated heightened
the overall longitudinal (b) family expressed anger, protocol. They second and third values and interval with mother- sympathetic and
readiness and WSFW. The resulting (c) parenting stress,(d) role included (a) a prior to and cut points, reported composite adrenocortical
capacity of sample included 73 overload, and (e) financial structured child following the were entered into the family stress in the reactivity.
individuals to deal girls and 49 boys from stress. Total stress: To interview modified reactivity protocol, signal detection infancy period
with the very real middle class families, create an overall measure from the Gesell and a fourth at the algorithm. The (N=115; x2 =5.76;
dangers in their representing 109 of stress in early childhood, School Readiness termination of the algorithm then p<.05). Half of the
environment (even White, 4 African the four composite family Screening Test, (b) assessment selected the variable children from
as it also results in American, 4 Asian, 2 stress measures and the digit span recitation procedures. and cut point that families in which the
chronic overarousal Hispanic, and 3 other SES measure, which were modified from the maximized the mother reported very
and associated race or biracial family already standardized, were Kaufman efficiency with which high family stress in
sequelae). We thus units. Of the 122 combined in a unit- Assessment Battery children at risk for infancy showed
anticipated that families who were weighted composite for Children, (c) 2–3 high reactivity evidence of
children who identified and agreed (M=0.03, SD=0.65). drops of lemon juice profiles were heightened
experience very to participate, 120 placed on the child’s identified. hypothalamic–
stressful early home completed all tongue, and (d) two pituitary–
and or school components of the emotion-evoking adrenocortical (HPA)
environments study. videotapes that responses, while only
23% of children from
111
would tend to were chosen to elicit lower stress families

develop high BSC. fear and sadness. (mean level of Total
Stress 5= -0.07) were
similarly responsive.
112
Table 2.4: P-curve Disclosure Table of Adversity-Cortisol Hypo-reactivity Literature

Original Quoted text from Sample Type of Adversity Stress Cortisol Measurement Data Analysis Plan Quoted text from Results P- Conclusion quoted
Paper original paper Characteristics Induction original paper value from original text
indicating Task with statistical
interest
Mezuk et al Individuals residing In total, 125 SES: Neighborhood SES was TSST: Stress Salivary Cortisol: Eight We conducted a Living in a lower SES t(785) = - 0.0028 In sum, our findings
2020 in lower-SES participants measured using a reactivity was salivary cortisol (SC) series of cross- neighborhood was 3.0 are broadly
neighborhoods will completed the composite of the following assessed using the samples (two prestress sectional analyses of associated with 16% consistent with prior
have blunted stress baseline assessment: five census tract variables: TSST, one of the and six poststress) were variance (ANOVAs) to higher baseline (SC1) work showing that
reactivity relative to 40 high-SES NHWs, 33 per capita income, percent most robust and taken to characterize explore how race and cortisol (p = .05), social disadvantage is
those living in high-SES AAs, 18 low- owner-occupied housing well-validated the HPA-axis response. neighborhood SES 8.4% shallower in- associated with HPA-
higher SES. SES NHWs, and 27 units, median home value, paradigms for relate to various crease in response to axis alterations, here
low-SES AAs. In percent not vacant housing evoking an HPA- features of HPA-axis the TSST (p = .004), indicated by both
addition, three high- units, and percent axis response. reactivity. We used and 1.0% steeper higher initial cortisol
SES and four low-SES households not in poverty multilevel modeling decline post-TSST (p levels and blunted
persons who (<$25,000). Individuals in (MLM) to estimate = .005) than living in TSST reactivity.
identified as Asian or the lowest and highest the association be- the higher
Hispanic tertiles for all five SES tween race, neighborhood SES
race/ethnicity variables were considered neighborhood SES, (Table 3, model 4).
participated. to reside in “low-” and and their interaction
“high-” SES neighborhoods, on HPA reactivity.
respectively.
Chiang et al We hypothesized Participants were 18 Major Life Events: TSST: Trier Social Salivary Cortisol: Six We first tested the Major Life Events: t(89) = 0.0035 We found that
2019 that all three types to 20-year-old Participants completed a Stress Task (TSST), saliva samples were main effects of stress Adolescents who -3.0 experiences of
of stress would be adolescents (n = 91; major life event checklist. a well-established collected using oral and psychological experienced more various stressors,
associated with sub- Mage = 18.37, SD = Example items included acute laboratory- swabs (Salimetrics) resources on cortisol major life events including major life
optimal biological .51; 57% female) parents divorced or based social throughout the and IL-6 response exhibited smaller events, daily
profiles, as drawn from a larger separated, family member stressor that laboratory visit. They trajectories by cortisol responses interpersonal stress,
evidenced by sample of 248 became seriously ill, a close activates the HPA were collected examining the cross- (linear: b(SE) = and early adversity,
dampened cortisol adolescents (Mage = friend moved quite far axis and increases immediately after the level interactions .07(.06), p = .195; were associated with
and heightened IL-6 18.31, SD = .77) away, you had a serious inflammation. neutral-content video between stress and quadratic: b(SE) = - smaller cortisol
responses to acute partaking in the falling out or ended a (baseline) and TSST, and time and between .03(.01), p = .006) responses to an acute
stress. second wave of a friendship with a close 30, 45, 60, and 75 psychological relative to those who laboratory-based
three-wave friend, you were suspended minutes after TSST resources and time. experienced fewer stressor.
longitudinal study on or expelled in school, and onset. Cortisol values major life events
the psychosocial your grades in school went were then natural log- (linear: b(SE) =
contributions to the down a lot. transformed to correct .21(.05), p < .001;
development of early for their skewed quadratic: b(SE) =
health risk from mid- distributions. −.05 (.01), p < .001).
adolescence to early
adulthood.
113
Raffington et al We first A total of 147 children SES: Parents self–reported TSST-C: The TSST– Salivary Cortisol: We tested whether Lower income was r(85) = 0.0195 We found that lower
2018 hypothesized that and parents their total combined C consisted of a Salivary cortisol values cortisol reactivity associated with a 0.25 income children have
lower SES would be participated in the monthly household income story preparation, were collected at 8 was associated with lower CSR slope, both a blunted CAR
associated with baseline after taxes and education story telling, and times: 10 min and SES and memory. representing both and blunted reactivity
lower memory measurement of an degree from 1 = none to 13 mental calculation immediately preceding Cortisol intercept reactivity and and recovery to acute
functioning, lower ongoing longitudinal = PhD with 9 or more part (5 min each) the stress task and at 0, and slope were recovery (see Table 3 stress. Interestingly,
CARs and stress study, of which 5 indicating higher education. and was 10, 20, 30, 40, and 50 regressed on SES to for parameter the main effect
hyporeactivity. children chose to Occupational status was performed in min after the stress test for main effects estimates and Figure indicating
Second, we discontinue their self–reported from 1 = front of 2 live task. All raw cortisol of SES. 1). hyporeactivity in CAR
predicted that participation during never employed to 15 = female judges and values were log and CSR was specific
cortisol the first session (final high civil service. We chose a video camera. transformed to correct to using income as
dysregulation would sample n = 142). 102 to focus on maternal, not for significant skew and the predictor and did
associate with lower children (M = 7.16 paternal, education and standardized to sample not replicate with
memory years, SD = 0.46, occupation to reduce the 1 (the first pre-stress education,
performance. Third, range = 6.08 – 7.98, number of SES predictors. sample). occupation or an SES
we hypothesized 46% female) were An SES composite score was composite. We
that lower SES randomly assigned to computed based on a single hypothesize that
would be related the TSST–C. The other factor analysis of family income may have
with smaller children functioned as income, maternal education specific effects
hippocampal a control group for an and occupational status because it is the most
volume that experimental task (not variable SES indicator
predicts lower reported here) and may be more
memory exploring effects of closely associated
performance. acute stress. with different levels
of family stress
related to financial
strain.
Allwood et al We hypothesized Participants were 51 The Perceived Stress Scale TSST-C: The Salivary Cortisol: Over Pearson correlations Subjective stress r(39) = 0.006 Contrary to
2017 that the healthy is a 14-item measure stressor tasks the course of each were conducted to (child-reported -.42 hypotheses, findings
associations children/adolescents designed to assess one’s included speech stress session, seven to examine the perceived stress) was revealed that
between stress (52.9% boys) and their perception of stress over (5 minutes nine saliva samples bivariate associations significantly compared to actual
(perceived or mothers. Participants the past month. Each item preparation, 5- were collected from between anxiety negatively associated experiences of stress,
objective) and ranged in age from 7– is rated on a 5-point Likert minute speech), each participant. The symptoms, perceived with baseline and perceived stress has
cortisol and 16 (M=12.41, scale with higher ratings mental arithmetic range of sensitivity was stress, SLEs, cortisol peak cortisol, as well greater associations
cardiovascular SD=2.30), and the indicating higher perceived (5 minutes) and .007-3.0 ug/dL, and and cardiovascular as with cortisol with both cortisol and
activity would be racial composition in stress (0 = “never,” 2 = mirror tracing (5 intra and inter- assay measures. Lastly, a change. In other cardiovascular
positive, such that this sample was 80.4% “sometimes,” 4 = “very minutes) tasks. coefficients of variation series of hierarchical words, as compared activity. Perhaps the
high stress would be Caucasian, 7.8% often”). Both child and The speech and less than 5 and 10%, regression analyses to children and normative stressors
associated with African American, parent completed the PSS, mental arithmetic respectively. Salivary were planned to adolescents with assessed in this study
higher cortisol, 7.8% Multi-racial, with parent ratings tasks were cortisol data is examine whether lower levels of (e.g., moving to a new
heart rate and 2.0% Hispanic, and reflecting their perception modified from the expressed in perceived stress perceived stress, school, parental
blood pressure. 2.0% Other. of their child’s stress. TSST for Children; micrograms per mediated or those with higher divorce) do not meet
the mirror-tracing deciliter (ug/dL). moderated the levels of perceived threshold to generate
task was adapted association between stress had lower the types of biological
from Allen and SLEs and biological baseline and peak alterations and stress
Matthews (1997). measures of stress. cortisol, and lower responses found in
changes in cortisol studies that assess
(see Figure 1; Table trauma and severe
4) adversities.
114
Fearon et al We explored the In total, 316 Cumulative Contextual TSST: Participants Salivary Cortisol: We modelled the As Figure 4 shows, t(143) = .02869 The results of our
2017 role of cumulative adolescents provided Adversity: Using data were told that Before TSST procedure, cortisol response for females there was 2.21 study, though in need
contextual adversity cortisol samples, 15 of completed by the child’s they would be participants (who had profile using a stronger (linear) Robust- of replication, provide
on the stress whom were excluded carer and the child at the given 3 minutes to not eaten or drunk polynomial functions increase in cortisol ness further evidence that
response, testing due to asthma steroid age 13 assessment, we prepare a 5- anything in the last to describe the response for those Quad: early caregiving may
whether this might pump use, which can dichotomized the following minute speech on hour) were first asked change in cortisol with high levels of t(143) = indeed be implicated
account for any interfere with cortisol indices, scored as present anything about to provide a saliva over time, including adversity, compared 2.03 in HPA axis
observed effects of measurements. A versus absent (zero versus themselves. After sample by directly filling terms for the to those with lower development into
attachment further 14 cases were one): overcrowding, the speech, one of a 2 milliliter plastic intercept (the adversity (Linear B= adolescence and in
security or lost due to problems community violence the ‘examiners’ sampling device baseline level), .053, 95% CI[.005, buffering the HPA axis
sensitivity, and with the labelling and exposure, house has no administered a (SaliCap) or using a linear, quadratic and .100], p= .030; from the effects of
whether higher storage of the running water, house has serial sevens short plastic straw to do cubic trends. The Quadratic B= −.006, chronic and extreme
quality of early care samples. Of these, no toilet, house has no subtraction task. so. After the TSST effects were then 95% CI[−.013, .001], adversity.
might moderate the 232 had been electricity, the parent This was stressors research modelled as a p= .106). In contrast,
impact of observed in the endorsed that members of maintained for 4 assistant entered the function of level 2 for males, the effects
cumulative Strange Situation at the family had gone for minutes without room and led the variables (across of adversity on the
contextual adversity 12 months, 217 had whole day without eating any intervention participant to the next individuals), such as linear (B= −.048, 95%
on HPA reactivity. completed because of a lack of food, or responses from room where a second attachment and CI[−.090, −.005], p=
assessments of primary caregiver is the ‘examiners’ saliva sample was parental sensitivity. .029) and quadratic
general sensitivity and unemployed, caregiver has irrespective of collected. After the (B= .007, 95%
212 completed the only primary-level how well the TSST, an unrelated CI[.0002, .013], p=
assessment of education, relationship participant was interview was .044) slopes were in
sensitivity during breakup with partner or performing. interrupted every 10 the opposite
feeding husband, partner has been minutes to collect a direction.
violent towards saliva sample.
mother/caregiver.
Laceulle et al It is hypothesized In total, 178 Social Loss or Social Defeat: Social Stress Task: Salivary Cortisol: We Associations With regard to the t(177) = .04813 Adolescents who
2017 that adversity is adolescents Participants were assessed This test was the assessed HPA-axis between adversity effect of adversity, 1.99 reported being
related to changes participated in the with the Life Stress last challenge of responses towards the and HPA-axis social defeat was exposed to social
in HPA-axis behavioural Interview LSI - social defeat the experimental GSST using four cortisol functioning were related to changes in defeat (but not
functioning. Based experiments both at group (i.e., if they had been sessions at age 16 samples (referred to as examined using both basal cortisol loss/illness) showed
on the existing age 16 and 19 and in a victim of bullying, sexual and 19. It involved CE1, CE2, CE3, CE4). Univariate ANOVA’s, and reaction to stress increases in basal
literature, effects the LSI at age 19 (no intimidation or violence or a standardized one for each of the during adolescence. cortisol and
are expected to be stress = 68; social were dumped after a protocol including cortisol measures. Specifically, decreases in reaction
larger for social defeat = 32 and serious relationship), or public speaking Dependent variables adolescents who to the stress task
defeat than for loss/illness = 61, 17 comparison group (i.e., and mental were the HPA-axis reported being from age 16 to 19,
loss/illness. participants randomly selected from the arithmetic, functioning measures exposed to social compared to their
Whether the HPA- experienced both pool of participants that inspired by the at T4. Independent defeat showed higher peers.
axis will respond to social defeat and experienced no loss or Trier Social Stress variables were the basal cortisol (B =
adversity with loss/illness and were defeat, or a loss or defeat Task, for the HPA-axis functioning −0.39, SE = 0.19, p =
hyper-reactivity or included in both that was rated 1 or 2 on the induction of measures at T3 and 0.040, ɳ2 = 0.029)
hypo-reactivity will adversity groups). For severity and/or and person- moderate the two adversity and lower reaction to
be explored. one participant from in- dependency scale). performance- dummies (i.e., social the stress task (B =
the loss/defeat and related social defeat (1) vs the rest 0.33, SE = 0.17, p =
social defeat group we stress. (0) and social loss (1) 0.048, ɳ2 = 0.030).
did not have any data, vs the rest (0)).
resulting in a sample
of 177 adolescents for
the current study
(46.3% girls).
115
Lucas- In line with past Two-parent families Parent Conflict: Parents TSST: In the Salivary Cortisol: Saliva Primary analyses Fathers who were r(139) = - .01244 More-negative
Thompson et al research, we with at least one child reported the frequency, modified TSST, was collected using the were conducted observed to behave .21 behavior as displayed
2017 hypothesized that between 10 and 17 intensity, and resolution of participants give a Salivette device separately on cortisol more negatively had by fathers during
marital conflict and years of age were their marital conflict using speech about (Sarstedt, Nu€mbrecht, and sAA AUCi adolescents who marital conflict
conflict appraisals recruited from the two questionnaires. Parents personal Germany). We reactivity, as well as produced marginally predicted lower levels
have unique effects community through reported the frequency and characteristics examined stress total amounts of less cortisol across of adolescent cortisol
on physiological advertisements; 153 intensity of conflict using (imagining that physiology via cortisol cortisol and sAA the TSST. and sAA produced
reactivity. youth from 98 families the five-item Conflict they are in front and sAA production in produced across the Adolescents of these during the TSST
(52% female; Mage = subscale of the Braiker– of a new response to the TSST as TSST (AUCg). fathers also displayed
12.92, SD = 2.16) and Kelley Partnership classroom; captured by area under Generalized significantly less
both of their parents Questionnaire; the mean of approximately 5 the curve (AUC). estimating equation cortisol reactivity to
participated in the each participant’s min) and then (GEE) models the TSST. In terms of
current study. responses was calculated. engage in out- (regression-based, conflict appraisals,
Observed Conflict loud mental nonparametric) were adolescents who
Behaviour: each parent was arithmetic used to adjust for appraised the
rated based on the degree (approximately 5 clustering of youth properties of their
to which specific min). within families. Main parents’ conflict
destructive conflict effects were tested more negatively
behaviors were displayed (0 first; then, produced
= absent to 2 = very strong interactions were significantly lower
display). tested separately by levels of sAA across
creating the TSST. Adolescent
multiplicative conflict property and
interaction terms threat appraisals
(after centering) and were also
controlling for lower significantly (and
order terms negatively)
associated with
cortisol AUCi
reactivity (Table 1).
Obasi et al Based on previous Participants (N = 60) Family Resources: Family TSST: Participants Salivary Cortisol: Four Individual difference Family resources had t(56) = - .0252 Participants who
2017 research findings, consisted of rural resources was measured by were to introduce samples of salivary predictors were a significant positive 2.30 reported lower
the following African Americans the Family Resource Scale. themselves to the cortisol were collected modeled as cross- relationship with PCR consistency of having
hypotheses were between the ages of The FRS is a 30-item self- “committee” in a during the TSST. Peak level interactions (γ21 = .001, t(56) = access to adequate
made: (1) Rural 18 and 22 (X = 20.0, report measure that free speech (5 cortisol levels are found between these 2.30, p = .025). More family resources in
African American SD = 1.1). The assesses an individual’s min) and convince in saliva 20–30 minutes cortisol parameters specifically, coming the household had a
emerging adults will majority of the perceived adequacy of them why they after the experience of and variables of from a family with blunted peak cortisol
exhibit blunted participants were concrete resources in the should be hired an acute stressor so interest (i.e., diminished levels of reactivity to acute
cortisol levels when female (n = 39, household. Resources for a job vacancy. sample #4 was designed perceived stress or basic resources was stress.
experiencing higher 63.9%), unmarried (n include physical necessities During the to represent the peak family resources). associated with a
levels of perceived = 58, 95.1%), and self- and shelter (8 items), arithmetic cortisol response (PCR) We tested whether blunted peak cortisol
stress; and (2) HPA- identified as 5th growth and support (9 component, to the TSST. some of the response to acute
axis reactivity to generation in items), necessities and participants were variability in PCR stress
acute stress will be response to health (7 items), interfamily asked to serially could be accounted
lower among immigrant status (n = support (2 items), child care subtract the for by family
participants with 54, 88.5%). (2 items), and personal number 13 from resources after
low family resources (2 items). 1,022 as quickly controlling for age,
resources. and as accurately sex, perceived stress,
as possible (5 and family resources
min). at baseline.
116
Tackett et al Hypothesis 1: We Of an initial sample of SES: We also examined TSST- C: Salivary Cortisol: Once Hierarchical Cortisol reactivity r(257) = .00121 These findings
2017 predicted that low 350 participants, all sociocontextual stress using Participants the child had given regression models and EI (summed z- .20 suggest that
socioeconomic Black, White, and different indices: an index engaged in an assent, they rinsed their were estimated to score of education Robust- cumulative effects of
status (low summed Hispanic participants of SES reflecting both adapted version mouth out with water examine the and income) ness: stress and
education/income were selected for the mother’s highest level of of the TSST-C, a and were instructed to incremental variance correlation in Table 3 r(257) = disadvantage may
and neighborhood present study education and overall laboratory drink a small cup of in cortisol reactivity - r(257) = .20 .22 result in differences
affluence) would (excluding smaller family income (referred to paradigm water. After 30 min of and recovery in stress response
predict lower categories reflecting in this paper as summed consisting of a sedentary activity predicted by physiology as early as
cortisol reactivity Asian, Multiracial, and education/ income), which storytelling and (questionnaire and summed middle childhood,
and recovery, above Other groups), is a commonly used math task computer task education/income, and that race-specific
and beyond resulting in a sample approach to indexing SES in designed to elicit completion), the neighborhood mechanisms account
prediction by of 296 children and the literature and an index a stress response baseline saliva sample affluence, or FAM for additional
race/ethnic group their primary female of SES reflecting overall to a psychosocial was collected. Saliva over and above variance in cortisol
membership, in caregivers (98% neighborhood affluence (or, stressor. samples were then race/ethnicity. reactivity and
hierarchical mothers). Children conversely, deprivation) collected 20 min after Dependent variables recovery. These
regression models. were primarily ages coded from census data, the start of the TSST were cortisol findings support the
9–10 years (M 9.82, which is another measure (excluding speech reactivity and cortisol “wear-and-tear”
SD 0.66; 140 males that has been used in preparation time, so recovery at T4. theory regarding the
[47%], full age range 8 previous research. from the beginning of effects of
–11). the speech delivery; socioeconomic status
T2), 40 min after the and contextual stress
start of the TSST (T3), on physiological
and 60 min after the functioning (McEwen,
start of the TSST (T4). 2004), even as early
nmol/L as preadolescence.
Wan et al 2017 We were interested Research participants Life Experiences Survey TSST: The TSST is Salivary Cortisol: Saliva Using simple In breast cancer r(17) = .04537 Overall, breast cancer
in exploring the role comprised a group of (LES): The LES is a a standardised samples were collected regression analyses, survivors, there was a -.464 survivors who
of stressful life breast cancer questionnaire designed to laboratory on oral swabs. we assessed the significant negative Robust- experienced a higher
events and their survivors and a group measure life changes (also protocol designed Participants were following: (1) correlation between ness: number of stressful
perceived impact on of women without a labelled as stressors) and to induce a instructed to place the Whether the number the number of r(17)= life events and
the blunted reactive history of breast their subjective impact. It moderate stress oral swabs under the of stressful life stressful life events -.552 perceived higher
cortisol stress cancer. Breast cancer comprises 47 items that list response. The tongue for three events experienced and the peak cortisol levels of disturbance
response noted in survivors (n = 19) and various common stressors. participant is minutes. A total of impacted the cortisol concentration (r = – associated with these
the same group of control group (n = 17) We calculated the asked to prepare seven saliva samples concentrations of .464, p = .035), events had lower
breast cancer following: (1) the total a speech that is were collected: Arrival, both groups of indicating that a levels of cortisol
survivors. We number of stressful life then delivered in after a 10-min participants; and (2) higher frequency of during the peak,
evaluated whether events, (2) the total number front of a panel of acclimatisation period; whether subjective stressful events lowest point (or
the number of of positive events, (3) the judges who have Anticipation, appraisal of the predicted lower trough) and the AUC
stressful life events total number of negative been instructed to immediately after the stressful life event, levels of cortisol. The in the reactive profile.
and the individual’s events, (4) the sum of the deliver no verbal speech preparation; affected participants’ perceived impact of The frequency and
subjective appraisal Impact score given to or non-verbal Arithmetic, immediately cortisol total stressful life impact of positive
of its impact, as positive events (Positive feedback; this is after the completion of concentration. events and peak stressful life events,
measured by the Impact Score), (5) the sum followed by a both speech and cortisol in particular, were
LES, would of the Impact score given to challenging verbal arithmetic tasks; and10- concentrations was negatively correlated
influence negative events (Negative arithmetic task in ,20-,40-,60- negatively correlated with the peak cortisol
physiological Impact Score) and (6) the front of the same minafterwards, during (r = –.552, p = .013). response and AUC,
cortisol profiles. sum of both positive and panel. which they completed whereas the trough
negative impact scores the LES. was negatively
(Total Impact Score). correlated with the
number and impact
117
of negative stressful
life events.
Afifi et al 2015 H1: Couples’ The sample included Economic Uncertainty: Stressful Salivary Cortisol: The Dyadic growth curve Wives with higher t(80) = - .0372 Wives with lower
communal coping 82 married Sixteen items were created Discussion: parents and adolescent modeling (using the economic uncertainty 2.12 economic uncertainty
and community (M=16.61years), by the authors to measure Couples were child provided baseline SPSS Mixed evidenced a blunted experienced a much
support will heterosexual parents parents’ economic asked to talk with saliva samples and then Procedure) was used cortisol response, or greater cortisol
contribute to their and one of their uncertainty in the each other for as took a survey. to examine the an increase to the
(a) lower baseline adolescent children preinteraction survey. e long as they could Immediately upon influence of inhibited/diminished discussion task
levels of cortisol (ages 12–18; M=14.7) EUS asked parents how up to 20 minutes completing the husbands’ and wives’ stress response, compared to
and sAA and (b) less from Southern uncertain or unsure they about their discussion task, the communication skills which was indicative everyone else,
reactance and (c) California. The were that in the next year uncertainties couple provided a on their cortisol and of a sharp, whereas wives with
greater recovery in families included 44 (time frame can be regarding their second set of saliva sAA response and immediate drop in higher economic
their cortisol and Caucasian couples, 25 adjusted) that they will lose family’s finances, samples. A third saliva recovery patterns to cortisol in response uncertainty
sAA from a stressful Latino couples, and 13 their job, no longer be able or things they sample was collected the discussion task to the stress task experienced a
discussion about mixed-race couples to pay their mortgage or were unsure 15 minutes after the (H1 and H2). The (rather than the blunted cortisol
economic where one partner rent, le for bankruptcy, and about regarding end of the discussion models represented expected increase) response.
uncertainty. was Caucasian and not be able to provide for their family’s and a fourth saliva two levels of analysis (See Table 4 - Mother
H2: Couples’ the other was Latino their child’s needs, among financial future, sample was collected with the multivariate x Economic
corumination and other issues (see Table 1 for that made them 40 minutes after the repeated measures Uncertainty).
interparental scale items and factor stressed and that end of the discussion. (i.e., four cortisol
conflict will loadings). e scale ranged tended to scores and four sAA
contribute to their from 1 to 5 with “1” being produce conflict scores for the
(a) higher baseline very unsure and “5” being between the two husband and a
levels of cortisol very sure (items were then of them. corresponding set of
and sAA and (b) reverse coded so that cortisol scores and
greater reactance higher scores equal greater sAA scores for the
and (c) slower economic uncertainty). wife) at Level 1 and
recovery from a Every item also had a “not the couple (with the
stressful discussion applicable” option (these couples’ score
about economic items were then omitted
uncertainty. from the scale).
118
Wieke de Since Kudielka et al. Eighty- five patients Burnout: Burnout Psychological Salivary Cortisol: Alpha- To test between- For cortisol, healthy F(2,89.5) .04066 This study assessed
Vente et al stated that ongoing were recruited complaints were measured Stressor: To study amylase and cortisol group differences in males showed earlier = 3.32 whether burnout is
2015 chronic stress is through occupational with the Maslach Burnout physiological were determined in reactivity-recovery and stronger cortisol characterised by
generally associated physicians, general Inventory-General Survey reactivity and saliva. Means of the and in overall mean reactivity dysregulation of the
with a hyperactive practitioners, and by (MBI-GS), which consists of recovery, first (−4 min. in relation activity during the immediately after sympathetic vagal
HPA axis and our self- referral to three subscales: emotional participants were to the start of the psychosocial stress cessation of the balance and the HPA
sample was participate in a RCT exhaustion (5 items), exposed to an stressor) and second procedure, mood, stressor than male axis. In contrast to
recruited within about treatment depersonalisation (4 items), acute (+5 min.) saliva samples cardiovascular, and patients. Mean prediction, reduced
weeks to a few efficacy of stress- and professional psychosocial were used as resting neuroendocrine reactivity cortisol reactivity to
months after having management training competence (6 items). stressor consisting values, indicative of reactivity and immediately after the an acute psychosocial
called themselves Twenty-seven Items were scored on 7- of a speech basal functioning (see recovery during the stressor of healthy stressor was observed
sick (burnout), we patients did not fulfill point Likert scales (0 = preparation task, Figure 1). For cortisol, psychosocial stress males was 0.91 in male patients,
expected to find the inclusion criteria never to 6 = always/daily) a mental the third (+20 min.) and procedure were ng/mL, which is which suggests
support for a (e.g., having major and mean subscale scores arithmetic task, fourth (+35 min.) saliva examined with almost equal to the hyporeactivity of the
hyperactive rather depression as a were calculated. Higher and a speech task. samples indicated ANOVA for repeated operational guideline HPA axis, rather than
than a hypoactive primary diagnosis) or scores reflect higher levels reactivity, and the fifth measures, using a for cortisol reactivity hyperactivity.
HPA axis. refused to take part in of emotional exhaustion, (+50 min.) indicated one within- (time), of 1 ng/mL .A cortisol
the study. As a distant/cynical attitudes sample recovery. one between- reaction could not be
reference group, forty towards work, and (group) subject observed in male
healthy individuals professional competence. factor design. patients (See Table
were recruited. 2).
Calhoun et al Given that relational Participants included The victimization subscales TSST: Adolescents Salivary Cortisol: Descriptive statistics Results indicated that t(60) = .028 Relational
2014 victimization is a 62 youth (73% female) of the Revised Peer participated in the Adolescents provided and correlations (or t higher levels of 2.25 victimization was
significant source of at the adolescent Experiences Questionnaire TSST during the salivary cortisol samples values, for gender relational associated with
stress during transition, between were used to assess overt laboratory using a passive drool and medication victimization were blunted HPA
adolescence, it was the ages of 12 and 16 and relational forms of assessment. procedure on three usage) were associated with a reactivity in this
hypothesized that years (M = 14.70, SD = victimization. The occasions throughout computed for all more hyporeactive study. Some evidence
relational 1.33). relational victimization the baseline laboratory study variables. cortisol response to suggests that
victimization would subscale assessed being the assessment described Primary study the social stressor individuals with
be associated with recipient of direct or above: immediately analyses consisted of task (See table 2; histories of chronic
dysregulated HPA indirect social aggression prior to the speech task latent difference Relational stress (characteristic
stress responses, as (e.g., “A teen left me out of (following a 20-min structural equation victimization and of the adolescents in
indicated by hypo- an activity or conversation relaxation period), 20 models for pre–post– cortisol reactivity). this study) may be
or hyperreactivity to that I really wanted to be min postspeech, and 40 post study designs. A more prone toward
the social stressor included in”). min postspeech. logarithmic blunted HPA
task. Changes in cortisol transformation was responses. Similar to
levels from baseline to applied to cortisol other chronic
stressor represent values prior to model stressors, relational
cortisol “reactivity,” identification. victimization during
while comparisons of adolescence may
baseline and serve as a stable
poststressor levels of source of stress that
cortisol are indicative of frequently engages
poststressor cortisol HPA stress responses,
“recovery” (i.e., return increasing allostatic
to baseline). load to the point that
HPA responses
become dysregulated.
119
Quas et al 2014 A small set of A total of 324 4- to 6- Chronic Stress: From the TSST—Child Salivary Cortisol: Saliva Scores in each The subgroups F(5, 284) .02795 In PAWS, the
children, it was year-olds involved in a measures of income and (TSST-C), which samples were collected sample were entered differed in SES, = 2.56 underaroused
hypothesized, longitudinal study of education, a composite required they using a dental role into latent profile calculated as the Robus- children had
would exhibit family social status, index of socioeconomic finish a story and before the TSST-C analysis (LPA), a form mean of standardized ness significantly lower
heightened biological responses status (SES) was created. complete mental began (baseline) and 10 of finite mixture scores based on F(4,223) family SES than did
reactivity across all to adversity, and Parents also completed arithmetic in front and 25 min after the modeling that parents’ highest = 2.48 the other subgroups.
systems (i.e., high health served as measures of their mental of one observer. TSST-C ended. identifies potential education level and Perhaps these
activation of the participants. Children health, financial stress, unobserved annual income, F (5, children had
SNS and the HPA were recruited from parental stress, and subgroups among a 284) = 2.56, p = .027. encountered a larger
axis, and high kindergarten marital/relationship set of indicators. Children in the range of other
deactivation of the classrooms in public functioning, both in the fall Models were fitted underaroused potentially stressful
PNS), a pattern schools in an urban of children’s participation in within a structural subgroup came from experiences,
suggested by area in Northern the larger study and again equation framework, substantially lower rendering the
studies in which California. in the spring. An overall and missing data SES families (M = laboratory
reactivity in Subsample: This index of chronic stress was were estimated using −1.31) than children assessment
multiple systems sample included 111 created by averaging full information in the other innocuous or
have each 11- and 12-year-olds parents’ standardized maximum likelihood subgroups. A one- unchallenging by
independently been in an urban public scores on the four using Mplus 6.11 way group ANOVA comparison.
linked to early school in measures in the fall and conducted predicting Alternatively,
childhood stress northwestern Canada spring. This created a single the adversity index exposure to mild
and suggested by whose parents cross-time marker of also revealed group chronic stress, which
Del Giudice et al. consented to a chronic family adversity differences, F (4, 223) often accompanies
(2011) among science study on that could be included as a = 2.48, p = .05. low SES, may already
children exposed to stress regulation. predictor of subgroup be compromising
chronic threats. reactivity differences. these children’s
general response
tendencies.
Badanes et al In study 1, we Participants— Stressful Home Stress Paradigm: Salivary Cortisol: Three A series of Although stress (in Chi2(1) = .04523 The results from
2011 worked with a Participants were 110 Environment: Financial Children saliva samples were hierarchical binary this case only 4.01 study 1 support our
sample of high risk preschoolers (53% Strain: assessed with the participated in a collected across both logistic regression financial strain) hypothesis that stress
preschoolers and female) enrolled in Family Finances 20-minute semi- the classroom and the models were predicted attenuated within a high risk
examined whether one of 14 full- day questionnaire consisted of structured mild home interactions in computed first cortisol, χ2(1) = 4.01, sample early in
known significant child care classrooms. one item addressing family stressor with the 20-min intervals; at predicting p = .04, both financial development is
stressors impacting The children were 2 to size, 2 items assessing support of their baseline before the task internalizing strain and maternal associated with HPA-
the family system 6 years of age (M = sources of income, 1 lead teacher at began; at the end of the problems (0, 1 above depression predicted axis attenuation.
(maternal 4.03, SEM = .07). question asking about child care and of a interaction assumed to the clinical cutoff for internalizing Specifically, the data
depression and Seventy-five percent amount of income earned, parent at home. reflect cortisol internalizing symptoms (see illustrate that serious
serious financial of children were and 5 items asking about reactivity to the task, symptoms as rated results above), and financial strain is
strain) predicted identified by parents insecurity of critical and 20-minutes after by mother, teacher, attenuated cortisol related to both
which children as white and 25% as resources. Items asked the task ended assumed or both) from all of predicted blunted basal cortisol
exhibited non-white. whether in the past 12 to reflect recovery to the available internalizing and attenuated stress
attenuated cortisol months the following was the task. demographic control symptoms (see reactivity in
across the day in true or false for the family: variables, the family results above), the preschoolers.
two normal 1) Your family went without risk variables, and effect of stress on
environments a telephone, 2) You didn’t the potential internalizing
(home and child pay all or part of the rent/ allostatic load symptoms was not
care) and across a mortgage because you did marker of reduced with the
stress paradigm not have enough money, 3) hypocortisolemia inclusion of
presented to the Your family was evicted for attenuated cortisol in
child in the failure to pay, 4) You didn’t the model.
120
presence of the pay all or part of the

mother or lead gas/electric/oil bill, 5) You
teacher. had your oil/gas/electricity
turned off for failure to pay.
Neuport et al The main goal of Participants in the SES: Socioeconomic status Cognitive Salivary Cortisol: Multilevel modeling An additional model r(72) = .3 .00094 The main purpose of
2006 the present study present study were was operationalized as Challenge: Physiological stress was was used to examine testing for education this study was to
was to examine from the Boston years of education. This Cognitively assessed by salivary age and education differences (i.e., investigate age and
potential age and oversample of the strategy was chosen challenging tests cortisol levels which differences in cortisol replacing age with SES differences in
SES differences in Midlife in the United because it captures the in four domains were collected via trajectories. At Level the continuous physiological
physiological States (MIDUS) well-established gradient of (vocabulary, Sarstedt Salivettes; the 1, each person’s education variable in reactivity to cognitive
reactivity to Survey. The Boston In- socioeconomic short-term participants inserted a variability (e.g., the previous model) stressors. Consistent
cognitive stressors. Depth Study of disadvantage, and it working memory, cotton-like swab into change in cortisol indicated that those with past work and in
We also sought to Management captures the primary speed, and their mouths for 30 levels over time) is with higher line with our
examine the role of Processes in Midlife educational benchmarks reasoning) were seconds and then represented by an education were more hypotheses, older
SES in reactivity. included 302 adults that provide the foundation given to placed it into a glass intercept and slope reactive than those adults were more
Specifically, we (aged 25-74 years) for subsequent participants. container. Each that become the with less education reactive compared to
expected higher SES from a probability stratification processes by Similar cognitive participant completed outcome variables in (see Model 2 of Table younger adults.
would be associated sample of the Greater occupation and earnings. tests have elicited between five and seven a Level 2 model in 3). Higher SES (in this
with greater Boston area. Eighty- Moreover, educational stress reactions in Salivette trials, which they may case, education) was
physiological eight (29%) individuals attainment has been the past studies. depending on how depend on person- also associated with
reactivity to provided salivary primary proxy for much time they level characteristics heightened reactivity.
cognitive tasks. cortisol readings socioeconomic status in required to complete (e.g., age and
during the cognitive previous investigations, the interview after the education).
tests. thereby allowing cognitive tests.
comparability with other
studies.
121
2.9 References
Afifi, T., Davis, S., Merrill, A. F., Coveleski, S., Denes, A., & Afifi, W. (2015). In the
Wake of the Great Recession: Economic Uncertainty, Communication, and
Biological Stress Responses in Families: In the Wake of the Great Recession.
Human Communication Research, 41(2), 268–302.
https://doi.org/10.1111/hcre.12048
Aiyer, S. M., Heinze, J. E., Miller, A. L., Stoddard, S. A., & Zimmerman, M. A. (2014).
Exposure to Violence Predicting Cortisol Response During Adolescence and
Early Adulthood: Understanding Moderating Factors. Journal of Youth and
Adolescence, 43(7), 1066–1079. https://doi.org/10.1007/s10964-014-0097-8
Allwood, M. A., Gaffey, A. E., Vergara-Lopez, C., & Stroud, L. R. (2017). Stress
through the mind of the beholder: Preliminary differences in child and maternal
perceptions of child stress in relation to child cortisol and cardiovascular activity.
Stress, 20(4), 341–349. https://doi.org/10.1080/10253890.2017.1336617
American Psychiatric Association. (2013). Diagnostic and statistical manual of mental

disorders (DSM-5®). American Psychiatric Pub.
Andreotti, C., Garrard, P., Venkatraman, S. L., & Compas, B. E. (2015). Stress-Related
Changes in Attentional Bias to Social Threat in Young Adults: Psychobiological
Associations with the Early Family Environment. Cognitive Therapy and
Research, 39(3), 332–342. https://doi.org/10.1007/s10608-014-9659-z
Arnsten, A. F. (2009). Stress signalling pathways that impair prefrontal cortex structure
and function. Nature reviews neuroscience, 10(6), 410-422.
https://doi.org/10.1038/nrn2648
Aschbacher, K., O’Donovan, A., Wolkowitz, O. M., Dhabhar, F. S., Su, Y., & Epel, E.
(2013). Good Stress, Bad Stress and Oxidative Stress: Insights from Anticipatory
122
Cortisol Reactivity. Psychoneuroendocrinology, 38(9), 1698–1708.

https://doi.org/10.1016/j.psyneuen.2013.02.004
Badanes, L. S., Watamura, S. E., & Hankin, B. L. (2011). Hypocortisolism as a potential

marker of allostatic load in children: Associations with family risk and
internalizing disorders. Development and Psychopathology, 23(3), 881–896.
https://doi.org/10.1017/S095457941100037X

Blankenship, S. L., Chad-Friedman, E., Riggins, T., & Dougherty, L. R. (2019). Early
parenting predicts hippocampal subregion volume via stress reactivity in
childhood. Developmental Psychobiology, 61(1), 125–140.
https://doi.org/10.1002/dev.21788
Boggero, I. A., Hostinar, C. E., Haak, E. A., Murphy, M. L., & Segerstrom, S. C. (2017).
Psychosocial functioning and the cortisol awakening response: Meta-analysis, P-
curve analysis, and evaluation of the evidential value in existing
studies. Biological psychology, 129, 207-
230. doi: 10.1016/j.biopsycho.2017.08.058
Bosch, N. M., Riese, H., Reijneveld, S. A., Bakker, M. P., Verhulst, F. C., Ormel, J., &
Oldehinkel, A. J. (2012). Timing matters: Long term effects of adversities from
prenatal period up to adolescence on adolescents’ cortisol stress response. The
TRAILS study. Psychoneuroendocrinology, 37(9), 1439–1447.
Boyce, W. T., Chesney, M., Alkon, A., Tschann, J. M., Adams, S., Chesterman, B.,
Cohen, F., Kaiser, P., Folkman, S., & Wara, D. (1995). Psychobiologic Reactivity
to Stress and Childhood Respiratory Illnesses: Results of Two Prospective
Studies. Psychosomatic Medicine, 12.
123

https://doi.org/10.1017/S0954579405050145
Brand, S. R., Brennan, P. A., Newport, D. J., Smith, A. K., Weiss, T., & Stowe, Z. N.
(2010). The impact of maternal childhood abuse on maternal and infant HPA axis
function in the postpartum period. Psychoneuroendocrinology, 35(5), 686–693.
Bremner, J. D., Vythilingam, M., Vermetten, E., Adil, J., Khan, S., Nazeer, A., Afzal, N.,
McGlashan, T., Elzinga, B., Anderson, G. M., Heninger, G., Southwick, S. M., &
Charney, D. S. (2003). Cortisol response to a cognitive stress challenge in
posttraumatic stress disorder (PTSD) related to childhood abuse.
Psychoneuroendocrinology, 28(6), 733–750. https://doi.org/10.1016/S0306-
4530(02)00067-7
7(12), 1274. https://doi.org/10.1038/s41398-017-0032-3
Busse, D., Yim, I. S., & Campos, B. (2017). Social context matters: Ethnicity,
discrimination and stress reactivity. Psychoneuroendocrinology, 83, 187–193.
Calhoun, C. D., Helms, S. W., Heilbron, N., Rudolph, K. D., Hastings, P. D., & Prinstein,
M. J. (2014). Relational victimization, friendship, and adolescents’ hypothalamic–
pituitary–adrenal axis responses to an in vivo social stressor. Development and
Psychopathology, 26(3), 605–618. https://doi.org/10.1017/S0954579414000261
Carpenter, L. L., Shattuck, T. T., Tyrka, A. R., Geracioti, T. D., & Price, L. H. (2011).
Effect of childhood physical abuse on cortisol stress response.
Psychopharmacology, 214(1), 367–375. https://doi.org/10.1007/s00213-010-
2007-4
124
Chen, G., Kong, Y., Deater-Deckard, K., & Zhang, W. (2018). Bullying Victimization
Heightens Cortisol Response to Psychosocial Stress in Chinese Children. Journal
of Abnormal Child Psychology, 46(5), 1051–1059.
https://doi.org/10.1007/s10802-017-0366-6
Chiang, J. J., Ko, A., Bower, J. E., Taylor, S. E., Irwin, M. R., & Fuligni, A. J. (2019).
Stress, Psychological Resources, and HPA and Inflammatory Reactivity During
Late Adolescence. Development and Psychopathology, 31(02), 699–712.
https://doi.org/10.1017/S0954579418000287
Coppens, E., Kempke, S., Van Wambeke, P., Claes, S., Morlion, B., Luyten, P., & Van
Oudenhove, L. (2018). Cortisol and Subjective Stress Responses to Acute
Psychosocial Stress in Fibromyalgia Patients and Control Participants:
Psychosomatic Medicine, 80(3), 317–326.
https://doi.org/10.1097/PSY.0000000000000551
Cordero, M. I., Moser, D. A., Manini, A., Suardi, F., Sancho-Rossignol, A., Torrisi, R.,
Rossier, M. F., Ansermet, F., Dayer, A. G., Rusconi-Serpa, S., & Schechter, D. S.
(2017). Effects of interpersonal violence-related post-traumatic stress disorder
(PTSD) on mother and child diurnal cortisol rhythm and cortisol reactivity to a
laboratory stressor involving separation. Hormones and Behavior, 90, 15–24.
https://doi.org/10.1016/j.yhbeh.2017.02.007
Cutuli, J. J., Wiik, K. L., Herbers, J. E., Gunnar, M. R., & Masten, A. S. (2010). Cortisol
Function Among Early School-aged Homeless Children.
Psychoneuroendocrinology, 35(6), 833–845.
Danese, A., & McEwen, B. S. (2012). Adverse childhood experiences, allostasis,

allostatic load, and age-related disease. Physiology & Behavior, 106(1), 29–39.
https://doi.org/10.1016/j.physbeh.2011.08.019
de Vente, W., van Amsterdam, J. G., Olff, M., Kamphuis, J. H., & Emmelkamp, P. M.
(2015). Burnout is associated with reduced parasympathetic activity and reduced
125
HPA axis responsiveness, predominantly in males. BioMed research

international, 2015.
Del Giudice, M., Ellis, B. J., & Shirtcliff, E. A. (2011). The Adaptive Calibration Model
of stress responsivity. Neuroscience and Biobehavioral Reviews, 35(7), 1562–
1592. https://doi.org/10.1016/j.neubiorev.2010.11.007
Del Rey, A., Chrousos, G., & Besedovsky, H. (2008). The hypothalamus-pituitary-
adrenal axis. Elsevier.
DePasquale, C. E., Donzella, B., & Gunnar, M. R. (2019). Pubertal recalibration of

cortisol reactivity following early life stress: A cross-sectional analysis. Journal
of Child Psychology and Psychiatry, 60(5), 566–575.
https://doi.org/10.1111/jcpp.12992
Dietz, L. J., Stoyak, S., Melhem, N., Porta, G., Matthews, K. A., Walker Payne, M., &
Brent, D. A. (2013). Cortisol Response to Social Stress in Parentally Bereaved
Youth. Biological Psychiatry, 73(4), 379–387.
https://doi.org/10.1016/j.biopsych.2012.08.016
Doom, J. R., Cook, S. H., Sturza, J., Kaciroti, N., Gearhardt, A. N., Vazquez, D. M.,
Lumeng, J. C., & Miller, A. L. (2018). Family conflict, chaos, and negative life
events predict cortisol activity in low-income children. Developmental
Psychobiology, 60(4), 364–379. https://doi.org/10.1002/dev.21602
Ellis, B. J., Essex, M. J., & Boyce, W. T. (2005). Biological sensitivity to context: II.
Empirical explorations of an evolutionary–developmental theory. Development
and Psychopathology, 17(02). https://doi.org/10.1017/S0954579405050157
Elzinga, B. M., Roelofs, K., Tollenaar, M. S., Bakvis, P., van Pelt, J., & Spinhoven, P.
(2008). Diminished cortisol responses to psychosocial stress associated with
lifetime adverse events. Psychoneuroendocrinology, 33(2), 227–237.
126
Elzinga, B. M., Spinhoven, P., Berretty, E., de Jong, P., & Roelofs, K. (2010). The role of
childhood abuse in HPA-axis reactivity in Social Anxiety Disorder: A pilot study.
Biological Psychology, 83(1), 1–6.
https://doi.org/10.1016/j.biopsycho.2009.09.006
England-Mason, G., Kimber, M., Khoury, J., Atkinson, L., MacMillan, H., & Gonzalez,
A. (2017). Difficulties with emotion regulation moderate the association between
childhood history of maltreatment and cortisol reactivity to psychosocial
challenge in postpartum women. Hormones and Behavior, 95, 44–56.
https://doi.org/10.1016/j.yhbeh.2017.07.007
Ewing-Cobbs, L., Prasad, M. R., Cox, C. S., Granger, D. A., Duque, G., & Swank, P. R.
(2017). Altered stress system reactivity after pediatric injury: Relation with post-
traumatic stress symptoms. Psychoneuroendocrinology, 84, 66–75.
Fearon, R. M. P., Tomlinson, M., Kumsta, R., Skeen, S., Murray, L., Cooper, P. J., &
Morgan, B. (2017). Poverty, early care and stress reactivity in adolescence:
Findings from a prospective, longitudinal study in South Africa. Development and
Psychopathology, 29(2), 449–464. https://doi.org/10.1017/S0954579417000104
Feldman, R., Vengrober, A., Eidelman-Rothman, M., & Zagoory-Sharon, O. (2013).

Stress reactivity in war-exposed young children with and without posttraumatic
stress disorder: Relations to maternal stress hormones, parenting, and child
emotionality and regulation. Development and Psychopathology, 25(4pt1), 943–
955. https://doi.org/10.1017/S0954579413000291
https://doi.org/10.1016/S0749-3797(98)00017-8
127
Fisher, P. A., Kim, H. K., Bruce, J., & Pears, K. C. (2012). Cumulative effects of prenatal
substance exposure and early adversity on foster children’s HPA-axis reactivity
during a psychosocial stressor. International Journal of Behavioral Development,
36(1), 29–35. https://doi.org/10.1177/0165025411406863
Flanagan, J. C., Baker, N. L., McRae-Clark, A. L., Brady, K. T., & Moran-Santa Maria,
M. M. (2015). Effects of adverse childhood experiences on the association
between intranasal oxytocin and social stress reactivity among individuals with
cocaine dependence. Psychiatry Research, 229(1–2), 94–100.
https://doi.org/10.1016/j.psychres.2015.07.064
Gola, H., Engler, H., Schauer, M., Adenauer, H., Riether, C., Kolassa, S., Elbert, T., &
Kolassa, I.-T. (2012). Victims of rape show increased cortisol responses to trauma
reminders: A study in individuals with war- and torture-related PTSD.
Green, J. G., McLaughlin, K. A., Berglund, P. A., Gruber, M. J., Sampson, N. A.,
Zaslavsky, A. M., & Kessler, R. C. (2010). Childhood Adversities and Adult
Psychiatric Disorders in the National Comorbidity Survey Replication I:
Associations With First Onset of DSM-IV Disorders. Archives of General
Psychiatry, 67(2), 113–123. https://doi.org/10.1001/archgenpsychiatry.2009.186
Grimm, S., Pestke, K., Feeser, M., Aust, S., Weigand, A., Wang, J., Wingenfeld, K.,
Pruessner, J. C., La Marca, R., Böker, H., & Bajbouj, M. (2014). Early life stress
modulates oxytocin effects on limbic system during acute psychosocial stress.
Social Cognitive and Affective Neuroscience, 9(11), 1828–1835.
https://doi.org/10.1093/scan/nsu020
Guilliams, T. G., & Edwards, L. (2010). Chronic stress and the HPA axis. The
standard, 9(2), 1-12.
128
Gunnar, M. R., Frenn, K., Wewerka, S. S., & Van Ryzin, M. J. (2009). Moderate versus
severe early life stress: Associations with stress reactivity and regulation in 10–
12-year-old children. Psychoneuroendocrinology, 34(1), 62–75.
Hackman, D. A., Betancourt, L. M., Brodsky, N. L., Hurt, H., & Farah, M. J. (2012).
Neighborhood disadvantage and adolescent stress reactivity. Frontiers in Human
Neuroscience, 6. https://doi.org/10.3389/fnhum.2012.00277
Hagan, M. J., Roubinov, D. S., Purdom Marreiro, C. L., & Luecken, L. J. (2014).
Childhood interparental conflict and HPA axis activity in young adulthood:
Examining nonlinear relations: Childhood Interparental Conflict and the HPA
Axis. Developmental Psychobiology, 56(4), 871–880.
https://doi.org/10.1002/dev.21157
Harkness, K. L., Stewart, J. G., & Wynne-Edwards, K. E. (2011). Cortisol reactivity to

social stress in adolescents: Role of depression severity and child maltreatment.
Heim, C., Newport, D. J., Mletzko, T., Miller, A. H., & Nemeroff, C. B. (2008). The link
between childhood trauma and depression: insights from HPA axis studies in
humans. Psychoneuroendocrinology, 33(6), 693-710.
Hengesch, X., Elwenspoek, M. M. C., Schaan, V. K., Larra, M. F., Finke, J. B., Zhang,
X., Bachmann, P., Turner, J. D., Vögele, C., Muller, C. P., & Schächinger, H.
(2018). Blunted endocrine response to a combined physical-cognitive stressor in
adults with early life adversity. Child Abuse & Neglect, 85, 137–144.
https://doi.org/10.1016/j.chiabu.2018.04.002
Hertzman, C. (1999). The Biological Embedding of Early Experience and Its Effects on
Health in Adulthood. Annals of the New York Academy of Sciences, 896(1), 85–
95. https://doi.org/10.1111/j.1749-6632.1999.tb08107.x
129
Hostinar, C. E., Johnson, A. E., & Gunnar, M. R. (2015). Early social deprivation and the
social buffering of cortisol stress responses in late childhood: An experimental
study. Developmental Psychology, 51(11), 1597–1608.
https://doi.org/10.1037/dev0000029
Ivanov, I., Yehuda, R., Greenblatt, E., Davidow, J., Makotkine, I., Alfi, L., & Newcorn, J.
H. (2011). The effect of trauma on stress reactivity in aggressive youth.
Psychiatry Research, 189(3), 396–402.
https://doi.org/10.1016/j.psychres.2011.05.046
Jaffee, S. R., McFarquhar, T., Stevens, S., Ouellet-Morin, I., Melhuish, E., & Belsky, J.
(2015). Interactive effects of early and recent exposure to stressful contexts on
cortisol reactivity in middle childhood. Journal of Child Psychology and
Psychiatry, 56(2), 138–146. https://doi.org/10.1111/jcpp.12287
Joos, C. M., McDonald, A., & Wadsworth, M. E. (2019). Extending the toxic stress
model into adolescence: Profiles of cortisol reactivity.
Psychoneuroendocrinology, 107, 46–58.
Kempke, S., Luyten, P., De Coninck, S., Van Houdenhove, B., Mayes, L. C., & Claes, S.
(2015). Effects of early childhood trauma on hypothalamic–pituitary–adrenal
(HPA) axis function in patients with Chronic Fatigue Syndrome.
Kern, S., & Laurent, H. K. (2019). Childhood abuse predicts affective symptoms via
HPA reactivity during mother-infant stress. Psychoneuroendocrinology, 107, 19–
25. https://doi.org/10.1016/j.psyneuen.2019.04.023
Kirschbaum, Clemens, Pirke, K.-M., & Hellhammer, D. (1993). The ‘Trier Social Stress
Test’ – A Tool for Investigating Psychobiological Stress Responses in a
Laboratory Setting. Neuropsychobiology, 28, 76–81.
https://doi.org/10.1159/000119004
130
Kliewer, W. (2016). Victimization and Biological Stress Responses in Urban

Adolescents: Emotion Regulation as a Moderator. Journal of Youth and
Adolescence, 45(9), 1812–1823. https://doi.org/10.1007/s10964-015-0398-6
Koss, K. J., George, M. R. W., Davies, P. T., Cicchetti, D., Cummings, E. M., & Sturge-
Apple, M. L. (2013). Patterns of Children’s Adrenocortical Reactivity to
Interparental Conflict and Associations with Child Adjustment: A Growth
Mixture Modeling Approach. Developmental Psychology, 49(2).
https://doi.org/10.1037/a0028246
Koss, K. J., Mliner, S. B., Donzella, B., & Gunnar, M. R. (2016). Early adversity,
hypocortisolism, and behavior problems at school entry: A study of
internationally adopted children. Psychoneuroendocrinology, 66, 31–38.
Krupnik, V., (2019). Trauma or adversity? Traumatology: An International Journal,

25(4), 256–261. http://dx.doi.org.proxy1.lib.uwo.ca/10.1037/trm0000169
Kuhlman, K. R., Vargas, I., Geiss, E. G., & Lopez-Duran, N. L. (2015). Age of Trauma
Onset and HPA Axis Dysregulation Among Trauma-Exposed Youth: Age of First
Trauma and HPA Axis Functioning. Journal of Traumatic Stress, 28(6), 572–579.
https://doi.org/10.1002/jts.22054
Laceulle, O., Nederhof, E., Aken, M., & Ormel, J. (2017). Adversity-driven changes in
hypothalamic-pituitary-adrenal axis functioning during adolescence. The trails
study. Psychoneuroendocrinology, 85.
1662. DOI: 10.1126/science.277.5332.1659
131
Lovallo, W. R., Cohoon, A. J., Acheson, A., Sorocco, K. H., & Vincent, A. S. (2019).
Blunted stress reactivity reveals vulnerability to early life adversity in young
adults with a family history of alcoholism. Addiction (Abingdon, England),
114(5), 798–806. https://doi.org/10.1111/add.14501
Lovallo, W. R., Enoch, M.-A., Sorocco, K. H., Vincent, A. S., Acheson, A., Cohoon, A.
J., Hodgkinson, C. A., & Goldman, D. (2017). Joint Impact of Early Life
Adversity and COMT Val158Met (rs4680) Genotypes on the Adult Cortisol
Response to Psychological Stress: Psychosomatic Medicine, 79(6), 631–637.
https://doi.org/10.1097/PSY.0000000000000481
Lovallo, W. R., Farag, N. H., Sorocco, K. H., Cohoon, A. J., & Vincent, A. S. (2012).
Lifetime Adversity Leads to Blunted Stress Axis Reactivity: Studies from the
Oklahoma Family Health Patterns Project. Biological Psychiatry, 71(4), 344–349.
Lucas-Thompson, R. G., Lunkenheimer, E. S., & Granger, D. A. (2017). Adolescent

Conflict Appraisals Moderate the Link Between Marital Conflict and
Physiological Stress Reactivity. Journal of Research on Adolescence, 27(1), 173–
188. https://doi.org/10.1111/jora.12264
Luecken, L. (2000). Attachment and loss experiences during childhood are associated
with adult hostility, depression, and social support. Journal of Psychosomatic
Research, 49, 85–91. https://doi.org/10.1016/S0022-3999(00)00151-3
Luecken, L. J. (1998). Childhood Attachment and Loss Experiences Affect Adult

Cardiovascular and Cortisol Function: Psychosomatic Medicine, 60(6), 765–772.
https://doi.org/10.1097/00006842-199811000-00021
Luecken, L. J., & Appelhans, B. M. (2006). Early parental loss and salivary cortisol in
young adulthood: The moderating role of family environment. Development and
Psychopathology, 18(01). https://doi.org/10.1017/S0954579406060160
132
MacMillan, H. L., Georgiades, K., Duku, E. K., Shea, A., Steiner, M., Niec, A., Tanaka,
M., Gensey, S., Spree, S., Vella, E., Walsh, C. A., De Bellis, M. D., Van der
Meulen, J., Boyle, M. H., & Schmidt, L. A. (2009). Cortisol Response to Stress in
Female Youths Exposed to Childhood Maltreatment: Results of the Youth Mood
Project. Biological Psychiatry, 66(1), 62–68.
Martinson, A., Craner, J., & Sigmon, S. (2016). Differences in HPA axis reactivity to
intimacy in women with and without histories of sexual trauma.
Matheson, K., Jorden, S., & Anisman, H. (2008). Relations Between Trauma Experiences
and Psychological, Physical and Neuroendocrine Functioning Among Somali
Refugees: Mediating Role of Coping with Acculturation Stressors. Journal of
Immigrant and Minority Health, 10(4), 291–304. https://doi.org/10.1007/s10903-
007-9086-2
McEwen, B. S. (1998). Stress, Adaptation, and Disease: Allostasis and Allostatic Load.
Annals of the New York Academy of Sciences, 840(1), 33–44.
https://doi.org/10.1111/j.1749-6632.1998.tb09546.x
McEwen, B. S., & Stellar, E. (1993). Stress and the Individual: Mechanisms Leading to
Disease. Archives of Internal Medicine, 153(18), 2093–2101.
https://doi.org/10.1001/archinte.1993.00410180039004
McEwen, B. S., & Wingfield, J. C. (2003). The concept of allostasis in biology and
biomedicine. Hormones and Behavior, 43(1), 2–15.
https://doi.org/10.1016/S0018-506X(02)00024-7
McFadyen-Ketchum, L. S., Hurwich-Reiss, E., Stiles, A. A., Mendoza, M. M., Badanes,

L. S., Dmitrieva, J., & Watamura, S. E. (2016). Self-regulation and economic
stress in children of Hispanic immigrants and their peers: Better regulation at a
cost?. Early education and development, 27(7), 914-931.
133
McLaughlin, K. A., Sheridan, M. A., Tibu, F., Fox, N. A., Zeanah, C. H., & Nelson, C.
A. (2015). Causal effects of the early caregiving environment on development of
stress response systems in children. Proceedings of the National Academy of
Sciences, 112(18), 5637–5642. https://doi.org/10.1073/pnas.1423363112
Meaney, M. J. (2010). Epigenetics and the biological definition of gene× environment

interactions. Child development, 81(1), 41-79. https://doi.org/10.1111/j.1467-
8624.2009.01381.x
Mezuk, B., Lexima, E., Kalesnikava, V., Fleming, J., Montgomery, J., Tuktur, W.,
Winston, J., Perrin, P. B., Green, T., & Wheeler, D. C. (2020). Stress-reactivity as
a contributor to racial and socioeconomic disparities: Rationale and baseline
results from the Richmond Stress and Sugar Study. Psychosomatic Medicine,
Publish Ahead of Print. https://doi.org/10.1097/PSY.0000000000000830
Mielock, A. S., Morris, M. C., & Rao, U. (2017). Patterns of cortisol and alpha-amylase
reactivity to psychosocial stress in maltreated women. Journal of Affective
Disorders, 209, 46–52. https://doi.org/10.1016/j.jad.2016.11.009
Monteleone, A. M., Patriciello, G., Ruzzi, V., Cimino, M., Giorno, C. D., Steardo, L.,
Monteleone, P., & Maj, M. (2018). Deranged emotional and cortisol responses to
a psychosocial stressor in anorexia nervosa women with childhood trauma
exposure: Evidence for a “maltreated ecophenotype”? Journal of Psychiatric
Research, 104, 39–45. https://doi.org/10.1016/j.jpsychires.2018.06.013
Mrug, S., Tyson, A., Turan, B., & Granger, D. A. (2016). Sleep problems predict cortisol
reactivity to stress in urban adolescents. Physiology & Behavior, 155, 95–101.
https://doi.org/10.1016/j.physbeh.2015.12.003
Negriff, S., Saxbe, D. E., & Trickett, P. K. (2015). Childhood maltreatment, pubertal
development, HPA axis functioning, and psychosocial outcomes: An integrative
biopsychosocial model: Maltreatment, Puberty, Cortisol, and Outcomes.
Developmental Psychobiology, 57(8), 984–993. https://doi.org/10.1002/dev.21340
134
Nejad, J. G., & Ghaseminezhad, M. (2016). A Cortisol Study; Facial Hair and Nails.
Journal of Steroids & Hormonal Science, 7(2). https://doi.org/10.4172/2157-
7536.1000177
Nelson, C. A. (2017). Hazards to Early Development: The Biological Embedding of

Early Life Adversity. Neuron, 96(2), 262–266.
https://doi.org/10.1016/j.neuron.2017.09.027
Nemeroff, C. B. (2004). Neurobiological Consequences of Childhood Trauma. 11.
Neupert, S. D., Miller, L. M. S., & Lachman, M. E. (2006). Physiological Reactivity to

Cognitive Stressors: Variations by Age and Socioeconomic Status. The
International Journal of Aging and Human Development, 62(3), 221–235.
https://doi.org/10.2190/17DU-21AA-5HUK-7UFG
Nicolson, N. A. (2008). Measurement of cortisol. In Handbook of physiological research

methods in health psychology (pp. 37–74). Sage Publications, Inc.
https://doi.org/10.4135/9781412976244.n3
Obasi, E. M., Shirtcliff, E. A., Cavanagh, L., Ratliff, K. L., Pittman, D. M., & Brooks, J.
J. (2017). Hypothalamic-Pituitary-Adrenal Reactivity to Acute Stress: An
Investigation into the Roles of Perceived Stress and Family Resources. Prevention
Science, 18(8), 923–931. https://doi.org/10.1007/s11121-017-0759-3
Ouellet-Morin, I., Boivin, M., Dionne, G., Lupien, S. J., Arsenault, L., Barr, R. G.,
Pérusse, D., & Tremblay, R. E. (2008). Variations in Heritability of Cortisol
Reactivity to Stress as a Function of Early Familial Adversity Among 19-Month-
Old Twins. Archives of General Psychiatry, 65(2), 211.
https://doi.org/10.1001/archgenpsychiatry.2007.27
Ouellet-Morin, I., Danese, A., Bowes, L., Shakoor, S., Ambler, A., Pariante, C. M.,
Papadopoulos, A. S., Caspi, A., Moffitt, T. E., & Arseneault, L. (2011). A
Discordant Monozygotic Twin Design Shows Blunted Cortisol Reactivity Among
135
Bullied Children. Journal of the American Academy of Child & Adolescent

Psychiatry, 50(6), 574-582.e3. https://doi.org/10.1016/j.jaac.2011.02.015
Ouellet-Morin, I., Odgers, C. L., Danese, A., Bowes, L., Shakoor, S., Papadopoulos, A.
S., Caspi, A., Moffitt, T. E., & Arseneault, L. (2011). Blunted Cortisol Responses
to Stress Signal Social and Behavioral Problems Among Maltreated/Bullied 12-
Year-Old Children. Biological Psychiatry, 70(11), 1016–1023.
Pagliaccio, D., Luby, J. L., Bogdan, R., Agrawal, A., Gaffrey, M. S., Belden, A. C.,
Botteron, K. N., Harms, M. P., & Barch, D. M. (2014). Stress-System Genes and
Life Stress Predict Cortisol Levels and Amygdala and Hippocampal Volumes in
Children. Neuropsychopharmacology, 39(5), 1245–1253.
https://doi.org/10.1038/npp.2013.327
Peckins, M. K., Dockray, S., Eckenrode, J. L., Heaton, J., & Susman, E. J. (2012). The
Longitudinal Impact of Exposure to Violence on Cortisol Reactivity in
Adolescents. Journal of Adolescent Health, 51(4), 366–372.
https://doi.org/10.1016/j.jadohealth.2012.01.005
Peckins, M. K., Roberts, A. G., Hein, T. C., Hyde, L. W., Mitchell, C., Brooks-Gunn, J.,
McLanahan, S. S., Monk, C. S., & Lopez-Duran, N. L. (2020). Violence exposure
and social deprivation is associated with cortisol reactivity in urban adolescents.
Psychoneuroendocrinology, 111, 104426.
https://doi.org/10.1016/j.psyneuen.2019.104426
Peckins, M. K., Susman, E. J., Negriff, S., Noll, J., & Trickett, P. K. (2015). Cortisol
profiles: A test for adaptive calibration of the stress response system in maltreated
and nonmaltreated youth. Development and Psychopathology, 27(4pt2), 1461–
1470. https://doi.org/10.1017/S0954579415000875
Pesonen, A.-K., Räikkönen, K., Feldt, K., Heinonen, K., Osmond, C., Phillips, D. I. W.,
Barker, D. J. P., Eriksson, J. G., & Kajantie, E. (2010). Childhood separation
experience predicts HPA axis hormonal responses in late adulthood: A natural
136
experiment of World War II. Psychoneuroendocrinology, 35(5), 758–767.

Pierrehumbert, B., Torrisi, R., Glatz, N., Dimitrova, N., Heinrichs, M., & Halfon, O.
(2009). The influence of attachment on perceived stress and cortisol response to
acute stress in women sexually abused in childhood or adolescence.
Quas, J. A., Yim, I. S., Oberlander, T. F., Nordstokke, D., Essex, M. J., Armstrong, J. M.,
Bush, N., Obradović, J., & Boyce, W. T. (2014). The symphonic structure of
childhood stress reactivity: Patterns of sympathetic, parasympathetic, and
adrenocortical responses to psychological challenge. Development and
Psychopathology, 26(4pt1), 963–982.
https://doi.org/10.1017/S0954579414000480
Raffington, L., Prindle, J., Keresztes, A., Binder, J., Heim, C., & Shing, Y. L. (2018).
Blunted cortisol stress reactivity in low–income children relates to lower memory
function. Psychoneuroendocrinology, 90, 110-121.
Raison, C. L., & Miller, A. H. (2003). When not enough is too much: the role of
insufficient glucocorticoid signaling in the pathophysiology of stress-related
disorders. American Journal of Psychiatry, 160(9), 1554-1565.
Rao, U., Hammen, C., Ortiz, L. R., Chen, L.-A., & Poland, R. E. (2008). Effects of Early
and Recent Adverse Experiences on Adrenal Response to Psychosocial Stress in
Depressed Adolescents. Biological Psychiatry, 64(6), 521–526.
Schalinski, I., Elbert, T., Steudte-Schmiedgen, S., & Kirschbaum, C. (2015). The Cortisol
Paradox of Trauma-Related Disorders: Lower Phasic Responses but Higher Tonic
Levels of Cortisol Are Associated with Sexual Abuse in Childhood. PLOS ONE,
10(8), e0136921. https://doi.org/10.1371/journal.pone.0136921
137
Selye, H. (1946). The general adaptation syndrome and the diseases of adaptation. The
journal of clinical endocrinology, 6(2), 117-230.
Selye, H. (1950). Stress and the General Adaptation Syndrome. British Medical Journal,
1(4667), 1383–1392.
Shalev, A., Porta, G., Biernesser, C., Zelazny, J., Walker-Payne, M., Melhem, N., &
Brent, D. (2019). Cortisol response to stress as a predictor for suicidal ideation in
youth. Journal of Affective Disorders, 257, 10–16.
Shenk, C. E., Putnam, F. W., Rausch, J. R., Peugh, J. L., & Noll, J. G. (2014). A
Longitudinal Study of Several Potential Mediators of the Relationship between
Child Maltreatment and PTSD Symptoms. Development and Psychopathology,
26(1), 81–91. https://doi.org/10.1017/S0954579413000916
Simonsohn, U., Nelson, L. D., & Simmons, J. P. (2014a). P-curve: A key to the file-
drawer. Journal of Experimental Psychology: General, 143(2), 534–547.
https://doi.org/10.1037/a0033242
Simonsohn, U., Nelson, L. D., & Simmons, J. P. (2014b). p-Curve and Effect Size:
Correcting for Publication Bias Using Only Significant Results. Perspectives on
Psychological Science, 9(6), 666–681.
https://doi.org/10.1177/1745691614553988
Simonsohn, U., Simmons, J. P., & Nelson, L. D. (2015). Better P-curves: Making P-curve
analysis more robust to errors, fraud, and ambitious P-hacking, a Reply to Ulrich
and Miller (2015). Journal of Experimental Psychology: General, 144(6), 1146–
1152. https://doi.org/10.1037/xge0000104
Sterling, P., Eyer, J., Fisher, S., & Reason, J. (1988). Handbook of life stress, cognition
and health. Allostasis; A new paradigm to explain arousal pathology. New York:
Wiley, 629-649.
138
Sturge-Apple, M. L., Davies, P. T., Cicchetti, D., & Manning, L. G. (2012). Interparental
Violence, Maternal
Emotional Unavailability and Children’s Cortisol Functioning in Family
Contexts. Developmental Psychology, 48(1), 237–249.
https://doi.org/10.1037/a0025419
Sumner, J. A., McLaughlin, K. A., Walsh, K., Sheridan, M. A., & Koenen, K. C. (2014).
CRHR1 genotype and history of maltreatment predict cortisol reactivity to stress
in adolescents. Psychoneuroendocrinology, 43, 71–80.
Suzuki, A., Poon, L., Papadopoulos, A. S., Kumari, V., & Cleare, A. J. (2014). Long term
effects of childhood trauma on cortisol stress reactivity in adulthood and
relationship to the occurrence of depression. Psychoneuroendocrinology, 50, 289–
299. https://doi.org/10.1016/j.psyneuen.2014.09.007
Tackett, J. L., Herzhoff, K., Smack, A. J., Reardon, K. W., & Adam, E. K. (2017). Does
socioeconomic status mediate racial differences in the cortisol response in middle
childhood?. Health Psychology, 36(7), 662.
Timothy, A., Benegal, V., Shankarappa, B., Saxena, S., Jain, S., & Purushottam, M.
(2019). Influence of early adversity on cortisol reactivity, SLC6A4 methylation
and externalizing behavior in children of alcoholics. Progress in Neuro-
Psychopharmacology and Biological Psychiatry, 94, 109649.
https://doi.org/10.1016/j.pnpbp.2019.109649
Tottenham, N., & Sheridan, M. A. (2010). A review of adversity, the amygdala and the
hippocampus: a consideration of developmental timing. Frontiers in human
neuroscience, 3, 68. https://doi.org/10.3389/neuro.09.068.2009
Trautmann, S., Kräplin, A., Dieterich, R., Richter, J., & Muehlhan, M. (2018). The role
of childhood trauma and stress reactivity for increased alcohol craving after
induced psychological trauma: An experimental analogue study.
139
Psychopharmacology, 235(10), 2883–2895. https://doi.org/10.1007/s00213-018-

4979-4
Trickett, P. K., Gordis, E., Peckins, M. K., & Susman, E. J. (2014). Stress Reactivity in
Maltreated and Comparison Male and Female Young Adolescents. Child
Maltreatment, 19(1), 27–37. https://doi.org/10.1177/1077559513520466
Troller-Renfree, S., McLaughlin, K. A., Sheridan, M. A., Nelson, C. A., Zeanah, C. H., &
Fox, N. A. (2017). The beneficial effects of a positive attention bias amongst
children with a history of psychosocial deprivation. Biological Psychology, 122,
110–120. https://doi.org/10.1016/j.biopsycho.2016.04.008
van der Hal-Van Raalte, E. A. M., Bakermans–Kranenburg, M. J., & van IJzendoorn, M.
H. (2008). Diurnal cortisol patterns and stress reactivity in child Holocaust
survivors reaching old age. Aging & Mental Health, 12(5), 630–638.
https://doi.org/10.1080/13607860802343134
Voellmin, A., Winzeler, K., Hug, E., Wilhelm, F. H., Schaefer, V., Gaab, J., La Marca,
R., Pruessner, J. C., & Bader, K. (2015). Blunted endocrine and cardiovascular
reactivity in young healthy women reporting a history of childhood adversity.
Wan, C., Couture-Lalande, M.-È., Lebel, S., & Bielajew, C. (2017). The role of stressful
life events on the cortisol reactivity patterns of breast cancer survivors.
Psychology & Health, 1–17. https://doi.org/10.1080/08870446.2017.1346194
Wirtz, P. H., Ehlert, U., Kottwitz, M. U., La Marca, R., & Semmer, N. K. (2013).
Occupational role stress is associated with higher cortisol reactivity to acute
stress. Journal of Occupational Health Psychology, 18(2), 121–131.
http://dx.doi.org.proxy1.lib.uwo.ca/10.1037/a0031802
140
Wust, S., Wolf, J., Hellhammer, D. H., Federenko, I., Schommer, N., & Kirschbaum, C.
(2000). The cortisol awakening response—Normal values and confounds. Noise
and Health, 2(7), 79.
Yehuda, R., Flory, J., Pratchett, L., Buxbaum, J., Ising, M., & Holsboer, F. (2010).
Putative biological mechanisms for the association between early life adversity
and the subsequent development of PTSD. Psychopharmacology, 212, 405–417.
https://doi.org/10.1007/s00213-010-1969-6
Zwolinski, J. (2008). Biopsychosocial responses to social rejection in targets of relational

aggression. Biological Psychology, 79(2), 260–267.
https://doi.org/10.1016/j.biopsycho.2008.06.006
141
Chapter 3
3 Stress, glucocorticoids, and the mesolimbic dopamine

pathway: How early-life experiences become
biologically embedded into the dopamine system
3.1 Introduction
Early-life experiences that result in persistent changes in physiology and health
are said to have become “biologically embedded” into the organism (Hertzman, 1999).
The adverse childhood experiences (ACE) study is one example of the biological
embedding of negative life experiences. The ACE study demonstrates that the quality of
the early childhood environment has a strong and significant impact on later health
outcomes (Felitti et al., 1998). For example, exposure to childhood abuse significantly
increases the risk of developing various diseases in adulthood, including heart disease,
liver disease, respiratory illnesses, and cancer (Felitti et al., 1998). More recently, in a
nationally representative sample of over 8000 Australian women, participants reporting
four or more childhood adversities had significantly higher rates of poor physical health,
sexually transmitted infections, and poor mental health. Childhood adversities were also
related to severe obesity and smoking status (Loxton et al., 2021). Thus, the relationship
between early-life adversity and negative health outcomes may be explained, at least in
part, by a greater likelihood of engaging in health-risk behaviors (Duffy, McLaughlin, &
Green, 2018).
Indeed, it appears that risk-behaviors involving the dopaminergic system such as,
smoking, drinking alcohol, and eating high-fat, high-sugar foods are at the center of
mechanisms that may explain the link between early-life adversity and later negative
health outcomes (Duffy et al., 2018; Felitti et al., 1998; Garrido, Weiler, & Taussig,
2018; Stevens et al., 2011). For instance, adolescents living within out-of-home care are
more likely to engage in risky sexual behavior and substance use behaviors (Stevens et
al., 2011). Similarly, in a sample of 500 children, ACE scores were predictive of health-
risk behaviors after controlling for age, sex, and minority status (Garrido et al., 2018).
Many of these health-risk behaviors, including substance use, involve alterations within
142
the dopaminergic system and suggest that the mesolimbic dopamine pathway may be
sensitive to the quality of the early-life environment (Gatzke-Kopp, 2011).
Considering the negative health-outcomes associated with exposure to adversity

and their relationship with dysregulation of the mesolimbic dopamine pathway, a deeper
understanding of the underlying biological mechanisms that mediate these relationships is
warranted. The purpose of this review then, is to summarize the rodent literature
demonstrating a link between stressful experiences, the HPA axis, and the mesolimbic
dopamine pathway, with a particular emphasis on how adversity may increase risk for the
development of substance abuse. Additionally, we will conduct a systematic review of
positron emission tomography (PET) imaging studies that explore the relationship
between early-life adversity and changes within the mesolimbic dopamine pathway in
humans. This review will highlight the plasticity of the mesolimbic dopamine pathway to
life-experiences, summarize the underlying mechanism that explains how the HPA axis
may modulate dopaminergic transmission via glucocorticoids, and provide an underlying
theoretical framework for understanding how the severity of life experiences relates to
the directionality of dopaminergic change.
3.1.1 The mesolimbic dopamine pathway: A brief review

The mesolimbic dopamine pathway is composed of a complex circuitry involving
the amygdala, hippocampus, medial prefrontal cortex (mPFC), nucleus accumbens,
ventral tegmental area (VTA) and ventral pallidum (Haber & Knutson, 2010; Pierce &
Kumaresan, 2006). Limbic structures including the amygdala, hippocampus, and mPFC
send glutamatergic projections to the nucleus accumbens. The nucleus accumbens sends
two primary GABAergic outputs, one to the ventral pallidum and another to the
VTA/substantia nigra. Both the VTA and ventral pallidum send GABAergic outputs to
the thalamus. Finally, glutamatergic outputs from the thalamus to the mPFC close the
mesolimbic circuitry. Critically, the flow of information between regions within the
mesolimbic dopamine pathway is heavily modulated by dopaminergic neurotransmission
(Haber & Knutson, 2010; Pierce & Kumaresan, 2006).
143
The mesolimbic dopamine pathway is involved in a variety of functions

including, reinforcement learning, motivation, decision-making, and reward-processing
(Wise, 2004). Seminal work by Phillips and Fibiger (1973) demonstrated the involvement
of the mesolimbic dopamine pathway in reward-processing. Intracranial self-stimulation
was elicited from electrodes located in midbrain dopamine regions. The authors reported
facilitory effects of d- and l-amphetamine on self-stimulation. Taken together their data
support the notion that dopaminergic systems subserve positive reinforcement and
reward-processing (Phillips & Fibiger, 1973).
Further support for the involvement of the mesolimbic dopamine pathway in

reward-processing is garnered from a large body of evidence demonstrating the
involvement of the dopamine system in the reinforcing properties of many drugs of abuse
(Haber & Knutson, 2010; Pierce & Kumaresan, 2006;Wise & Ropmpre, 1989; Wise,
1996; Wise, 2004). For example, contingent presentation of cocaine cues results in
striatal dopamine release (Ito et al., 2002), and dopamine antagonists in the striatum
reduce drug seeking behaviours (Vanderschuren, Di Ciano, & Everitt, 2005). Finally, in
drug-addicted individuals, the mesolimbic dopamine pathway is disrupted as evidenced
by alterations in dopamine release and D2 receptors in the striatum (Leyton & Vezina,
2014; Leyton, 2021; Nader et al., 2006; Volkow, Fowler, & Wang, 2004). Taken together
then, these results demonstrate a key role of the mesolimbic dopamine system in reward-
processing and highlight its involvement in substance use disorders (Nader et al., 2002;
Nader et al., 2006; Pierce & Kumaresan, 2006). Indeed, there is abundant evidence from
both animal literature and human epidemiological investigations that stress and early-life
experiences are related to substance use via modulation of the mesolimbic dopamine
pathway (See Levis, Baram, & Mahler, 2021 for a recent review).
3.1.2 Stress and drugs: Animal studies

The animal literature has demonstrated that stressful experiences exert long-term
changes within the mesolimbic dopamine pathway and increase the likelihood of drug
administration (Haney et al., 1995; LeSage, Stafford, & Glowa, 1999; Piazza and Le
Moal 1998; Schenk et al., 1987; Sinha, 2000). Early work with rodents showed that
isolated rats more readily acquired cocaine self-administration as compared to rats
144
housed in groups (Schenk et al., 1987). Other researchers however, failed to replicate this
finding and report that isolate versus group housing did not result in significant
differences in cocaine self-administration (Boyle et al., 1991; Phillips et al., 1994).
Nevertheless, it appears that various forms of stressors are indeed related to drug self-
administration. For example, rodents that experienced social stress in the form of
aggressive attack by a same-sex opponent self-administered more cocaine as compared to
controls (Haney et al., 1995). Physical stressors in the form of footshock induced
reinstatement of heroin and cocaine seeking after prolonged drug-free periods (Erb,
Shaham, & Steward, 1996; Shaham, Erb, & Stewart, 2000). Finally, in rodents, early-life
adversity in the form of maternal deprivation resulted in hypersensitivity to the effects of
morphine (Vazquez et al., 2005). In animal studies using primates, Morgan et al. (2002)
reported that cocaine was a reinforcer in subordinate but not dominant monkeys. Taken
together then, the association between stress and drug self-administration demonstrates
that environmental factors influence individual differences in vulnerability to drug self-
administration (Levis et al., 2021; Piazza & Le Moal, 1996; Piazza & Le Moal, 1998;
Schenk et al., 1987; Shaham et al., 2000).
Critically however, the type and controllability of the stressor appears to produce
differing results. For example, in one study, physical stress was induced by either a hot
plate or by repeated footshocks and emotional stress was induced by forcing rats to
witness another rat being subjected to repeated footshocks. Interestingly, emotional but
not physical stress was associated with an increased rate of cocaine self-administration,
and the authors argue that the uncontrollability and unpredictability of the emotional
stressor contribute to its impact on drug self-administration (Ramsey & Van Ree, 1993).
Similarly, Goeders and Guerin (1994) report that rats receiving response-independent
footshock acquire cocaine self-administration at higher rates as compared to rats
receiving the same number of footshocks with some degree of control. These results
highlight the importance of taking into consideration the type and controllability of
stressors when examining the relationship between stress and drug self-administration.
145
3.1.3 Stress and drugs: Human studies

In humans, the link between exposure to stress and the development of substance
abuse disorders mirrors findings reported in the animal literature. Empirical and
epidemiological evidence confirm that both acute stressors as well as adverse-life
experiences increase the risk for subsequent substance abuse disorders. For example,
acute exposure to a stress imagery task increases cravings for cocaine and alcohol in
cocaine abusers (Sinha, Catapano, O’Malley, 1999; Sinha et al., 2000; Sinha et al., 2003).
Moderately stressful life experiences, such as family disruptions and unemployment, also
have a significant impact on substance use problems. An early meta-analysis reported
that high levels of stress and unemployment were significant predictors of drug abuse in
opiate addicts (Brewer et al., 1998). More recent studies have replicated these findings
(Karlsgodt, Lukas, & Elman, 2003; Low et al., 2012; Sussman & Dent, 2000); for
example, in a sample of over 1000 adolescents, romantic break up stress and family
disruptions were predictors of substance use (Low et al., 2012).
In addition to acute and moderate stressors, previous exposure to traumatic life

experiences are also linked to heightened risk for the development of substance use
disorders. Numerous studies have reported that exposure to traumatic life events,
including physical and sexual abuse, increase the risk for drug abuse (Clark et al., 2001;
Cisler et al., 2011; Enoch, 2011; Keyes, Hatzenbuehler, & Hastin 2011; Kosten, 1986;
Kuksis et al., 2017; Liebschutz, Saitz, & Lloyd-Travaglini, 2002; Mandavia et al., 2016).
For instance, in a nationally representative sample of adolescents, exposure to traumatic
events was related to increases in both PTSD symptoms and binge drinking (Cisler et al.,
2011). In another study, the severity of addiction was positively correlated with the
severity of life trauma as well was with the number of stressful life experiences (Kuksis
et al., 2017). Finally, a review of epidemiological studies investing the relationship
between various stressors and alcohol consumption reported that exposure to terrorism
and disasters, childhood maltreatment, stressful life events, and discrimination were all
associated with alcohol use and alcohol use disorders (Keyes et al., 2011).
Data obtained from individuals seeking addiction treatment is perhaps the most
compelling evidence to emphasize the link between adverse life-experiences and
146
subsequent drug abuse in humans (Cole, Sprang, & Silman, 2019; Ouimette et al., 2000;
Pirard et al., 2005; Rice et al., 2001). Across various observations, a high percentage of
individuals seeking treatment for substance use report having experienced previous
trauma. For example, in a sample of adolescents seeking treatment for substance abuse,
71.5% reported previous trauma exposure (Cole et al., 2019). The percentage of
individuals reporting previous trauma or abuse varies across studies; however, overall it
appears that approximately 60% of all women and at least 25% of all men entering
addiction treatment centers report having experienced previous physical or sexual abuse
(Pirard et al., 2005). Overall then, we can be fairly confident that stress and early-life
adversity increase the risk for the development of substance use disorders. Importantly,
substance abuse involves alterations within the dopaminergic system and this suggests
that the mesolimbic dopamine pathway may be sensitive to stressful experiences (Gatzke-
Kopp, 2011). To better understand how life-experiences and stress have lasting
consequences for the function of the mesolimbic dopamine pathway however, an
investigation of underlying mechanisms that may mediate the link between stress and the
dopamine system must be explored.
3.1.4 Mechanisms of action: The link between the HPA axis and
the mesolimbic dopamine pathway
Activation of the stress response systems, including the hypothalamic-pituitary-
adrenal (HPA) axis, facilitate adaptive physiological and behavioural changes that
increase the chances of survival in the face of stress or threat (Smith & Vale, 2006). The
physiological response of the HPA axis involves increased heart rate, respiration, and
metabolism. Critically, activation of the HPA axis ultimately results in the production of
glucocorticoids—cortisol in humans and corticosterone in animals (Del Rey et al., 2008).
Glucocorticoid production results in lipolysis and glyconeogenesis which increase the
availability of glucose (Del Rey et al., 2008; Marinelli & Piazza, 2002), and increased
glucose availability in the face of stressors is adaptive in that it enables sufficient energy
resources to the organism for “fight or flight” responses to stress (Del Rey et al., 2008).
Notably, stress, glucocorticoids, and the mesolimbic dopamine pathway have long
been proposed to form a pathophysiological chain that ultimately increases risk for the
147
development of addictions (Piazza & Le Moal, 1996). Based on animal models, the
hypothesis was formed that elevated corticosterone levels produced by stress are related
to heightened activity of the mesolimbic dopamine pathway. By extension, alterations
within the mesolimbic dopamine pathway increase vulnerability for the development of
substance abuse problem (Piazza & Le Moal,1996). The relationship between the HPA
axis and the mesolimbic dopamine pathway is of great importance as it provides a
potential mechanism that explains stress-induced plasticity within the mesolimbic
dopamine pathway.
3.1.5 Mechanisms of action: The role of glucocorticoids

The connection between stress, the HPA axis and the mesolimbic dopamine
pathway is evidenced by reports that glucocorticoids influence behaviours mediated by
dopaminergic neurotransmission (Campbell & Carroll, 2001; Goeders & Clampitt, 2002;
See Marinelli & Piazza, 2002 for a review). For example, in individuals seeking
treatment for cocaine dependence, exposure to stress and drug cues increased cortisol
levels and cocaine cravings (Sinha et al., 2003). Furthermore, administration of
mifepristone—a glucocorticoid receptor antagonist, reduced ethanol intake in rodents
(Fahlke et al., 1994; Koenig & Olive, 2004). Indeed, a review of the animal literature
concluded that glucocorticoid administration potentiates behavioural responses to
psychostimulants as well as drug self-administration (Marinelli & Piazza, 2002). Overall
then, the HPA axis and glucocorticoids in particular, are related to behavioural responses
mediated by the mesolimbic dopamine pathway; however, an investigation of the
underlying biological mechanisms explaining the link between the HPA axis,
glucocorticoids, and the dopamine pathway is required.
Evidence that a substantial number of dopamine neurons contain glucocorticoid

receptors provides a mechanism that explains the link between stress, the HPA axis, and
the mesolimbic dopamine pathway (Harfstrand et al., 1986; Hensleigh & Pritchard,
2013). In fact, in an early report, glucocorticoid receptor-immunoreactive cells were
found in up to 75% of the total population of midbrain dopaminergic neurons (Harfstrand
et al., 1986). More recently, using fluorescence immunohistochemistry in rodents,
glucocorticoid receptors were found in 47% of dopaminergic neurons (Hensleigh &
148
Pritchard, 2013). Importantly, evidence of glucocorticoid expression in dopamine

neurons indicates a mechanism for stress-induced plasticity with the mesolimbic
dopamine pathway (Hensleigh & Pritchard, 2013), and provides support for the
hypothesis that the dopamine pathway is indeed sensitive to environmental signals
(Gatzke-Kopp, 2011).
While the presence of glucocorticoid receptors on dopamine neurons suggests the

mesolimbic dopamine pathway may be sensitive to stress hormones, it does not clarify
whether glucocorticoids potentiate or inhibit dopaminergic activity. Overall, evidence
demonstrates that glucocorticoid release potentiates dopaminergic activity. For example,
in rodents, glucocorticoids have stimulant effects on mesolimbic dopamine neurons; for
instance, administration of corticosterone increased extra-cellular dopamine levels (Butts
et al., 2011; Piazza et al., 1996b). Additionally, activation of glucocorticoid receptors by
in vivo injection of dexamethasone potentiated the synaptic strength of glutamatergic
synapses in VTA dopamine neurons (Daftary et al., 2009). The authors conclude that
stress-induced potentiation of VTA dopamine synaptic transmission facilitates cellular
adaptations within the mesolimbic dopamine system (Daftary et al., 2009).
To further demonstrate the potentiating effect of glucocorticoids on dopamine

transmission, studies that block the function of glucocorticoids through a variety of
methods report blunted dopamine release. For example, while administration of
corticosterone in rat prefrontal cortex increased dopamine release, blockade of
glucocorticoid receptors in the prefrontal cortex reduced stress-induced dopamine release
in this region (Butts et al., 2011). Moreover, inhibiting glucocorticoid receptors reduced
locomotor responses to morphine (Marinelli et al., 1998), as well as morphine-induced
and dexamethasone-induced release of dopamine (Daftary et al., 2009; Marinelli et al.,
1998). Finally, suppression of glucocorticoid secretion via adrenalectomy decreased both
basal levels of dopamine as well as dopamine release in the nucleus accumbens by over
50% (Piazza et al., 1996a). Critically, the effects of adrenalectomy were reversed with the
administration of glucocorticoids within the physiological range (Piazza et al., 1996a).
These studies suggest that glucocorticoids facilitate dopamine-dependent behavioural
responses via potentiating dopamine release (Marinelli et al., 1998; Marinelli & Piazza,
149
2002). Taken together, the observation that dopamine neurons express glucocorticoid
receptors, and that glucocorticoids can increase dopaminergic transmission, emphasize
that stressful experiences may impact the mesolimbic dopamine pathway via their effect
on the HPA axis and glucocorticoid secretion.
3.1.6 Stress and dopamine: Animal literature

Much like the HPA axis then, the mesolimbic dopamine pathway is hypothesized
to exhibit plasticity to a wide range of environmental signals (Gatzke-Kopp, 2011).
Studies conducted with rodents confirm that exposure to both acute and chronic stressors
are related to variation in dopamine levels within regions of the mesolimbic pathway (See
Douma & de Kloet, 2020; Holly & Miczek, 2016 for reviews). Critically, acute stressors
seem to be related to heightened dopamine release, whereas chronic stressors are
associated with blunted dopamine levels and even loss of dopaminergic neurons.
Regardless of the direction of dopaminergic modulation however, the observation that the
mesolimbic pathway is indeed impacted by exposure to acute and chronic stressors
demonstrates its sensitivity to environmental signals.
In the case of acute stressors, in vivo microdialysis studies with rodents showed
that dopamine levels in the prefrontal cortex and nucleus accumbens were elevated by
130-160% of baseline in response to threat and social defeat (Tidey & Miczek, 1996;
Miczek et al., 2011). Consistent with these findings, in a study using fast-scan cyclic
voltammetry and multiunit recordings, aggressive confrontations in defeated rats were
associated with increases in phasic dopamine transmission in the mesolimbic pathway
(Anstrom, Miczek, & Budygin, 2009). Furthermore, in vivo administration of various
drugs as well as acute stress have both been shown to increase the strength of excitatory
synapses in the mesolimbic dopamine pathway (Saal et al., 2003). In contrast to these
findings, one report indicated that a single acute stressor significantly decreased
dopamine levels in the striatum; however, it is noteworthy to mention that the acute
stressor used in the experiment was quite severe and involved a single immobilization for
150 minutes (Rasheed et al., 2010). Overall then, it appears that acute stressors are related
to heightened dopaminergic release in mesolimbic regions (Abercrombie et al., 1989;
150
Valenti, Lodge, & Grace, 2011; See Douma & de Kloet, 2020; Holly & Miczek, 2016 for
reviews).
In contrast to acute stressors, chronic stressors seem to be associated with blunted

dopamine levels (Enman et al., 2015; Gambarana et al., 1999; Imperato et al., 1992;
Mangiavacchi et al., 2001; Miczek et al., 2011; Pascucci et al., 2007; Puglisi-Alegra et
al., 1991; Shimamoto et al., 2011). For example, while acute social threat increases
dopamine levels in the nucleus accumbens, chronic social defeat is related to blunted
dopamine levels (Miczek et al., 2011; Shimamoto et al., 2011). Similarly, chronic
restraint stress results in reduced basal dopamine concentrations in the nucleus
accumbens (Mangiavacchi et al., 2001). To emphasize the difference in dopamine levels
after acute versus prolonged stress, Pascucci et al. (2007) reported that rats experiencing
restraint stress for 240 minutes showed an initial increase in dopamine in the nucleus
accumbens; however, the initial rise in dopamine was followed by a decrease below
resting levels as the stressor continued. That said, in two reports, chronic stressors in the
form of isolate rearing and maternal separation result in increased dopamine responses to
d-amphetamine (Hall et al., 1998) and stress (Meaney, Brake, & Gratton, 2002). Overall
however, these findings demonstrate that chronic stress is related to reduced dopamine
levels (See Douma & de Kloet, 2020; Holly & Miczek, 2016 for reviews).
The finding that stressful experiences are related to dopamine levels within the
mesolimbic pathway confirms that environmental factors and life-experiences have an
enduring impact on the function of the dopamine system. In fact, exposure to chronic
restraint stress in rats resulted in significant loss of dopaminergic neurons in the
mesolimbic pathway in a time-dependent manner. Chronic restraint stress (8 h/day, 5
days/week) resulted in a loss of 57% of dopaminergic neurons in the VTA and 61% of
dopaminergic neurons in the substantia nigra pars compacta by the 16th week of stress
(Sugama & Kakinuma, 2016; Sugama et al., 2016; See Goldstein & Kopin, 2018 for a
review). Evidence of neuronal loss within dopaminergic regions as a function of exposure
to chronic stress demonstrates that life-experiences become biologically embedded within
the mesolimbic dopamine pathway (Smith, Castro, & Zigmond, 2002). Moreover, the
links between exposure to stressful life-experiences, glucocorticoids, and dopaminergic
151
transmission demonstrate that the plasticity of the mesolimbic dopamine system to

stressful life-events is at least partially mediated by HPA axis function.
3.1.7 The link between the HPA axis and the mesolimbic
dopamine pathway in humans
While in rodents, the data overwhelmingly suggests that glucocorticoid release
plays a role in stress-induced vulnerability to addiction (Campbell & Carroll, 2001;
Goeders & Clampitt, 2002; Piazza et al., 1991; Piazza et al., 1996). In humans, the
evidence connecting the HPA axis, the mesolimbic dopamine pathway, and dopamine-
mediated behaviours is mixed. For example, a study of cocaine addicted individuals
reported blunted levels of the catecholamine metabolite homovanillic acid (HVA), and
higher basal cortisol levels as compared to controls (Gerra et al., 2009). Importantly,
these changes were characteristic of cocaine addicts with a history of childhood neglect,
suggesting that childhood adversities contribute to dysregulation within both the HPA
and mesolimbic dopamine pathways and may confer risk for the development of
substance use disorders (Gerra et al., 2009). In contrast, other studies with human
participants reported that cortisol levels were not related to the reinforcing effects of d-
amphetamine (Alessi et al., 2003; Wachtel et al., 2001). Similarly, Harris et al. (2003)
reported that in human participants, pharmacological blockade of glucocorticoids did not
alter the pleasurable effects of methamphetamine (Harris et al., 2003). The authors
conclude that results obtained from human studies do not mirror those found in the
animal literature. Additionally, PET studies investigating the relationship between
adverse-life experiences and mesolimbic dopamine reactivity have provided conflicting
results, with some studies reporting heightened dopamine levels after exposure to early-
life adversity, and others reported blunted dopamine levels (See Appendix B-1).
Therefore, it appears that while the mesolimbic dopamine pathway is sensitive to signals
of environmental quality in humans, the directionality of dopaminergic profiles after
exposure to adversity is unclear.
152
3.1.8 Theoretical frameworks and models

Gatzke-Kopp (2011) proposed that the mesolimbic dopamine pathway in humans
exhibits sensitivity to indicators of environmental quality; however, the directionality of
dopaminergic change remains inconsistent with evidence present for both
hyperdopaminergic and hypodopaminergic profiles after exposure to adversity. In theory,
changes in dopaminergic profiles are proposed to reflect adaptive changes to adversity,
and the severity and chronicity of adversity is said to influence the direction of change in
dopamine function (Gatzke-Kopp, 2011). According to this framework, more severe or
prolonged forms of adversity result in a hypodopaminergic profile, by extension, a
hypodopaminergic profile results in the individual favoring choices that are least costly in
terms of effort instead of selecting options with the highest reward value (Gatzke-Kopp,
2011). In a related framework, Cabib and Puglisi-Allegra (2012) proposed that
heightened dopamine levels support the expression of active coping strategies against
events appraised as stressors, while blunted dopamine levels support passive coping with
stressful situations that are appraised as unescapable or uncontrollable (Cabib & Puglisi-
Allegra, 2012). Behaviorally, elevated dopamine levels are associated with attempts to
escape or overcome the source of the stressor, whereas blunted dopamine levels are
related to helplessness (Cabib & Puglisi-Allegra, 2012). Indeed, this model explains the
rodent findings that under acutely stressful conditions, dopamine levels are heightened,
whereas under chronic or severely stressful conditions, dopamine levels are blunted.
Moreover, this proposal is supported by a computational model that provides an
explanation for the shift from active to passive coping strategies under acute versus
chronic stress (Fiore et al., 2015).
Evidence from rodent models demonstrating that acute versus chronic stressors
result in differing dopaminergic profiles within the mesolimbic pathway, emphasizes the
need to take the type and timing of adversity into consideration. Typically, chronic
stressors are more severe than acute stressors; therefore, perhaps variation across studies
in the severity of adversity experienced by participants may partially explain the
inconsistent findings in the human literature. In a recent review and p-curve meta-
analysis that explored the link between early-life adversity and HPA axis reactivity, we
153
found that moderately adverse life events were associated with cortisol hyper-reactivity,
while traumatic experiences were associated with cortisol hypo-reactivity (Hosseini-
Kamkar et al., 2021). These results underscore the importance of differentiating between
moderate adversity (e.g., low SES, work-related stress) and severe trauma (sexual abuse,
exposure to war-related violence) when investing the human literature (Hosseini-Kamkar
et al., 2021, Krupnik, 2019).
3.1.9 Stress-response Adaptations to Variability in Early-life

experiences (SAVE) model
To elaborate on the adaptive function of the HPA axis and the mesolimbic
pathway in the face of stressors, we are proposing the Stress-response Adaptations to
Variability in Early-life experiences (SAVE) model. Based on the established link
between the HPA axis and the mesolimbic dopamine pathway, and the theoretical model
proposed by Cabib and Puglisi-Allegra (2012), the SAVE model predicts an inverted-U
shaped relationship between severity of adversity and both HPA axis and dopamine
reactivity. The model predicts that there will be blunted HPA and dopamine reactivity in
individuals exposed to traumatic life experiences, while exposure to moderate adversities
will be related to HPA axis and dopamine hyper-reactivity. The SAVE model is based on
the biological role that glucocorticoid release plays in glucose metabolism, and the
tendency of the body to “save” resources unless it is necessary and efficient to utilize
them. When exposed to stressful experiences, glucocorticoids are produced and result in
increased availability of energy resources to facilitate “fight or flight” responses (Del Rey
et al, 2008). Under chronically stressful or traumatic conditions, “fight or flight” may not
be feasible, and thus, the organism’s physiology may adapt in a manner that does not
waste energy resources. If the organism perceives the situation as completely
inescapable, there is little utility in releasing glucocorticoids and mobilizing energy
resources unnecessarily (Cabib & Puglisi-Allegra, 2012). On the other hand, if an
organism faces moderate levels of adversity throughout its lifetime, the organism will
have likely learned that it is possible to “fight or flee” those moderately stressful
conditions, in which case it is adaptive for the organism to release glucocorticoids and
mobilize energy resources to facilitate overcoming the stressful situation. Therefore,
154
under the SAVE model, exposure to extremely low and high degrees of adversity will
result in blunted biological reactivity to stress as well as blunted dopamine levels;
because when exposed to conditions of low adversity, there is no need to mobilize energy
resources. Likewise, when exposed to severe adversity, the organism may exhibit
helplessness in response to the stressor. In contrast, under conditions of moderate
adversity, it is most adaptive for the organism to release glucocorticoids to mobilize
energy resources; and thus, the model predicts that individuals experiencing moderate
levels of adverse life-experiences will demonstrate biological hyper-reactivity to stress
with heightened HPA axis and dopamine reactivity.
The finding that moderate adversity is related to HPA axis hyper-reactivity, while
severe trauma is associated with HPA axis hypo-reactivity is consistent with the
predictions made by the SAVE model. Moreover, the animal literature exploring the link
between acute versus chronic stressors and dopamine levels is also consistent with the
SAVE model. What remains to be determined however, is whether variation in the
severity of adversity across studies using human participants explains the inconsistent
findings in studies exploring the link between early-life experiences and mesolimbic
dopamine function. Cabib and Puglisi-Allegra (2012) proposed a model under which
uncontrollable stressors are associated with a hypodopaminergic profile. Given that
severe stressors like traumatic experiences are more likely to be uncontrollable relative to
moderately stressful life events, we predict that the severity of adversity will be
differentially related to dopamine levels. The SAVE model expands upon the model
proposed by Cabib and Puglisi-Allegra (2012) by highlighting the link between the HPA
axis and the mesolimbic dopamine pathway and proposing an inverted-U shaped function
to describe the relationship between severity of adversity and both HPA axis reactivity
and mesolimbic dopamine reactivity.
3.1.10 Current study and hypotheses

The human literature that investigates the link between adversity and
dopaminergic function remains inconsistent, with evidence of both dopamine hyper- and
hypo-reactivity as a function of exposure to adversity (See Appendix B-1). We predict
that this inconsistency can be explained, at least partially, by variation in the severity of
155
adversity experienced by participants across studies. To address this, we will conduct a

systematic review of PET studies with human participants that investigate the association
between adversity versus trauma and mesolimbic dopamine function. Based on the SAVE
model, we predict that the experience of moderately adverse life events will be associated
with a hyperdopaminergic profile, whereas severely traumatic experiences will be related
to a hypodopaminergic profile.
3.2 Methods
3.2.1 Data sources and study selection
We conducted a systematic review of human PET imaging studies that
investigated the association between exposure to adverse life-experiences and
dopaminergic function (Pre-registration: https://osf.io/cf2rj). We used three databases,
EMBASE, MEDLINE, and Web of Science to conduct our literature search. Our search
concepts were “adversity”, “dopaminergic function”, and “PET studies”. Our key terms
were as follows: adversity (trauma*OR advers* OR maltreatment OR PTSD OR abuse
OR stress*), dopaminergic function (dopamine OR D1 receptor* OR D2 receptor* OR
dopamine transporter*), PET studies (PET stud* OR Raclopride OR FDOPA). See
Appendix B, Tables 2-4 for search terms entered in each database and the number of
results generated for each search term. A date restriction was used in the search to
exclude any studies that were conducted prior to 2001. Furthermore, only articles in the
English language were included. A “grey literature” search was also conducted using
Google Scholar and references listed in other articles. Overall, our search yielded 1234
results, however, after the removal of 281 duplicates, 953 articles went to abstract and
title review.
3.2.2 Screening and inclusion

We conducted the screening and review of our articles using Rayyan – Intelligent
Systematic Reviews (https://rayyan.ai.). Three reviewers conducted the screening for
inclusion of articles in the systematic review. Reviewer 1 screened all 953 abstracts and
titles for inclusion, Reviewer 2 screened 477 articles, and Reviewer 3 screened the
remaining 476. Therefore, two reviewers (Reviewer 1 and 2, or Reviewer 1 and 3)
156
screened all 953 abstracts. Disagreements between the two reviewers were then resolved
by the third reviewer. The screening for abstracts was conducted blindly, and each
reviewer was unaware of the decisions made by the others. After screening all 953 titles
and abstracts, 27 articles met criteria for full text review. Of these 27 articles, 1 article
was excluded because it was the wrong study type (i.e., review or meta-analysis), 2
articles were excluded because they did not have a measure of trauma or adversity, 2
articles were excluded because the full text for the article was unavailable, and 9 were
excluded because they did not report an effect of adversity/trauma on dopamine function
(See Figure 3.1 for Flowchart). In total then, 13 articles were included in our systematic
review.

1.) Must include empirical studies with human participants
2.) Must include a measure of adversity or trauma.
Traumatic life experiences were defined in accordance with DSM-5 criteria:

Exposure to: death, threatened death, actual or threatened serious injury,
or actual or threatened sexual violence, in the following way(s): direct
exposure; witnessing the trauma; learning that a relative or close friend
was exposed to a trauma; indirect exposure to aversive details of the
trauma, usually in the course of professional duties (e.g., first responders,
medics) (Diagnostic and Statistical Manual of Mental Disorders, Fifth
Edition [DSM–5]; American Psychiatric Association, 2013).
Adversity was defined as negative or stressful life experiences that do not meet
criterion A of DSM-V for the development of trauma-related disorders (e.g., low
SES, family conflict, work-related stress).
3.) Dopaminergic function as assessed via PET studies.
157
Identification
Unique articles identified from Database Searches Excluded in Abstract Review
N = 953 N = 926

or perspectives)
N = 251

Records Identified full text unavailable)
N = 27 N=3

Excluded or perspectives)
Screening
(did not measure trauma

or adversity)
N = 24 N=2
Records Identified
Excluded (did not report effect of
N = 22
trauma/adversity on DA function
N=9
Eligibility
Records Identified
N = 13

Included
N = 13
Figure 3.1: Flowchart - Early-life adversity and dopamine function

158
3.3 Results
At first glance, findings from PET studies in humans investigating the relationship
between adverse life-experiences and dopaminergic function appear mixed in that there is
evidence for both hyper and hypo-dopaminergic profiles after exposure to adverse life-
events (See Appendix B-1 for a summary). It is also unclear whether the severity of
adversity is associated with the direction of dopaminergic change (See Appendix B).
Upon closer investigation of the literature however, it becomes apparent that there is
substantial variability across studies in terms of the type of adversity (i.e., adversity
versus trauma), the radioligand used in the PET study, and whether they measured
dopamine release or dopamine receptor binding potential. For example, some studies
obtained basal measures of dopamine (Sebold et al., 2019; Egerton et al., 2014), while
others measured dopamine reactivity (Bloomfield et al., 2019; Bloomfield et al., 2017;
2019 Casey et al., 2010). Additionally, some studies measured dopamine reactivity to a
psychosocial stress task (Kasanova et al., 2016; Mizrahi et al., 2014; Pruessner et al.,
2014), while others measured dopamine reactivity to psychostimulants (Dahoun et al.,
2019; Oswald et al., 2011; Oswald et al., 2014; Casey et al., 2010). Finally, the majority
of studies measured dopamine release in the striatum, however, two studies measured D2
receptor availability (Sebold et al., 2019; Schifani et al., 2018), and two studies used
prefrontal cortex rather than striatum as a region of interest (Kasanova et al., 2016;
Schifani et al., 2018). Taken together, these observations demonstrate that there is
substantial variability in methodology across studies and these differences may contribute
to the mixed results.
In a previous study investigating the link between adverse life-experiences and

HPA axis reactivity to a psychosocial stressor, we found that adversity was related to
heightened cortisol reactivity, while trauma was associated with blunted cortisol
reactivity (Hosseini-Kamkar et al., 2021). Importantly, we confined our search of the
literature to cortisol reactivity to a psychosocial stressor. Here, if we consider only
studies that measured dopamine reactivity, the results appear more consistent. Moderate
levels of adversity (e.g., low SES, daily stress) are associated heightened dopamine
reactivity (See Table 3.1). Trauma (e.g., sexual and physical abuse) on the other hand,
159
appears to be associated with blunted dopamine reactivity in most studies (See Table 3.1).
Two studies however (Oswald et al., 2011 and Oswald et al., 2014) reported that trauma
was associated with heightened dopamine reactivity. It is important to note however, that
these two studies measured dopamine reactivity to a psychostimulant (AMPH), and
psychostimulants themselves increase dopamine levels. In Table 3.2, we show the results
for PET studies that reported findings on adversity versus trauma and dopamine reactivity
to a psychosocial stress task or to glucocorticoid administration. Confining the literature
to this subset of studies shows that consistent with the literature on cortisol reactivity
(Hosseini-Kamkar et al., 2021), adversity is associated heightened dopamine reactivity,
while trauma is associated with blunted dopamine reactivity (See Table 3.2). Given the
small number of studies included in this systematic review however, these results must be
interpreted with caution.
Table 3.1: Overview of human PET studies of adversity versus trauma and
dopamine reactivity
Source Trauma/Adversity Reactivity/Basal Dopamine Brain Region Hypo/Hyper
Measure
Egerton et al Adversity Psychosocial – MIST DA Release Striatum Hyper
2017 Reactivity
Mizrahi et al Adversity Psychosocial – MIST DA Release Striatum Hyper
2014 Reactivity
Casey et al Adversity Psychostimulant DA Release Dorsal Putamen Hyper
2010 Reactivity- AMPH
Pruessner et al Adversity Psychosocial –MIST DA Release Striatum Hyper
2004 Reactivity
Bloomfield et al Trauma Psychosocial – MIST DA Synthesis Striatum Hypo
2019 Reactivity
Dahoun et al Trauma Dex-amphetamine DA Release Ventral Hypo
2019 Reactivity Striatum
2017 Reactivity
Oswald et al Trauma Psychostimulant DA Release Striatum Hyper
Oswald et al Trauma Psychostimulant DA Release Striatum Hyper
160
Table 3.2: Overview of human PET studies of adversity versus trauma and
dopamine reactivity to MIST or Dex administration
Source Trauma/Adversity Reactivity/Basal Dopamine Brain Region Hypo/Hyper
Measure
Egerton et al Adversity Psychosocial – MIST DA Release Striatum Hyper
2017 Reactivity
Mizrahi et al Adversity Psychosocial – MIST DA Release Striatum Hyper
2014 Reactivity
Pruessner et al Adversity Psychosocial –MIST DA Release Striatum Hyper
2004 Reactivity
2019 Reactivity
Dahoun et al Trauma Dex-amphetamine DA Release Ventral Hypo
2019 Reactivity Striatum
2017 Reactivity
161
3.4 Discussion
Early-life experiences have been shown to become biologically embedded and exert
long-lasting influences on health and well-being (Felliti et al., 1998). One biological
system that is typically studied in association with adversity is the HPA axis given its role
in responding to stressors. A closer investigation of the link between exposure to
adversity and health-related outcomes however, reveals that health-risk behaviours
involving the mesolimbic dopamine pathway are at the center of mechanisms explaining
the relationship between adversity and later health outcomes (Duffy et al., 2018). Indeed,
in rodent models, a pathophysiological link between exposure to stress, glucocorticoid
release, dopaminergic reactivity, and vulnerability to addictions has long been proposed
(Piazza & Le Moal, 1996b). Human PET studies that explore the association between
early-life adversity and mesolimbic dopamine function however, provide conflicting
results—with some studies reporting a hyperdopaminergic profile after exposure to
adversity, and others reporting a hypodopaminergic profile (See Appendix B-1).
This chapter systematically reviewed the human literature on exposure to

adversity and dopaminergic function, and found there was substantial variability across
studies in the type of adversity measured (e.g., moderate adversity versus severe trauma),
the PET radioligand used, and whether the studies measured dopamine release or D2
receptors, as well as the region of interest studied. When the criteria were limited to
exploring the link between adverse life-experiences and dopamine reactivity however, a
relatively consistent pattern of findings begins to emerge. Specifically, consistent with
findings exploring the link between adverse life-experiences and HPA axis reactivity
(Hosseini-Kamkar et al., 2021), moderate levels of adversity are associated with a
hyperdopaminergic profile, while severe forms of trauma are related to a
hypodopaminergic profile (See Tables 3.1 and 3.2). The majority of studies fit this
pattern of results, however, two studies reported that trauma was related to heightened
dopamine reactivity (Oswald et al., 2011; Oswald et al., 2014). Importantly, these two
studies measured dopamine reactivity to psychostimulant (AMPH) administration rather
than psychosocial stressors. The finding that amphetamine-induced dopamine reactivity
was positively associated with trauma is consistent with a report in the animal literature
162
(Hall et al., 1998). In the rodent literature, overall chronic stress was associated with
blunted dopamine levels (See Douma & de Kloet, 2020; Holly & Miczek 2016 for
reviews). However, one study reported heightened dopamine reactivity in response to d-
amphetamine administration after chronic stress in the form of isolate rearing (Hall et al.,
1998). This finding mirrors the human literature demonstrating that in general, trauma is
associated with reduced dopamine reactivity; however, perhaps when dopamine reactivity
is measured in response to psychostimulant administration, the opposite pattern of results
may follow.
In rodents, the differential response of the dopaminergic system to acute versus

chronic stressors is proposed to reflect an adaptive response in the face of controllable
versus uncontrollable stressors (Cabib & Puglishi-Allegra, 2012; Fiore et al., 2015).
Specifically, heightened dopamine levels are proposed to support active coping strategies
to events appraised as controllable, while blunted dopamine levels support passive coping
strategies with stressors that are unescapable or uncontrollable (Cabib & Puglisi-Allegra,
2012). By highlighting the link between the HPA axis and the mesolimbic dopamine
pathway, the SAVE model expands upon the model proposed by Cabib and Puglisi-
Allegra (2012) and proposes an inverted-U shaped function to describe the relationship
between severity of adversity and both HPA axis reactivity and mesolimbic dopamine
reactivity. Based on the SAVE model, we predicted that severely traumatic life-
experiences would be related to blunted HPA axis and dopamine reactivity, in contrast,
moderately adverse life-events would be associated with heightened HPA axis and
dopamine reactivity. The logic behind the SAVE model is based on the biological role
that glucocorticoid release plays in glucose metabolism, and the tendency of the body to
“save” resources unless it is necessary and efficient to utilize them. When exposed to
moderate or controllable stressors, glucocorticoids are produced and result in increased
availability of energy resources to facilitate “fight or flight” responses (Del Rey et al,
2008). Under uncontrollable stressful situations or traumatic conditions however, “fight
or flight” may not be feasible, and thus, the organism’s physiology may adapt in a
manner that does not waste energy resources.
163
Indeed, our previous work examining how moderate adversity versus trauma
relates to HPA axis reactivity provides partial support for the SAVE model (Hosseini-
Kamkar et al., 2021). Moreover, here we review preliminary evidence from human PET
imaging studies investigating how adverse life-experiences are associated dopamine
reactivity and we observed a pattern of results generally consistent with the predictions of
the SAVE model. Overall, it appears that both HPA axis and dopamine reactivity are
blunted in the face of traumatic life-experiences, while they are heightened after exposure
to moderately adverse life experiences. These results underscore the importance of
differentiating between moderately adverse life-experiences and severely traumatic
experiences when probing the human literature (Krupnik, 2019; Hosseini-Kamkar, 2021).
The established relationships between exposure to stress, HPA axis reactivity, and the
mesolimbic dopamine pathway demonstrates the plasticity of the dopamine pathway to
environmental signals (Gatzke-Kopp, 2011). In the human literature however,
inconsistencies remain in terms of whether exposure to early life-adversity is related to a
hyperdopaminergic or hypodopaminergic profile (Gatzke-Kopp, 2011). Here, we suggest
that these mixed findings, can be partially explained by taking into consideration the
severity of adverse life-experiences.
Substantial evidence indicates that substance abuse problems are associated with
dysregulation within the mesolimbic dopamine pathway (Nutt et al., 2015; Volkow et al.,
2004; Schwabe et al., 2011). Studies conducted with human participants report that
substance abuse problems are associated with a hypodopaminergic profile in the form of
diminished dopamine release (Nutt et al., 2015; Volkow et al., 2004; but see Leyton &
Vezina, 2014; Leyton, 2021). Based on this SAVE model and available evidence then, the
prediction can be made that individuals exposed to severely traumatic life events would
exhibit blunted HPA axis and dopamine reactivity, and this hypodopaminergic profile
would increase their vulnerability to developing substance abuse problems. Indeed,
numerous studies have provided evidence that traumatic life-experiences increase the risk
of developing substance abuse problems (Kosten et al., 1986; Keyes et al., 2011; Enoch
et al., 2011; Liebschutz et al., 2002; Clark et al., 2001; Mandavia et al., 2016; Kuksis et
al., 2017; Cisler et al., 2011).
164
While the rodent literature has clearly demonstrated a physiological chain

between exposure to stressful events, glucocorticoid release, dopamine reactivity, and
substance abuse (Piazza & Le Moal., 1996), the research conducted with human
participants is at a preliminary stage. In animal models, acute stress is related to increased
glucocorticoid and dopamine reactivity, while chronic stress is associated with blunted
reactivity. Based on available evidence, it appears that in humans, moderately adverse
life-events are related to heightened dopamine reactivity, while severely traumatic life
events are associated with blunted reactivity. These findings are consistent with reports
investigating the link between adverse-life experiences and HPA- axis reactivity
(Hosseini-Kamkar, 2021), and with available theory (Cabib & Puglisi-Allegra, 2012).
Due to the limited number of studies conducted with humans however, these results must
be interpreted with caution. In this review, only 13 articles met criteria for inclusion, and
from those 13, only 9 measured dopamine reactivity. Therefore, while these results are
intriguing and consistent with animal research and theoretical models, the paucity of
empirical research with humans makes drawing firm conclusions premature. This
warrants the need for further investigation of the link between adverse-life experiences
and dopaminergic function in human participants. Our results to date highlight the
importance of differentiating between moderately adverse life-experiences and severely
traumatic experiences when conducting research with human participants. Furthermore,
based on the available literature, it appears that the type of paradigm used to elicit
dopamine reactivity may result in differing findings. For example, it seems that after
exposure to trauma, dopamine reactivity to psychosocial stressors is blunted, however,
amphetamine-induced dopamine reactivity is heightened (See Table 3.2). These results
demonstrate that variability in the reactivity paradigm used across research groups may
contribute to the mixed findings in the literature.
3.1 Conclusion
Gatzke-Kopp (2011) proposed that much like the HPA axis, the mesolimbic
dopamine pathway demonstrates plasticity to indicators of environmental quality. What
remained unclear however, was whether exposure to adversity is related to a
hypodopaminergic or hyperdopaminergic profile (Gatzke-Kopp, 2011). The links
165
between stress, the HPA axis and the mesolimbic dopamine pathway are well-established
in animal models. Indeed, in rodent models, acute stressors are associated with
heightened dopamine levels, while chronic stressors are related to blunted dopamine
levels. Moreover, the rodent literature demonstrates that the HPA axis and
glucocorticoids modulate dopaminergic transmission. In humans, we previously reported
that moderate adversity was associated with heightened glucocorticoid reactivity, while
severe trauma was related to blunted glucocorticoid reactivity (Hosseini-Kamkar et al.,
2021). Here, we propose the SAVE model, predicting an inverted-U shaped relationship
between severity of adversity and HPA axis and dopamine reactivity. We systematically
reviewed human PET studies linking adverse early-life experiences to dopamine
reactivity. Although preliminary, our results are consistent with the predictions made by
the SAVE model in that moderately adverse life-experiences appear to be related to
heightened dopamine reactivity, while traumatic life-experiences appear to be related to
blunted dopamine reactivity. Given the role of the mesolimbic dopamine pathway in
substance abuse disorders, these findings may partially explain the relationship between
exposure to trauma and substance abuse disorders.
166
3.2 References
Abercrombie, E. D., Keefe, K. A., DiFrischia, D. S., & Zigmond, M. J. (1989).
Differential Effect of Stress on In Vivo Dopamine Release in Striatum, Nucleus
Accumbens, and Medial Frontal Cortex. Journal of Neurochemistry, 52(5), 1655–
Alessi, S. M., Greenwald, M., & Johanson, C.-E. (2003). The prediction of individual
differences in response to D-amphetamine in healthy adults. Behavioural
Pharmacology, 14(1), 19–32.
Anstrom, K. K., Miczek, K. A., & Budygin, E. A. (2009). Increased phasic dopamine
signaling in the mesolimbic pathway during social defeat in rats. Neuroscience,
161(1), 3–12. https://doi.org/10.1016/j.neuroscience.2009.03.023
Bloomfield, M. A. P., McCutcheon, R. A., Kempton, M., Freeman, T. P., & Howes, O.
(2019). The effects of psychosocial stress on dopaminergic function and the acute
stress response. ELIFE, 8. https://doi.org/10.7554/eLife.46797
Bloomfield M., McCutcheon R., Dahoun T., Kempton M., Valmaggia L., Freeman T.,
Glover V., Pruesner J., Morgan C., Kapur S., & Howes O. (2017). The effects of
long-term psychosocial adversity on dopaminergic function and the acute stress
response. European Neuropsychopharmacology, 27(Supplement 4), S718.
Boyle, A. E., Gill, K., Smith, B. R., & Amit, Z. (n.d.). Differential Effects of an Early
Housing Manipulation on Cocaine-Induced Activity and Self-Administration in
Laboratory Rats. 6.
Butts, K. A., Weinberg, J., Young, A. H., & Phillips, A. G. (2011). Glucocorticoid
receptors in the prefrontal cortex regulate stress-evoked dopamine efflux and
aspects of executive function. Proceedings of the National Academy of Sciences,
108(45), 18459–18464. https://doi.org/10.1073/pnas.1111746108
167

Casey K.F., Cherkasova M.V., Benkelfat C., Dagher A., & Leyton M. (2010). Childhood
family function predicts dopamine response to d-amphetamine challenge: A PET
[11C] raclopride study. Biological Psychiatry, 67(9 SUPPL. 1), 169S-170S.
Clark, H. W., Masson, C. L., Delucchi, K. L., Hall, S. M., & Sees, K. L. (2001). Violent
traumatic events and drug abuse severity. Journal of Substance Abuse Treatment,
20(2), 121–127. https://doi.org/10.1016/S0740-5472(00)00156-2
Cisler, J. M., Amstadter, A. B., Begle, A. M., Resnick, H. S., Danielson, C. K., Saunders, B.
E., & Kilpatrick, D. G. (2011). PTSD symptoms, potentially traumatic event
exposure, and binge drinking: A prospective study with a national sample of
adolescents. Journal of anxiety disorders, 25(7), 978-987.
Cole, J., Sprang, G., & Silman, M. (2019). Interpersonal Trauma Exposure, Trauma
Symptoms, and Severity of Substance Use Disorder among Youth Entering
Outpatient Substance Abuse Treatment. Journal of Child & Adolescent Trauma,
12(3), 341–349. https://doi.org/10.1007/s40653-018-0239-3
Daftary, S. S., Panksepp, J., Dong, Y., & Saal, D. B. (2009). Stress-induced,
glucocorticoid-dependent strengthening of glutamatergic synaptic transmission in
midbrain dopamine neurons. Neuroscience Letters, 452(3), 273–276.
https://doi.org/10.1016/j.neulet.2009.01.070
Dahoun, T., Nour, M. M., McCutcheon, R. A., Adams, R. A., Bloomfield, M. A. P., &
Howes, O. D. (2019). The relationship between childhood trauma, dopamine
168
release and dexamphetamine-induced positive psychotic symptoms: A [C-11]-(+)-

PHNO PET study. TRANSLATIONAL PSYCHIATRY, 9.
https://doi.org/10.1038/s41398-019-0627-y

Duffy, K. A., McLaughlin, K. A., & Green, P. A. (2018). Early life adversity and health-
risk behaviors: Proposed psychological and neural mechanisms. Annals of the
New York Academy of Sciences, 1428(1), 151–169.
https://doi.org/10.1111/nyas.13928
Egerton A., Howes O.D., Houle S., McKenzie K., Valmaggia L.R., Bagby M.R., Tseng
H.-H., Bloomfield M.A., Kenk M., Bhattacharyya S., Suridjan I., Chaddock C.A.,
Winton-Brown T.T., Allen P., Rusjan P., Remington G., Meyer-Lindenberg A.,
McGuire P.K., & Mizrahi R. (2017). Elevated Striatal Dopamine Function in
Immigrants and Their Children: A Risk Mechanism for Psychosis. Schizophrenia
Bulletin, 43(2), 293–301. https://doi.org/10.1093/schbul/sbw181
Egerton A., Valmaggia L.R., Howes O.D., Day F., Chaddock C.A., Allen P., Winton-
Brown T.T., Bloomfield M.A.P., Bhattacharyya S., Lappin J., Murray R.M., &
McGuire P.K. (2014). Effect of childhood adversity on brain dopamine function
in adulthood. Early Intervention in Psychiatry, 8(SUPPL. 1), 37.
https://doi.org/10.1111/eip.12186
Enman, N. M., Arthur, K., Ward, S. J., Perrine, S. A., & Unterwald, E. M. (2015).
Anhedonia, reduced cocaine reward, and dopamine dysfunction in a rat model of
Posttraumatic Stress Disorder. Biological Psychiatry, 78(12), 871–879.
Enoch, M.-A. (2011). The role of early life stress as a predictor for alcohol and drug
dependence. Psychopharmacology, 214(1), 17–31.
https://doi.org/10.1007/s00213-010-1916-6
169
Erb, S., Shaham, Y., & Stewart, J. (1996). Stress reinstates cocaine-seeking behavior
after prolonged extinction and a drug-free period. Psychopharmacology, 128(4),
408. https://doi.org/10.1007/s002130050150
Fahlke, C., Engel, J. A., Peter Eriksson, C. J., Hȧrd, E., & Söderpalm, B. (1994).
Involvement of corticosterone in the modulation of ethanol consumption in the
rat. Alcohol, 11(3), 195–202. https://doi.org/10.1016/0741-8329(94)90031-0
https://doi.org/10.1016/S0749-3797(98)00017-8
Fiore, V. G., Mannella, F., Mirolli, M., Latagliata, E. C., Valzania, A., Cabib, S., Dolan,
R. J., Puglisi-Allegra, S., & Baldassarre, G. (2015). Corticolimbic catecholamines
in stress: A computational model of the appraisal of controllability. Brain
Structure and Function, 220(3), 1339–1353. https://doi.org/10.1007/s00429-014-
0727-7
Gambarana, C., Masi, F., Tagliamonte, A., Scheggi, S., Ghiglieri, O., & Montis, M. G. D.
(1999). A Chronic Stress that Impairs Reactivity in Rats Also Decreases
Dopaminergic Transmission in the Nucleus Accumbens: A Microdialysis Study.
Journal of Neurochemistry, 72(5), 2039–2046. https://doi.org/10.1046/j.1471-
4159.1999.0722039.x

170
Gerra, G., Leonardi, C., Cortese, E., Zaimovic, A., Dell’Agnello, G., Manfredini, M.,
Somaini, L., Petracca, F., Caretti, V., Raggi, M. A., & Donnini, C. (2009).
Childhood neglect and parental care perception in cocaine addicts: Relation with
psychiatric symptoms and biological correlates. Neuroscience & Biobehavioral
Reviews, 33(4), 601–610. https://doi.org/10.1016/j.neubiorev.2007.08.002
https://doi.org/10.1007/s00213-002-1007-4
Goeders, N. E., & Guerin, G. F. (1994). Non-contingent electric footshock facilitates the
acquisition of intravenous cocaine self-administration in rats.
Psychopharmacology, 114(1), 63–70. https://doi.org/10.1007/BF02245445
Goldstein, D. S., & Kopin, I. J. (2018). Linking Stress, Catecholamine Autotoxicity, and
Allostatic Load with Neurodegenerative Diseases: A Focused Review in Memory
of Richard Kvetnansky. Cellular and Molecular Neurobiology, 38(1), 13–24.
https://doi.org/10.1007/s10571-017-0497-x
Haber, S. N., & Knutson, B. (2010). The Reward Circuit: Linking Primate Anatomy and
Human Imaging. Neuropsychopharmacology, 35(1), 4–26.
https://doi.org/10.1038/npp.2009.129
Hall, F. S., Wilkinson, L. S., Humby, T., Inglis, W., Kendall, D. A., Marsden, C. A., &
Robbins, T. W. (1998). Isolation Rearing in Rats: Pre- and Postsynaptic Changes
in Striatal Dopaminergic Systems. Pharmacology Biochemistry and Behavior,
59(4), 859–872. https://doi.org/10.1016/S0091-3057(97)00510-8
Haney, M., Maccari, S., Le Moal, M., Simon, H., & Vincenzo Piazza, P. (1995). Social
stress increases the acquisition of cocaine self-administration in male and female
rats. Brain Research, 698(1–2), 46–52. https://doi.org/10.1016/0006-
8993(95)00788-R
171
Harfstrand, A., Fuxe, K., Cintra, A., Agnati, L. F., Zini, I., Wikstrom, A. C., Okret, S.,
Yu, Z. Y., Goldstein, M., & Steinbusch, H. (1986). Glucocorticoid receptor
immunoreactivity in monoaminergic neurons of rat brain. Proceedings of the
National Academy of Sciences, 83(24), 9779–9783.
Harris, D. S., Reus, V. I., Wolkowitz, O. M., Mendelson, J. E., & Jones, R. T. (2003).
Altering Cortisol Level does not Change the Pleasurable Effects of
Methamphetamine in Humans. Neuropsychopharmacology, 28(9), 1677–1684.
https://doi.org/10.1038/sj.npp.1300223
Hensleigh, E., & Pritchard, L. M. (2013). Glucocorticoid receptor expression and sub-
cellular localization in dopamine neurons of the rat midbrain. Neuroscience
Letters, 556, 191–195. https://doi.org/10.1016/j.neulet.2013.09.067
https://doi.org/10.1007/s00213-015-4151-3
Imperato, A., Angelucci, L., Casolini, P., Zocchi, A., & Puglisi-Allegra, S. (1992).
Repeated stressful experiences differently affect limbic dopamine release during
and following stress. Brain Research, 577(2), 194–199.
https://doi.org/10.1016/0006-8993(92)90274-D
172
Ito, R., Dalley, J. W., Robbins, T. W., & Everitt, B. J. (2002). Dopamine release in the
dorsal striatum during cocaine-seeking behavior under the control of a drug-
associated cue. Journal of Neuroscience, 22(14), 6247-6253.
Karlsgodt, K. H., Lukas, S. E., & Elman, I. (2003). Psychosocial Stress and the Duration
of Cocaine Use in Non‐treatment Seeking Individuals with Cocaine Dependence.
The American Journal of Drug and Alcohol Abuse, 29(3), 539–551.
https://doi.org/10.1081/ADA-120023457
Kasanova, Z., Hernaus, D., Vaessen, T., van Amelsvoort, T., Winz, O., Heinzel, A.,
Pruessner, J., Mottaghy, F. M., Collip, D., & Myin-Germeys, I. (2016). Early-Life
Stress Affects Stress-Related Prefrontal Dopamine Activity in Healthy Adults, but
Not in Individuals with Psychotic Disorder. PloS One, 11(3), e0150746.
https://doi.org/10.1371/journal.pone.0150746
Keyes, K. M., Hatzenbuehler, M. L., & Hasin, D. S. (2011). Stressful life experiences,
alcohol consumption, and alcohol use disorders: The epidemiologic evidence for
four main types of stressors. Psychopharmacology, 218(1), 1–17.
https://doi.org/10.1007/s00213-011-2236-1
Koenig, H. N., & Olive, M. F. (2004). The glucocorticoid receptor antagonist

mifepristone reduces ethanol intake in rats under limited access
conditions. Psychoneuroendocrinology, 29(8), 999-1003.
Kosten, T. R. (1986). A 2.5-Year Follow-up of Depression, Life Crises, and Treatment

Effects on Abstinence Among Opioid Addicts. Archives of General Psychiatry,
43(8), 733. https://doi.org/10.1001/archpsyc.1986.01800080019003
Krupnik, V. (2019). Trauma or adversity?. Traumatology, 25(4), 256.
Kuksis, M., Di Prospero, C., Hawken, E. R., & Finch, S. (2017). The correlation between
trauma, PTSD, and substance abuse in a community sample seeking outpatient
treatment for addiction. Canadian Journal of Addiction, 8(1), 18–24.
173
LeSage, M. G., Stafford, D., & Glowa, J. R. (1999). Preclinical research on cocaine self-
administration: Environmental determinants and their interaction with
pharmacological treatment. Neuroscience & Biobehavioral Reviews, 23(5), 717–
741. https://doi.org/10.1016/S0149-7634(99)00015-9
Levis, S. C., Baram, T. Z., & Mahler, S. V. (2021). Neurodevelopmental origins of

substance use disorders: Evidence from animal models of early‐life adversity and
addiction. European Journal of Neuroscience.
Leyton, M., & Vezina, P. (2014). Dopamine ups and downs in vulnerability to addictions:
a neurodevelopmental model. Trends in pharmacological sciences, 35(6), 268-
276.
Leyton, M. (2021). Ubiquitous Dopamine Deficit Hypotheses in Cocaine Use Disorder

Lack Support. American Journal of Psychiatry, 178(5), 469-469.
Liebschutz, J., Saitz, R., & Lloyd-Travaglini, C. (2002). The relationship between sexual
and physical abuse and substance abuse consequences. Journal of Substance
Abuse Treatment, 8.
Low, N. C., Dugas, E., O’Loughlin, E., Rodriguez, D., Contreras, G., Chaiton, M., &
O’Loughlin, J. (2012). Common stressful life events and difficulties are
associated with mental health symptoms and substance use in young adolescents.
BMC Psychiatry, 12(1), 116. https://doi.org/10.1186/1471-244X-12-116
Loxton, D., Forder, P. M., Cavenagh, D., Townsend, N., Holliday, E., Chojenta, C., &
Melka, A. S. (2021). The impact of adverse childhood experiences on the health
and health behaviors of young Australian women. Child Abuse & Neglect, 111,
104771. https://doi.org/10.1016/j.chiabu.2020.104771
Mandavia, A., Robinson, G. G. N., Bradley, B., Ressler, K. J., & Powers, A. (2016).
Exposure to Childhood Abuse and Later Substance Use: Indirect Effects of
Emotion Dysregulation and Exposure to Trauma. Journal of Traumatic Stress,
29(5), 422–429. https://doi.org/10.1002/jts.22131
174
Mangiavacchi, S., Masi, F., Scheggi, S., Leggio, B., Montis, M. G. D., & Gambarana, C.
(2001). Long-term behavioral and neurochemical effects of chronic stress
exposure in rats. Journal of Neurochemistry, 79(6), 1113–1121.
https://doi.org/10.1046/j.1471-4159.2001.00665.x
Marinelli, M., Aouizerate, B., Barrot, M., Le Moal, M., & Piazza, P. V. (1998).
Dopamine-dependent responses to morphine depend on glucocorticoid receptors.
Proceedings of the National Academy of Sciences, 95(13), 7742–7747.
Marinelli, Michela, & Piazza, P. V. (2002). Interaction between glucocorticoid hormones,

stress and psychostimulant drugs*. European Journal of Neuroscience, 16(3),
387–394. https://doi.org/10.1046/j.1460-9568.2002.02089.x
Meaney, M. J., Brake, W., & Gratton, A. (2002). Environmental regulation of the
development of mesolimbic dopamine systems: A neurobiological mechanism for
vulnerability to drug abuse? Psychoneuroendocrinology, 27(1–2), 127–138.
https://doi.org/10.1016/S0306-4530(01)00040-3
Miczek, K. A., Nikulina, E. M., Shimamoto, A., & Covington, H. E. (2011). Escalated or
Suppressed Cocaine Reward, Tegmental BDNF, and Accumbal Dopamine
Caused by Episodic versus Continuous Social Stress in Rats. Journal of
Neuroscience, 31(27), 9848–9857. https://doi.org/10.1523/JNEUROSCI.0637-
11.2011
Mizrahi, R., Kenk, M., Suridjan, I., Boileau, I., George, T. P., McKenzie, K., Wilson, A.
A., Houle, S., & Rusjan, P. (2014). Stress-induced dopamine response in subjects
at clinical high risk for schizophrenia with and without concurrent cannabis use.
Neuropsychopharmacology : Official Publication of the American College of
Neuropsychopharmacology, 39(6), 1479–1489.
https://doi.org/10.1038/npp.2013.347
Nader, M. A., Sinnott, R. S., Mach, R. H., & Morgan, D. (2002). Cocaine-and food-
maintained responding under a multiple schedule in rhesus monkeys:
175
environmental context and the effects of a dopamine

antagonist. Psychopharmacology, 163(3), 292-301.
Nader, M. A., Morgan, D., Gage, H. D., Nader, S. H., Calhoun, T. L., Buchheimer, N., ...
& Mach, R. H. (2006). PET imaging of dopamine D2 receptors during chronic
cocaine self-administration in monkeys. Nature neuroscience, 9(8), 1050-1056.
Nutt, D. J., Lingford-Hughes, A., Erritzoe, D., & Stokes, P. R. A. (2015). The dopamine
theory of addiction: 40 years of highs and lows. Nature Reviews Neuroscience,
16(5), 305–312. https://doi.org/10.1038/nrn3939
Oswald, L. M., Wand, G. S., Kuwabara, H., Wong, D. F., Zhu, S., & Brasic, J. R. (2014).
History of childhood adversity is positively associated with ventral striatal
dopamine responses to amphetamine. PSYCHOPHARMACOLOGY, 231(12),
2417–2433. https://doi.org/10.1007/s00213-013-3407-z
Oswald L.M., Wand G.S., Wong D.F., Kuwabara H., Brasic J., Chamroonrat W.,
Guevara M., & Zaidi E. (2011). History of early childhood trauma is positively
associated with ventral striatal dopamine responses to amphetamine.
Neuropsychopharmacology, 36(SUPPL. 1), S287.
https://doi.org/10.1038/npp.2011.292
Ouimette, P. C., Kimerling, R., Shaw, J., & Moos, R. H. (2000). Physical and Sexual
Abuse Among Women and Men with Substance Use Disorders. Alcoholism
Treatment Quarterly, 18(3), 7–17. https://doi.org/10.1300/J020v18n03_02
Pascucci, T., Ventura, R., Latagliata, E. C., Cabib, S., & Puglisi-Allegra, S. (2007). The
Medial Prefrontal Cortex Determines the Accumbens Dopamine Response to
Stress through the Opposing Influences of Norepinephrine and Dopamine.
Cerebral Cortex, 17(12), 2796–2804. https://doi.org/10.1093/cercor/bhm008
Phillips, A. G., & Fibiger, H. C. (1973). Dopaminergic and noradrenergic substrates of

positive reinforcement: differential effects of d-and l-
amphetamine. Science, 179(4073), 575-577.
176
Phillips, G. D., Howes, S. R., Whitelaw, R. B., Wilkinson, L. S., Robbins, T. W., &
Everitt, B. J. (1994). Isolation rearing enhances the locomotor response to cocaine
and a novel environment, but impairs the intravenous self-administration of
cocaine. Psychopharmacology, 115(3), 407-418.
Piazza, P. V., Barrot, M., Rouge-Pont, F., Marinelli, M., Maccari, S., Abrous, D. N.,
Simon, H., & Le Moal, M. (1996). Suppression of glucocorticoid secretion and
antipsychotic drugs have similar effects on the mesolimbic dopaminergic
transmission. Proceedings of the National Academy of Sciences, 93(26), 15445–
15450. https://doi.org/10.1073/pnas.93.26.15445
Piazza, P. V., Maccari, S., Deminiere, J. M., Le Moal, M., Mormede, P., & Simon, H.
(1991). Corticosterone levels determine individual vulnerability to amphetamine
self-administration. Proceedings of the National Academy of Sciences, 88(6),
2088–2092. https://doi.org/10.1073/pnas.88.6.2088
Piazza, P. V., & Le Moal, M. (1998). The role of stress in drug self-
administration. Trends in pharmacological sciences, 19(2), 67-74.
Piazza, P. V., Rouge-Pont, F., Deroche, V., Maccari, S., Simon, H., & Le Moal, M.
(1996a). Glucocorticoids have state-dependent stimulant effects on the
mesencephalic dopaminergic transmission. Proceedings of the National Academy
of Sciences, 93(16), 8716–8720. https://doi.org/10.1073/pnas.93.16.8716
Piazza, P. V., & Le Moal, M. (1996b). Pathophysiological basis of vulnerability to drug

abuse: role of an interaction between stress, glucocorticoids, and dopaminergic
neurons. Annual review of pharmacology and toxicology, 36(1), 359-378.
Pierce, R. C., & Kumaresan, V. (2006). The mesolimbic dopamine system: The final
common pathway for the reinforcing effect of drugs of abuse? Neuroscience &
177
Pirard, S., Sharon, E., Kang, S. K., Angarita, G. A., & Gastfriend, D. R. (2005).
Prevalence of physical and sexual abuse among substance abuse patients and
impact on treatment outcomes. Drug and Alcohol Dependence, 8.
Pruessner, J., Champagne, F., Meaney, M., & Dagher, A. (2004). Dopamine release in
response to a psychological stress in humans and its relationship to early life
maternal care: A positron emission tomography study using [C-11]raclopride.
Journal Of Neuroscience, 24(11), 2825–2831.
https://doi.org/10.1523/JNEUROSCI.3422-03.2004
Puglisi-Allegra, S., Imperato, A., Angelucci, L., & Cabib, S. (1991). Acute stress induces
time-dependent responses in dopamine mesolimbic system. Brain Research,
554(1), 217–222. https://doi.org/10.1016/0006-8993(91)90192-X
Ramsey, N. F., & Van Ree, J. M. (1993). Emotional but not physical stress enhances
intravenous cocaine self-administration in drug-naive rats. Brain Research,
608(2), 216–222. https://doi.org/10.1016/0006-8993(93)91461-Z
Rice, C., Mohr, C. D., Boca, F. K. D., Mattson, M. E., Young, L., Brady, K., & Nickless,
C. (2001). Self-Reports of Physical, Sexual and Emotional Abuse in an
Alcoholism Treatment Sample(*). Journal of Studies on Alcohol, 62(1), 114–114.
Saal, D., Dong, Y., Bonci, A., & Malenka, R. C. (2003). Drugs of Abuse and Stress
Trigger a Common Synaptic Adaptation in Dopamine Neurons. Neuron, 37(4),
577–582. https://doi.org/10.1016/S0896-6273(03)00021-7
Schenk, S., Lacelle, G., Gorman, K., & Amit, Z. (1987). Cocaine self-administration in
rats influenced by environmental conditions: Implications for the etiology of drug
abuse. Neuroscience Letters, 81(1–2), 227–231. https://doi.org/10.1016/0304-
3940(87)91003-2
Schifani C., Tseng H.-H., Kenk M., Tagore A., Kiang M., Wilson A.A., Houle S., Rusjan
P.M., & Mizrahi R. (2018). Cortical stress regulation is disrupted in schizophrenia
178
but not in clinical high risk for psychosis. Brain, 141(7), 2213–2224.
https://doi.org/10.1093/brain/awy133
Schwabe, L., Dickinson, A., & Wolf, O. T. (2011). Stress, habits, and drug addiction: A
psychoneuroendocrinological perspective. Experimental and Clinical
Psychopharmacology, 19(1), 53–63.
Sebold, M., Spitta, G., Gleich, T., Dembler-Stamm, T., Butler, O., Zacharias, K., Aydin,
S., Garbusow, M., Rapp, M., Schubert, F., Buchert, R., Gallinat, J., & Heinz, A.
(2019). Stressful life events are associated with striatal dopamine receptor
availability in alcohol dependence. JOURNAL OF NEURAL TRANSMISSION,
126(9, SI), 1127–1134. https://doi.org/10.1007/s00702-019-01985-2
Shaham, Y., Erb, S., & Stewart, J. (2000). Stress-induced relapse to heroin and cocaine
seeking in rats: A review. Brain Research Reviews, 33(1), 13–33.
https://doi.org/10.1016/S0165-0173(00)00024-2
Shimamoto, A., DeBold, J. F., Holly, E. N., & Miczek, K. A. (2011). Blunted accumbal
dopamine response to cocaine following chronic social stress in female rats:
Exploring a link between depression and drug abuse. Psychopharmacology,
218(1), 271–279. https://doi.org/10.1007/s00213-011-2364-7
Sinha, R., Catapano, D., & O’Malley, S. (1999). Stress-induced craving and stress
response in cocaine dependent individuals. Psychopharmacology, 142(4), 343–
351. https://doi.org/10.1007/s002130050898
Sinha, R., Fuse, T., Aubin, L.-R., & O’Malley, S. S. (2000). Psychological stress, drug-
related cues and cocaine craving. Psychopharmacology, 152(2), 140–148.
https://doi.org/10.1007/s002130000499
Sinha, Rajita. (2002). Sinha R. How does stress increase risk of drug abuse and relapse?
Psychopharmacology (Berl) 158: 343-359. Psychopharmacology, 158, 343–359.
https://doi.org/10.1007/s002130100917
179
Sinha, Rajita, Talih, M., Malison, R., Cooney, N., Anderson, G. M., & Kreek, M. J.
(2003). Hypothalamic-pituitary-adrenal axis and sympatho-adreno-medullary
responses during stress-induced and drug cue-induced cocaine craving states.
Psychopharmacology, 170(1), 62–72. https://doi.org/10.1007/s00213-003-1525-8
Smith, A. D., Castro, S. L., & Zigmond, M. J. (2002). Stress-induced Parkinson’s

disease: A working hypothesis. Physiology & Behavior, 77(4–5), 527–531.
Smith, S. M., & Vale, W. W. (2006). The role of the hypothalamic-pituitary-adrenal axis
in neuroendocrine responses to stress. Dialogues in Clinical Neuroscience, 8(4),
383. https://doi.org/10.31887/DCNS.2006.8.4/ssmith
Sugama, S., & Kakinuma, Y. (2016). Loss of dopaminergic neurons occurs in the ventral
tegmental area and hypothalamus of rats following chronic stress: Possible
pathogenetic loci for depression involved in Parkinson’s disease. Neuroscience
Research, 111, 48–55. https://doi.org/10.1016/j.neures.2016.04.008
Sugama, S., Sekiyama, K., Kodama, T., Takamatsu, Y., Takenouchi, T., Hashimoto, M.,
Bruno, C., & Kakinuma, Y. (2016). Chronic restraint stress triggers dopaminergic
and noradrenergic neurodegeneration: Possible role of chronic stress in the onset
of Parkinson’s disease. Brain, Behavior, and Immunity, 51, 39–46.
https://doi.org/10.1016/j.bbi.2015.08.015
Sussman, S., & Dent, C. W. (2000). One-Year Prospective Prediction of Drug Use from
Stress-Related Variables. Substance Use & Misuse, 35(5), 717–735.
https://doi.org/10.3109/10826080009148418
Tidey, J. W., & Miczek, K. A. (1996). Social defeat stress selectively alters
mesocorticolimbic dopamine release: An in vivo microdialysis study. Brain
Research, 721(1–2), 140–149. https://doi.org/10.1016/0006-8993(96)00159-X
180
Valenti, O., Lodge, D. J., & Grace, A. A. (2011). Aversive Stimuli Alter Ventral
Tegmental Area Dopamine Neuron Activity via a Common Action in the Ventral
Hippocampus. Journal of Neuroscience, 31(11), 4280–4289.
https://doi.org/10.1523/JNEUROSCI.5310-10.2011
Vanderschuren, L. J., Di Ciano, P., & Everitt, B. J. (2005). Involvement of the dorsal
striatum in cue-controlled cocaine seeking. Journal of Neuroscience, 25(38),
8665-8670.
Vazquez, V., Penit-Soria, J., Durand, C., Besson, M. J., Giros, B., & Dauge, V. (n.d.).
Maternal Deprivation Increases Vulnerability to Morphine Dependence and
Disturbs the Enkephalinergic System in Adulthood. 10.
Volkow, N., Fowler, J., & Wang, G. (2004). The addicted human brain viewed in the
light of imaging studies: Brain circuits and treatment strategies.
NEUROPHARMACOLOGY, 47(S), 3–13.
https://doi.org/10.1016/j.neuropharm.2004.07.019
Wachtel, S. R., Charnot, A., & de Wit, H. (2001). Acute hydrocortisone administration
does not affect subjective responses to d-amphetamine in humans.
Wise, R. A., & Rompre, P.-P. (1989). Brain Dopamine and Reward. Annual Review of
Psychology, 40(1), 191–225.
https://doi.org/10.1146/annurev.ps.40.020189.001203
Wise, Roy A. (2004). Dopamine, learning and motivation. Nature Reviews Neuroscience,
5(6), 483–494. https://doi.org/10.1038/nrn1406
181
182
Chapter 4
4 Ventral striatal activity links adversity and reward

processing in children
4.1 Introduction
Individual differences in reward-based learning and decision-making take root
early in life and predict a host of later outcomes including physical and psychological
health, financial well-being, academic achievement, and social adjustment (Mischel,
Shoda, & Rodriguez, 1989; Moffitt et al., 2011; Schlam, Wilson, Shoda, Mischel, &
Ayduk, 2013; Shoda, Mischel, & Peake, 1990). Understanding the origins of early
differences in reward processing is therefore an important research goal.
The current study focused on potential links between reward-based learning and
decision-making early in development and exposure to adverse life events. In humans,
adversity has been linked to a variety of alterations in reward processing, including both
potentiated and attenuated motivation to approach prospective rewards. For example,
adversity experienced by traumatized children is associated with heightened impulsivity
and hyperactivity (Laucht et al., 2007), increased risk of substance use and addiction
(Sinha, 2008), riskier sexual behavior (Cinq-Mars, Write, Cyr, & McDuff, 2004; Noll,
Shenk, & Putnam, 2009; Senn, Carey, Vanable, Coury-Doniger, & Urban, 2007), and
heightened incidence of obesity (Davis et al., 2014; Farr et al., 2015; Non et al., 2016),
but also anhedonia and depression (Bogdan & Pizzagalli, 2006; Pizzagalli & Bogdan,
2007). From the use of animal models, it is known that adult rodents who, as pups, were
separated from their mother show greater impulsivity, sensitivity to rewards, and
behavioral inflexibility compared to adults who were reared naturally (Hall et al., 1998;
Lovic, Keen, & Fletcher, 2011). And in rhesus monkeys, individuals at the bottom of a
social hierarchy show greater self-administration of cocaine compared to individuals at
the apex of the hierarchy (Morgan et al., 2002). Why – on a very basic level – adversity is
associated with alterations in reward processing, and whether this association extends to
non-traumatized children remains unclear.
183
Our hypothesis is that variation in the quality of early experience will have
implications for the functional calibration of the brain’s reward system later in childhood.
The reward system is an evolutionarily well-conserved network of subcortical and
cortical brain regions that includes the ventral tegmental area (VTA), the ventral striatum
(VS), ventromedial and orbital prefrontal cortex (vmPFC, OFC), and dorsal anterior
cingulate cortex (dACC). Although functionally multifaceted, these regions all utilize
dopaminergic signals to support different aspects of reward-based learning and decision-
making. According to developmental and evolutionary theory (Gatzke-Kopp; Gluckman
& Hanson, 2004; Meaney, 2001; Meaney, 2010), biological systems, such as the reward
system, enjoy considerable functional plasticity early in development, making them
highly attuned to indicators of environmental quality. Thus, adjustments in dopamine-
related phenotypic traits – including dopamine availability, dopamine receptor density,
and dopamine-mediated behavioral traits such as learning rate to rewards, activity level,
and novelty seeking – can occur in response to variation in environmental quality,
including variation that falls within a normative range (Gatzke-Kopp, 2011; Gluckman &
Hanson, 2004). These adjustments may be adaptive in the short-term, but carry the
burden of increased risk of adverse outcomes such as addiction and psychopathology. As
such, adverse early-life experiences can become “biologically embedded” in the
developing brain (Hertzman, 1999; Hertzman, 2012) and exert a lasting influence on the
physical and psychological health of the affected individual (Brake, Zhang, Diorio,
Meaney, & Gratton, 2004; Hall, Wilkinson, Humby, & Robbins, 1999; Hall et al., 1998;
Léonhardt, Matthews, Meaney, & Walker, 2007; Nelson, 2013).
Accumulating evidence, particularly from the use of animal models, is consistent

with the idea that adversity impacts reward-based learning and decision-making via its
influence on dopamine signaling in the reward system (Abercrombie, Keefe, DiFrischia,
& Zigmond, 1989; Hall et al., 1998; Hosking & Winstanley, 2011; Piazza & Le Moal,
1996; Lovic et al., 2011). In rodent and primate models, for example, adversity has been
associated with elevated levels of basal dopamine (DA) in the ventral striatum (VS),
potentiated dopamine response to amphetamine administration in the VS (Piazza & Le
Moal, 1996; Hall et al., 1996), lower D2 receptor density in the VS (Hall et al. 1998;
Morgan et al., 2002), and potentiated responses in dopamine-mediated traits such as
184
novelty-seeking and activity level (Lovic, et al., 2011; Piazza & Le Moal, 1996). These
effects appear to extend to humans, as early-life adversity has been associated with
elevated levels of striatal dopamine (Egerton et al., 2016; Preussner et al., 2004), and
increased striatal dopamine release to rewarding stimuli such as amphetamine (Oswald et
al., 2014). Whether these associations hold for normative levels of adversity – as might
be predicted (Gatzke-Kopp, 2011; Gluckman & Hanson, 2004) – is unknown.
The goals of the present study were therefore twofold. The first goal was to
examine whether normative variation in exposure to adverse early-life events is
associated with variation in children’s reward-based learning and decision-making. To
test this question, we recruited a sample of typically-developing children and assessed
exposure to a variety of events that while adverse, would be considered within the range
of normative experience. We then assessed reward-based learning and decision-making
using tasks that are dependent, at least in part, on dopamine signaling in the reward
system. The first was the Probabilistic Selection Task (PST), a measure of reward-based
approach-avoidance learning that has been shown, via genetics methods and the study of
Parkinson’s patients, to be sensitive to variation in levels of striatal dopamine (Frank,
Seeberger, & O’Reilly, 2004). The second was a Delay Discounting (DD) task, a measure
of inter-temporal (or impulsive) choice which has been shown to be associated with
striatal reactivity to the provision of rewards (Hariri et al., 2006). Having tested for an
association between adversity and reward-based learning and decision-making, the
second goal of the study was to examine whether possible associations between adversity
and reward processing were explainable at least in part, by the functional response of the
reward system to small gains and losses. To test for this possibility, children were
administered a reinforcement learning task as brain activity was measured via functional
magnetic resonance imaging (fMRI). The task provides a means of measuring behavioral
and brain response to both gains and losses separately (Pessiglione et al., 2006).
Importantly for the goals of the current study, learning rates and ventral striatal response
– specifically to gains – covary with pharmacologically-induced changes in striatal
dopamine, with these effects measureable using fMRI (Pessiglione et al., 2006). Use of
this task in combination with fMRI, then, provided a validated, albeit indirect measure of
reward-related dopaminergic transmission.
185
Our predictions were as follows:
(a) Greater experience of early-life adversity within the normative range would be
associated with potentiated reward-learning and more impulsive decision-making.
(b) The association between adversity and reward-based learning and decision-
making would be explainable, at least in part, by differences in the functional
response of regions within the reward system to the receipt of small gains.
4.2 Methods
Two studies examined the association between normative variation in early-life
adversity and children’s reward-based learning and decision-making. The first focused on
associations between adversity and reward-related behaviors; the second focused on
neurophysiological mediators of adversity and reward-behavior associations.
4.2.1 Method Study 1
4.2.1.1 Participants
Trained research assistants recruited 40 (24 females) children between the ages of
9-12 years (M = 10.75, SD = 0.95) from a database of London, Ontario families who
voluntarily participate in psychological research. Children who had experienced “trauma”
(e.g., physical/sexual abuse, witnessing death or severe injury of a family member/close
friend) or had any developmental, neurological, or psychiatric disorders were excluded
from the study. Parents provided written consent to their children’s participation; children
provided verbal assent. Child participants received a $25 gift card for participating.
Parents were compensated for travel and parking expenses. All aspects of the study were
conducted in accordance with the Declaration of Helsinki.
4.2.1.2 Measure of early-life adversity

To measure normative variation in exposure to adverse early-life events, parents
of child participants were administered the Early Life Experiences Questionnaire
(ELEQ). The measure consists of a list of 22 different adverse events (e.g., changing
schools, moving to a new neighborhood, loss of a pet, loss of a grandparent) that would
186
all be emotionally challenging for a child, but within the range of normative experience.
For each event, parents indicated how frequently and how intensely their child had
experienced the event. Parents could also add frequency and intensity estimates for
additional events that were not included in list of 22-events. A score for individual events
was computed as the product of frequency and intensity, and a Total Adversity score was
calculated for each participant as the sum of all individual event scores.
Use of the ELEQ was motivated by our interest in normative variation in

adversity exposure. Other instruments, such as The Juvenile Victimization Questionnaire,
or The Adverse Childhood Experiences Questionnaire, while valuable measures, profile
exposure to more severe and non-normative forms of adversity including maltreatment,
physical abuse, sexual abuse, warfare, violence, and victimization (Felitti et al., 1998;
Finkelhor, Hamby, Ormrod, & Turner, 2005).
4.2.1.3 Measure of reward-based learning and decision-making

Reward-based learning was assessed by means of the Probabilistic Selection Task
(PST), a measure of reinforcement learning which has previously been shown to be
sensitive to variation in levels of striatal dopamine (Frank et al., 2004). On each trial, one
of two pairs (AB and CD) of stimuli was presented and participants selected one member
of the pair for possible reward. For AB pairs, choice of A(B) returned reward on
80%(20%) of trials; for CD pairs, choice of C(D) returned reward on 70%(30%) of trials
(Frank et al., 2004). Based on participants’ choices while learning these pairings, we
estimated separate learning rates for gains and losses (i.e., αwin and αloss) using a
reinforcement learning (RL) model (Sutton & Barto, 1998). Learning rates, especially for
gains, correlate positively with tonic levels of DA in the VS (Pessiglione et al., 2006),
and were therefore considered a suitable means of assessing the impact of adversity on
DA-mediated behavior.
Reward-based decision-making was assessed by means of the Delay Discounting

task (DD) (Fishburn & Rubinstein, 1982; Hariri et al., 2006; van den Bos & McClure,
2013). On each of 88 separate trials, participants chose between a small immediate
reward and a larger delayed reward. Across trials, the value of the immediate reward
187
varied between $0 and $20 in increments of $1, and the delay was set to 7, 30, 90, or 180
days. The value of the delayed reward remained fixed at $20. Each unique combination
of immediate reward value and delay duration was presented in random order. To model
each participant’s choices, we first estimated indifference points for each delay using a
logistic regression method, where an indifference point is defined as the point along the
range of immediate reward values at a particular delay where the participant transitions
from choosing the immediate reward to choosing the delayed reward. We then estimated
how steeply participants discount the value of future rewards by fitting indifference
points with a hyperbolic discounting function. Larger values of the discounting parameter
reflect a preference for immediate over delayed rewards and have been associated with
higher VS reactivity (Hariri et al., 2006).
4.3 Results – Study 1

To test for whether variation in normative exposure to adverse early-life events
was associated with children’s reward-based learning, we correlated Total Adversity
scores computed from the ELEQ with reward and loss-related learning from PST testing
phase data. Adversity was positively correlated with reward-related learning, r(38) =
0.547, p = .002 (See Figure 4.1A), but not loss-related learning, r(38) = 0.189, p = .242
(See Figure 4.1B). To test whether variation in normative exposure to adverse early-life
events was associated with children’s reward-based decision-making, we correlated Total
Adversity scores computed from the ELEQ with delay discounting parameter estimates.
Adversity was positively correlated with discounting parameter magnitude, r(38) = .35, p
= .027, indicating that higher adversity scores were associated with more impulsive
choice (See Figure 4.1C). Age was not related to any learning (reward-related r(38) =
0.097, p = .551; loss-related r(38) = -0.054, p = .741) or decision-making measures
(delay-discounting r(38) = 0.143, p = .377) .
Given evidence that variation in normative exposure to adverse early-life events

was associated with variation in reward-based learning and decision-making, we were
then interested in whether this association might be explainable, at least in part, in terms
of variation in the brain’s response to the provision of small gains and losses. This was
the goal of Study 2.
188
Figure 4.1: Associations Between Adversity and Reward-Based Learning and

Decision-Making. (A) Adversity is positively correlated with reward-related learning
r(38) = 0.547, p < .05(B) Adversity is not significantly correlated with loss-related
learning r(38) = 0.189, p = .242 (C) Adversity is positively correlated with impulsive
decision-making r(38) = 0.35, p < .05.
4.4 Methods Study 2

4.4.1 Participants
Participants included 26 (12 females) children ranging between 9- and 12-years of
age (M = 10.69, SD = 1.01). Twelve of the 26 children in the imaging protocol also
189
participated in Study 1. Inclusion and exclusion criteria, consent and assent procedures,
and compensation procedures were all identical to Study 1. All aspects of the study were
conducted in accordance with the Declaration of Helsinki.
4.4.2 Measure of reward-based learning

To measure reward-based learning, participants were administered a simple
reinforcement learning task presented as a game in which they could win points
(Pessiglione et al., 2006). To increase children’s motivation to play, children were told
that they would be given a monetary reward that scaled with the number of points earned.
In fact, all children received the same monetary reward regardless of the number of
points earned.
On each trial, participants were presented a pair of stimuli and chose one stimulus
for possible gain (+10 points) or loss (-10 points). There were three types of stimulus
pairs: gain pairs (AB), loss pairs (CD), and null (EF) pairs. For AB (or gain) pairs, choice
of A returned 10 (0) points on 80% (20%) of trials, and choice of B returned 10 (0) points
on 20% (80%) of trials. For CD (or loss) pairs, choice of C returned -10 (0) points on
80% (20%) of trials and choice of D returned -10 (0) points on 20% (80%) of trials. And
for EF (or null) pairs, choice of either E or F returned 0 points.
Individual trials were 4000ms in duration, and consisted of a 3000ms stimulus
presentation/response period and a 1000ms feedback period. Individual trials were
jittered in their presentation through use of an inter-trial interval that randomly varied in
duration from 1 to 5s in 1s increments. During the stimulus/response period, one
stimulus-pair was presented, and participants selected one member of the pair by means
of a button-press to an MRI-compatible button-box. Left button presses with the 2D
finger were computer-coded as a choice of the stimulus presented on the left side of the
screen; right button presses with the 3D finger were computer-coded as a choice of the
stimulus presented on the right side of the screen. Responses did not terminate the
stimulus-pair. Following the stimulus/response period, feedback appeared in the center of
the screen for 1000 ms with either “+10”, “-10”, “0” or “Too Slow” if no response was
registered in the preceding stimulus/response period. Position of the stimuli relative to
190
fixation (i.e., left versus right) varied randomly from trial to trial. The order of gain-pair,
loss-pair, and null trials was randomized for each participant.
Individual trials were administered in three 7-minute runs. Each run consisted of
44 trials and included 20 AB trials, 20 CD trials, and 4 EF trials. All stimuli were
motivationally neutral Snodgrass figures (Snodgrass & Vanderwart, 1980), with a unique
set of 6 stimuli used for each of the three runs. The task was presented using a Lenovo
laptop computer running E-Prime 2 software (Schneider, Eschman, & Zuccolotto, 2002).
4.4.3 MRI data acquisition

To mitigate fear and discomfort associated with the functional neuroimaging
procedure, children were first exposed to MRI-like environment in the form of a mock
scanner facility. MRI images were then acquired with a 3-Tesla Siemens Magnetom
Prisma scanner and a Siemens Prisma 32-channel head coil. Functional T2*-weighted
images were acquired with an ascending, interleaved slice order using a multiband echo-
planar imaging (EPI) pulse sequence (TR = 686 ms; TE = 30 ms; FOV=192 x 192 mm;
flip angle = 54°; voxel size = 3 mm3, 64 x 64 matrix). We selected this sequence to
maximize the sampling rate and permit better modeling of motion-related noise. A total
of 3 runs of functional data were collected from each participant. Each functional run
consisted of 650 volumes and lasted approximately 7 minutes.
After the completion of all 3 functional runs, we collected a high-resolution T1-

weighted anatomical image using a 3D MPRAGE pulse sequence (192 slices; voxel size
= 1 mm3, 256 x 256 matrix). The entire MRI procedure took approximately 1 hour to
complete and participants were compensated with $10.00 cash and a $25.00 gift card.
4.4.4 fMRI data pre-processing

fMRI data were preprocessed using Statistical Parametric Mapping 12 (SPM12).
For each run, volumes were spatially aligned to the first volume of the run that was
acquired. Realignment parameters were archived and subsequently used to estimate
participant motion during data acquisition. Participants with motion in excess of 3 mm of
translation or 1.5 degrees of rotation were dropped (18 participants were removed due to
motion, and the imaging sample consisted of the remaining 26 participants). Motion
191
parameters for each run were used as covariates of no interest in subsequent linear
modeling. After coregistering functional and anatomical images, data were spatially
normalized to Montreal Neurological Institute (MNI) space, and spatially smoothed via
convolution with an 8mm full-width at half-maximum Gaussian kernel.
4.4.5 Event-related modeling

Functional volumes were modeled by means of a general linear model (GLM)
with separate predictors for six nuisance predictors (i.e., subject motion) and the
following four events of interest:
1.) Wins: on gain-pair trials, instances when the participant gained 10 points;
2.) Misses: on gain-pair trials, instances when the participant gained 0 points;
3.) Losses: on loss-pair trials, instances when the participant lost 10 points;
4.) Avoids: on loss-pair trials, instances when participants lost 0 points.
Predictors for events of interest were created by convolving a vector of event

onsets with a canonical hemodynamic response function (HRF), where, for all events,
onsets were defined as the point in time that feedback was presented.
4.4.6 Reward-related Region of Interest (ROI) analysis

Reward-related Regions of Interest (ROIs) were defined through the use of the
Neurosynth platform. A reverse-inference map of 671 studies that included the term
“reward” in the abstract or introduction (http://www.neurosynth.org/; Yarkoni, Poldrack,
Nichols, Van Essen, & Wager, 2011) was generated and the results of the meta-analysis
were thresholded using a false discovery rate (FDR) correction for multiple comparisons
of 0.01. We subsequently visualized the results in Mango. From this visualization, we
identified and specified the coordinates of 4 ROIs (See Figure 4.4 A/B), including the left
and right ventral striatum (VS), and the left and right ventromedial prefrontal cortex
(vmPFC). We then used the MarsBar Region of Interest toolbox for SPM to extract and
average beta coefficients for all voxels within each of the four ROI’s, separately for each
of the four predictors (i.e., Wins, Misses, Losses, and Avoids – See Section 4.4.5). These
coefficients were then used to compute selected contrasts and brain-behavior correlations.
Following Pessiglione et al. (2007), gain-related activity was computed as the contrast of
192
Wins – Misses corrected for multiple comparisons at p < .05 via Bonferroni’s procedure,
and loss-related activity as the contrast of Losses – Avoids corrected for multiple
comparisons at p < .05 via Bonferroni’s procedure. To examine associations of gain- and
loss-related activity to behavior and early adversity, statistically significant contrasts of
Wins – Misses or Losses – Avoids were correlated with learning rates to wins and losses
and Total Adversity scores respectively.
4.5 Results – Study 2

4.5.1 Behavioral results
Over time, children learned to select A on AB trials (See Figure 4.2, pink) and
avoid C on CD trials (See Figure 4.2, blue), indicating that they learned to select stimuli
that maximized gains and minimized losses. Gain-pair accuracy, defined as the
proportion of AB trials of which participants chose A, was uncorrelated with loss-pair
accuracy, defined as the proportion of CD trials of which participants chose D (r(24) =
0.334, p = .095). Neither gain-pair accuracy (r(24) = 0.075, p = .716) nor loss-pair
accuracy (r(24) = 0.040, p = .846) were associated with age.
193
Figure 4.2: Observed choice to gain-pair and loss-pair stimuli. Over the course of
the trials, participants learned to select the stimulus that more frequently results in a
reward (pink) and avoid the stimulus that more frequently results in a loss (blue)
To test for possible associations between learning and adversity, we first

correlated gain- and loss-pair accuracy with Total Adversity scores. Total Adversity was
positively associated with gain-pair, r(24) = .39, p = .048, but not loss-pair accuracy,
r(24) = .25, p = .218 (See Figure 4.3A). Then, to specifically examine whether
dopamine-mediated aspects of reward-based learning explained the association between
adversity and choice behavior, we modeled choice behavior using an RL model, with
separate learning rates for gains (αwin) and losses (αloss). Consistent with findings from the
PST observed in Study 1, adversity was associated with learning rates for gains, r(24) =
0.40, p = .042, but was not associated with learning rates for losses r(24) = 0.1, p = .627
(See Figure 4.3B). Importantly, the association between adversity and choice behavior
was fully mediated by learning rate to gains (See Figure 4.3C and Fig. 4.3D).
194
Figure 4.3: Associations Between Adversity And Reward-Related and Loss-Related

Learning. (A) Adversity was positively correlated with gain-pair accuracy (pink) r(24) =
0.39, p < 0.05, but not loss-pair accuracy, r(24) = 0.25, p =.218 (blue). (B) Adversity was
positively correlated with learning rate to gains (pink) r(24) = 0.40, p < 0.05, but not
learning rate to losses r(24) = 0.1, p = .627 (blue). (C and D) The association between
adversity and gain-pair accuracy was fully mediated by learning rate to gains. ACME =
Average causal mediation effect; ADE = Average direct effect.
4.5.2 ROI activation, relation to behavior

Both left VS and left vmPFC showed significant gain- but not loss-related
activation. Gain-related activity predicted learning rates to gains (See Figure 4.4C/D)
both for the left VS, r(24) = 0.54, p < .05, and the left vmPFC, r(24) = 0.47, p < .05.
195
Figure 4.4: Neuroimaging Results. (A and B) Neurosynth meta-analysis of 671 studies

that included the word “reward”. Both the VS and vmPFC are preferentially related to the
term “reward”. Based on the meta-analysis ROIs were created for both the VS [-12,10,-
9; 12, 10, -9] as well as the vmPFC [2,62,-10; -2,62,-10], visible in blue. (C and D)
Reward-relate activity correlated positively with learning rate to gains in the VS r(24) =
0.54, p < 0.05, and vmPFC r(24) = 0.47, p < 0.05.
4.5.3 Adversity-behavior association partially explained by VS

activation
To examine whether the association between adversity and reward-related
learning could be explained, at least in part, by reward-related activity in our ROIs, we
tested separate mediation models for the VS and vmPFC. Reward-related activity in the
left VS partially mediated the association between adversity and learning rate for gains
(See Figure 4.5A and 4.5B). No other mediation models were significant.
196
Figure 4.5: Mediation Analysis. (A and B) Reward-related activity in the VS partially

mediated the relationship between adversity and learning rate to gains. ACME = Average
causal mediation effect; ADE = Average direct effect.
4.6 Discussion
Individual differences in children’s reward-based learning and decision-making
predict many important physical and psychological outcomes. The results of the present
study are important as they contribute to an understanding of the origin of these
individual differences.
First, we found that early life adversity predicts differences in typically

developing children’s reward-based learning and decision-making. Adversity experienced
early in development has been linked to increased impulsivity and reward incentive
salience, both in animals and in humans (Chugani et al., 2001; Hall, 1998; Pollak et al.,
2010). In rats, maternal-separation and isolate-rearing increase impulsivity and
hyperactivity, with effects more pronounced in measures of impulsive action than
impulsive choice (Lovic et al., 2011). In humans, exposure to adversity early in life is
associated with heightened ADHD symptomology including greater impulsivity and
hyperactivity (Laucht et al., 2007). In the current study, we found that children who had
experienced more frequent and intense adverse events early in life discounted temporally
displaced rewards more steeply and showed potentiated reward-based learning as
compared to children who had experienced less frequent and intense adverse early life
197
events. Interestingly, the association between adversity and learning was specific to
aspects of approach learning, including the likelihood of selecting previously rewarded
stimuli and RL-model estimates of learning rate to gains. There were no corresponding
associations between adversity and avoidance learning.
Second, we found that the link between adversity and reward-processing could be
explained, at least in part, by differences in ventral striatal response to rewards.
Consistent with earlier studies (Pessiglione et al., 2006), rewards were associated with
activity in the VS and the vmPFC, and reward-related activity in these regions predicted
reward-related learning, including RL-model estimates of learning rate to gains.
Interestingly, of these two regions, it was the VS that partially mediated the association
between adversity and learning rate to gains. Taken together then, our findings point to a
link between adversity, VS physiology, and reward-related behavior.
One plausible explanation for our findings is that early adversity contributes to
hyper-dopaminergic functioning in the VS. Indeed, pharmacologically induced changes
in striatal DA impact VS activity and reward-based learning in ways that are similar to
variations in adversity; an increase in striatal DA induced by the anti-Parkinsonian
medication L-DOPA for example, increases learning rates and VS response to gains but
has no effect on learning and striatal response to losses (Pessiglione et al., 2006). The
idea that early adversity contributes to hyper-dopaminergic functioning in the VS is
certainly consistent with evidence from animal studies and recent human imaging studies
(Egerton et al., 2016; Oswald et al., 2014). In rodents, for example, early adversity has
been linked to increases in tonic and phasic DA, as reflected in measures of basal DA
(Abercrombie et al., 1989; Hall et al., 1998) and evoked response to amphetamine
administration (Piazza & Le Moal 1996) respectively. Similarly in humans, adversity
experienced in childhood has been associated with elevated levels of dopamine in
adulthood (Egerton et al., 2016), and increased ventral striatal dopamine response to
amphetamine. In other cases however, there appears to be a blunted sensitivity to rewards
as measured by fMRI (Boecker et al., 2014; Dillon et al., 2009; Hanson et al., 2016;
Mehta, Gore-Langton, Golembo, & Colvert, 2010; Weller & Fisher, 2012), which may be
198
indicative of hypo-dopaminergic striatal functioning; these contradictory findings may be

a result of differences in the timing, type, and severity of adversity experienced.
Such discrepancies highlight several critical limitations of the present study. One
limitation concerns the fact that we used the ELEQ to measure children’s exposure to
adversity. The use of the ELEQ reflects our interest in normative variation in adversity;
however, this measure has not been widely used and awaits proper validation against
other more standardized measures such as ACES. A second limitation is that we did not
collect information about the pubertal status of our participants. Adolescence is a
developmental period during which individuals exhibit a hypersensitivity to rewards (Van
Leijenhorst et al., 2010), and therefore a completely unblemished picture of reward
processing in late childhood would remove any possible influence of pubertal onset. That
said, none of our measures of reward-based learning and decision-making were correlated
with age, suggesting that the current findings do not simply document variation related to
maturation.
The findings have important implications for our understanding of human

development by providing evidence that the functional calibration of the reward system is
sensitive to variation in adversity that falls within the normative range. Early childhood is
considered a sensitive period as variations in the quality of experience during this time
have lasting implications for behavioral, neurophysiological, and neuroendocrine
organization. This may be especially true of the reward system, given its role in learning
and motivated behavior specifically, and its importance for adaptation to different
environments more generally (for discussion see Gatzke-Kopp 2011; Frankenhuis &
Weerth, 2013). For example, using tasks that engaged the reward system, our results
show that children who experienced greater adversity exhibited potentiated reward-
related learning and impulsive decision-making. Importantly, the adverse life events do
not fall outside a range that could be considered normative; thus, these results highlight
the sensitivity of the reward system to indicators of environmental quality.
Our findings have potentially important health and policy implications as well.
Early variation in dopamine-mediated traits, such as reward incentive salience,
199
impulsivity, and striatal reactivity to rewards, have been linked to increased lifetime risk
of substance use/addiction, relationship problems, poor financial decision-making, and
low educational achievement (Moffitt et al., 2011). A more detailed understanding of the
impact of adversity on the developing brain may ultimately help to alleviate some of the
financial burden and human suffering related to these psychosocial issues.
Of course, much remains to be learned about the impact of early adversity on

neurocognitive development and reward processing. It is unclear whether mechanisms
relating adversity and reward processing are molecular (e.g., D2-receptor down-
regulation; Hall et al., 1998), anatomical (e.g., volumetric changes within the reward
system or in the integrity of fiber tracts connecting the reward system to other brain
regions, etc.), or involve interactions with other systems such as the stress system (see
Gatzke-Kopp, 2011; Piazza, Maccari, Deminiére, Le Moal, Morméde, & Simon 1991;
Piazza & Le Moal, 1996; Pruessner et al., 2004). These questions represent critical
avenues for future research.
200
4.7 References
Abercrombie, E. D., Keefe, K. a, DiFrischia, D. S., & Zigmond, M. J. (1989). Differential
effect of stress on in vivo dopamine release in striatum, nucleus accumbens, and
medial frontal cortex. Journal of Neurochemistry, 52(5), 1655–1658.
http://doi.org/10.1111/j.1471-4159.1989.tb09224.x
Boecker, R., Holz, N. E., Buchmann, A. F., Blomeyer, D., Plichta, M. M., Wolf, I., …
Laucht, M. (2014). Impact of early life adversity on reward processing in young
adults: EEG-fMRI results from a prospective study over 25 years. PloS One, 9(8),
e104185. http://doi.org/10.1371/journal.pone.0104185
Bogdan, R., & Pizzagalli, D. A. (2006). Acute stress reduces reward responsiveness:
implications for depression. Biological psychiatry, 60(10), 1147-1154.
http://dx.doi.org/10.1016/j.biopsych.2006.03.037
Brake, W. G., Zhang, T. Y., Diorio, J., Meaney, M. J., & Gratton, A. (2004). Influence of
early postnatal rearing conditions on mesocorticolimbic dopamine and behavioural
responses to psychostimulants and stressors in adult rats. European Journal of
Neuroscience, 19(7), 1863–1874. http://doi.org/10.1111/j.1460-9568.2004.03286.x
Casement, M. D., Shaw, D. S., Sitnick, S. L., Musselman, S. C., & Forbes, E. E. (2015).
Life stress in adolescence predicts early adult reward-related brain function and
alcohol dependence. Social Cognitive and Affective Neuroscience, 10(3), 416–23.
http://doi.org/10.1093/scan/nsu061
Chugani, H. T., Behen, M. E., Muzik, O., Juhász, C., Nagy, F., & Chugani, D. C. (2001).
Local brain functional activity following early deprivation: a study of
postinstitutionalized Romanian orphans. NeuroImage, 14(6), 1290–301.
http://doi.org/10.1006/nimg.2001.0917
Cinq-Mars, C., Wright, J., Cyr, M., & McDuff, P. (2004). Sexual at-risk behaviors of
sexually abused adolescent girls. Journal of Child Sexual Abuse, 12(2), 1-18.
http://dx.doi.org/10.1300/J070v12n02_01
201
Davis, C. R., Dearing, E., Usher, N., Trifiletti, S., Zaichenko, L., Ollen, E., ... &
Mantzoros, C. S. (2014). Detailed assessments of childhood adversity enhance
prediction of central obesity independent of gender, race, adult psychosocial risk
and health behaviors. Metabolism, 63(2), 199-206.
http://dx.doi.org/10.1016/j.metabol.2013.08.013
Dillon, D. G., Holmes, A.J., Birk, J.L., Brooks, N., Lyons-Ruth, K., & Pizzagalli, D.A.
(2009). Childhood Adversity Is Associated with Left Basal Ganglia Dysfunction
During Reward Anticipation in Adulthood. Biological Psychiatry, 66(3, 206-213.
doi:10.1016/j.biopsych.2009.02.019
Egerton, A., Valmaggia, L. R., Howes, O. D., Day, F., Chaddock, C. A., Allen, P., …
McGuire, P. (2016). Adversity in childhood linked to elevated striatal dopamine
function in adulthood. Schizophrenia Research, 176(2), 171–176.
http://doi.org/10.1016/j.schres.2016.06.005
Farr, O. M., Ko, B. J., Joung, K. E., Zaichenko, L., Usher, N., Tsoukas, M., ... &
Mantzoros, C. S. (2015). Posttraumatic stress disorder, alone or additively with early
life adversity, is associated with obesity and cardiometabolic risk. Nutrition,
Metabolism and Cardiovascular Diseases, 25(5), 479-488.
http://dx.doi.org/10.1016/j.numecd.2015.01.007
… Hans, P. (1998). Relationship of Childhood Abuse and Household Dysfunction to
Many of the Leading Causes of Death in Adults. American Journal of Preventive
Medicine, 14(4), 245–258. http://doi.org/10.1016/S0749-3797(98)00017-8
Finkelhor, D., Hamby, S. L., Ormrod, R., & Turner, H. (2005). The Juvenile
Victimization Questionnaire: Reliability, validity, and national norms. Child Abuse
& Neglect, 29(4), 383–412. http://doi.org/10.1016/j.chiabu.2004.11.001
Fishburn, P. C., & Rubinstein, A. (1982). Time Preference. International Economic

Review, 23(3), 677-694. doi: 10.2307/2526382
202
Frank, M. J., Seeberger, L. C., & O’Reilly, R. C. (2004). By carrot or by stick: cognitive
reinforcement learning in parkinsonism. Science, 306(5703), 1940–1943. doi:
10.1126/science.1102941
Frankenhuis, W., & Weerth, C. de. (2013). Does early-life exposure to stress shape or
impair cognition? Current Directions in Psychological Science, 22, 407–412.
doi: 10.1177/0963721413484324

http://doi.org/10.1016/j.neubiorev.2010.09.013
Gluckman, P. D., & Hanson, M. A. (2004). Living with the past: evolution, development,
and patterns of disease. Science, 305(5691), 1733-1736. doi
10.1126/science.1095292
Goeders, N. E. (2003). The impact of stress on addiction. European

Neuropsychopharmacology, 13(6), 435–441.
http://doi.org/10.1016/j.euroneuro.2003.08.004
Hall, F., Wilkinson, L., Humby, T., Inglis, W., Kendall, D., Marsden, C., & Robbins, T.
(1998). Isolation Rearing in Rats: Pre- and Postsynaptic Changes in Striatal
Dopaminergic Systems. Pharmacology Biochemistry and Behavior, 59(4), 859–872.
doi: http://doi.org/10.1016/S0091-3057(97)00510-8
Hall, F. S. (1998). Social Deprivation of Neonatal, Adolescent, and Adult Rats Has
Distinct Neurochemical and Behavioral Consequences. Critical Reviews in
Neurobiology, 12(1–2), 129–162. doi: 10.1615/CritRevNeurobiol.v12.i1-2.50
203
Hall, F. S., Wilkinson, L. S., Humby, T., & Robbins, T. W. (1999). Maternal deprivation
of neonatal rats produces enduring changes in dopamine function. Synapse, 32(1),
37–43. doi: 10.1002/(SICI)1098-2396(199904)32:1<37::AID-SYN5>3.0.CO;2-4
Hanson, J. L., Albert, D., Iselin, A.R., Carré, J.M., Dodge, K.A., & Hariri, A.R. (2016).
Cumulative stress in childhood is associated with blunted reward-related brain
activity in adulthood, Social Cognitive and Affective Neuroscience,11(3), 405-412.
doi: 10.1093/scan/nsv124
Hariri, A. R., Brown, S. M., Williamson, D. E., Flory, J. D., de Wit, H., & Manuck, S. B.
(2006). Preference for Immediate over Delayed Rewards Is Associated with
Magnitude of Ventral Striatal Activity. Journal of Neuroscience, 26(51), 13213–
13217. doi: http://doi.org/10.1523/JNEUROSCI.3446-06.2006
Health in Adulthood. Annals of the New York Academy of Sciences, 896(1), 85–95.
http://doi.org/10.1111/j.1749-6632.1999.tb08107.x
Hertzman, C. (2012). Putting the concept of biological embedding in historical

perspective. Proceedings of the National Academy of Sciences of the United States
of America, 17160–7. http://doi.org/10.1073/pnas.1202203109
Kahnt, T., Park, S. Q., Cohen, M. X., Beck, A., Heinz, A., & Wrase, J. (2009). Dorsal
Striatal–midbrain Connectivity in Humans Predicts How Reinforcements Are Used
to Guide Decisions. Journal of Cognitive Neuroscience, 21(7), 1332–1345.
http://doi.org/10.1162/jocn.2009.21092
Laucht, M., Skowronek, M. H., Becker, K., Schmidt, M. H., Esser, G., Schulze, T. G., &
Rietschel, M. (2007). Interacting effects of the dopamine transporter gene and
psychosocial adversity on attention-deficit/hyperactivity disorder symptoms among
204
15-year-olds from a high-risk community sample. Archives of General Psychiatry,

64(5), 585–90. http://doi.org/10.1001/archpsyc.64.5.585
Léonhardt, M., Matthews, S. G., Meaney, M. J., & Walker, C.-D. (2007). Psychological
stressors as a model of maternal adversity: Diurnal modulation of corticosterone
responses and changes in maternal behavior. Hormones and Behavior, 51(1), 77–88.
http://doi.org/10.1016/j.yhbeh.2006.08.008
Lighthall, N. R., Gorlick, M. A., Schoeke, A., Frank, M. J., & Mather, M. (2013). Stress
modulates reinforcement learning in younger and older adults. Psychology and
Aging, 28(1), 35–46. http://doi.org/10.1037/a0029823
Lovic, V., Keen, D., Fletcher, P. J., & Fleming, A. S. (2011). Early-life maternal
separation and social isolation produce an increase in impulsive action but not
impulsive choice. Behavioral Neuroscience, 125(4), 481-491. doi:
10.1037/a0024367
Mather, M., & Lighthall, N. R. (2012). Both Risk and Reward are Processed Differently
in Decisions Made Under Stress. Current Directions in Psychological Science,
21(2), 36–41. http://doi.org/10.1177/0963721411429452
Meaney, M. J. (2001). Maternal care, gene expression, and the transmission of individual
differences in stress reactivity across generations. Annual review of
neuroscience, 24(1), 1161-1192. doi 10. 1146/annurev.neuro.24.1.1161
Meaney, M. J., Brake, W., & Gratton, A. (2002). Environmental regulation of the
development of mesolimbic dopamine systems: A neurobiological mechanism for
vulnerability to drug abuse? Psychoneuroendocrinology, 27(1-2), 127–138.
http://doi.org/10.1016/S0306-4530(01)00040-3
Meaney, M. J. (2010). Epigenetics and the biological definition of gene× environment

interactions. Child development, 81(1), 41-79. doi 10.1111/j.1467-
8624.2009.01381.x
Mehta, M., Gore-Langton, E., Golembo, N., & Colvert, E. (2010). Hyporesponsive
205
Reward Anticipation in the Basal Ganglia following Severe Institutional Deprivation

Early in Life. Journal of Cognitive Neuroscience, 22(10), 2316-2325.
Mischel, W., Shoda, Y., & Rodriguez, M. (1989). Delay of gratification in children.
Science, 244(4907), 933–938. http://doi.org/10.1126/science.2658056
Moffitt, T. E., Arseneault, L., Belsky, D., Dickson, N., Hancox, R. J., Harrington, H., …
Caspi, A. (2011). A gradient of childhood self-control predicts health, wealth, and
public safety. Proceedings of the National Academy of Sciences of the United States
of America, 108(7), 2693–8. http://doi.org/10.1073/pnas.1010076108
Montague, P. R., Hyman, S. E., & Cohen, J. D. (2004). Computational roles for
dopamine in behavioural control. Nature, 431(7010), 760–767.
http://doi.org/10.1038/nature03015
Morgan, D., Grant, K. A., Gage, H. D., Mach, R. H., Kaplan, J. R., Prioleau, O., …
Nader, M. A. (2002). Social dominance in monkeys: dopamine D2 receptors and
cocaine self-administration. Nature Neuroscience, 5(2), 169–74.
http://doi.org/10.1038/nn798
Nelson, C. A. (2013). Biological embedding of early life adversity. JAMA

pediatrics, 167(12), 1098-1100. doi 10.1001/jamapediatrics.2013.3768
Noll, J. G., Shenk, C. E., & Putnam, K. T. (2009). Childhood sexual abuse and adolescent
pregnancy: A meta-analytic update. Journal of pediatric psychology, 34(4), 366-
378. https://doi.org/10.1093/jpepsy/jsn098
Non, A. L., Román, J. C., Gross, C. L., Gilman, S. E., Loucks, E. B., Buka, S. L., &
Kubzansky, L. D. (2016). Early childhood social disadvantage is
associated with poor health behaviours in adulthood. Annals of human
biology, 43(2), 144-153. http://dx.doi.org/10.3109/03014460.2015.1136357
Oswald, L. M., Wand, G. S., Kuwabara, H., Wong, D. F., Zhu, S., & Brasic, J. R. (2014).
History of childhood adversity is positively associated with ventral striatal dopamine
206
responses to amphetamine. Psychopharmacology, 231(12), 2417-2433.

http://doi.org/10.1007/s00213-013-3407-z
Pessiglione, M., Seymour, B., Flandin, G., Dolan, R. J., & Frith, C. D. (2006).
Dopamine-dependent prediction errors underpin reward-seeking behaviour in
humans. Nature, 442(7106), 1042–5. http://doi.org/10.1038/nature05051
Piazza, P., & Maccari, S. (1991). Corticosterone levels determine individual vulnerability
to amphetamine self-administration. Proceedings of the …. Retrieved from
http://www.pnas.org/content/88/6/2088.short
Piazza, P., & Maccari, S., Deminiére, J.M., Le Moal, M., Morméde, & Simon, H. (1991).
Corticosterone levels determine individual vulnerability to amphetamine self-
administration. Proceedings of the National Academy of Sciences, 88, 2088-2092.
doi: 10.1073/pnas.88.6.2088
Piazza, P., & Le Moal, M. (1996). Pathophysiological basis of vulnerability to drug

abuse: role of an interaction between stress, glucocorticoids, and dopaminergic
neurons. Annual Review of Pharmacology and Toxicology, 36, 359-378.
doi: 10.1146/annurev.pa.36.040196.002043
Pizzagalli, D. A., Bogdan, R., Ratner, K. G., & Jahn, A. L. (2007). Increased perceived
stress is associated with blunted hedonic capacity: potential implications for
depression research. Behaviour research and therapy, 45(11), 2742-2753.
http://dx.doi.org/10.1016/j.brat.2007.07.013
Pruessner, J. C., Champagne, F., Meaney, M.J., Dagher, A. (2004). Dopamine Release in
Response to a Psychological Stress in Humans and Its Relationship to Early Life
Maternal Care: A Positron Emission Tomography Study Using [11C]Raclopride.
Journal of Neuroscience, 24(11), 2825–2831.
http://doi.org/10.1523/JNEUROSCI.3422-03.2004
207
Schlam, T. R., Wilson, N. L., Shoda, Y., Mischel, W., & Ayduk, O. (2013). Preschoolers’
Delay of Gratification Predicts their Body Mass 30 Years Later. The Journal of
Pediatrics, 162(1), 90–93. http://doi.org/10.1016/j.jpeds.2012.06.049
Schneider, W., Eschman, A., & Zuccolotto, A. (2002). E-prime reference guide.
Psychology Software Tools .
Senn, T. E., Carey, M. P., Vanable, P. A., Coury-Doniger, P., & Urban, M. (2007).
Characteristics of sexual abuse in childhood and adolescence influence sexual risk
behavior in adulthood. Archives of Sexual Behavior, 36(5), 637-645. doi
10.1007/s10508-006-9109-4
Shoda, Y., Mischel, W., & Peake, P. K. (1990). Predicting adolescent cognitive and self-
regulatory competencies from preschool delay of gratification: Identifying
diagnostic conditions. Developmental Psychology, 26(6), 978–986.
http://doi.org/10.1037/0012-1649.26.6.978
Sinha, R. (2008). Chronic stress, drug use, and vulnerability to addiction. Annals of the
New York Academy of Sciences, 1141(1), 105-130. doi 10.1196/annals.1441.030
Snodgrass, J. G., & Vanderwart, M. (1980). A standardized set of 260 pictures: norms for
name agreement, image agreement, familiarity, and visual complexity. Journal of
experimental psychology: Human learning and memory, 6(2), 174.
http://dx.doi.org/10.1037/0278-7393.6.2.174
Sutton, R. S., & Barto, A. G. (1998). Reinforcement learning: An introduction (Vol. 1,

No. 1). Cambridge: MIT press.
van den Bos, W., & McClure, S. M. (2013). Towards A General Model of Temporal
Discounting. Journal of the Experimental Analysis of Behavior, 99(1), 58–73. doi:
http://doi.org/10.1002/jeab.6
Van Leijenhorst, L., Zanolie, K., Van Meel, C. S., Westenberg, P. M., Rombouts, S. A.,
& Crone, E. A. (2010). What motivates the adolescent? Brain regions mediating
reward sensitivity across adolescence. Cerebral cortex, 20(1), 61-69. doi
208
https://doi.org/10.1093/cercor/bhp078
Weller, J. A., & Fisher, P. A. (2012). Decision-Making Deficits Among Maltreated

Children. Child Maltreatment, 18(3), 184-194. doi: 10.1177/1077559512467846
Yarkoni, T., Poldrack, R. A., Nichols, T. E., Van Essen, D. C., & Wager, T. D. (2011).
Large-scale automated synthesis of human functional neuroimaging data. Nature
Methods, 8, 665–670. doi:10.1038/nmeth.1635
209
Chapter 5
5 General Discussion
5.1 Developmental plasticity and adverse life experiences
Developmental plasticity is the extent to which an individual’s current phenotype
is influenced by past experiences (Stamps, 2016). Thus, developmental plasticity
necessitates an underlying susceptibility to indicators of environmental quality. Initially,
individual differences in sensitivity to environmental factors were thought to reflect
genotypic differences that rendered some individuals more sensitive to environmental
variation, and others less sensitive (Pluess, 2015). More recently however, evidence
suggests that plasticity to the quality of the early-life environment may be induced by
environmental factors themselves and are not necessarily the result of genetically-derived
differences (Pluess, 2015). Regardless of the source of sensitivity to environmental
factors however, in developmentally plastic individuals, the quality of the early-life
environment can affect the neural and hormonal state of the organism at later stages of
development (Stamps, 2016).
Indeed, overwhelming evidence indicates that adversity experienced early in life

has long-term health-related consequences (Anda et al., 2006; Chapman et al., 2004;
Dube et al., 2003; Edwards et al., 2003; Felitti et al., 1998; Shonkoff, Boyce, & McEwen,
2009; Slavich & Irwin, 2014). These studies suggest that early-life experiences may
become “biologically embedded” in the developing organism (Berens, Jensen, & Nelson,
2017; Hertzman et al., 1999). The brain is the central organ that interprets and responds
to variation in environmental factors, including stressful experiences (McEwen, 2008).
Thus, when investigating developmental plasticity, it is important to consider plasticity of
regions within the nervous system that are involved in processing and responding to
stressful situations. In fact, substantial evidence suggests that the stress-response systems,
including the HPA axis, are plastic in the face of variation in environmental quality (Liu
et al., 1997; Meaney, 2010). In addition to the HPA axis, the mesolimbic dopamine
pathway has been proposed to be sensitive to indicators of environmental quality
(Gatzke-Kopp, 2011).
210
5.2 Summary of thesis and common themes

5.2.1 Aims of current thesis
The overall aim of this thesis then, was to explore whether and how the HPA axis
and the mesolimbic dopamine pathway exhibit developmental plasticity in the face of
adverse life-experiences. The empirical literature reporting associations between adverse
life-experiences and both cortisol and dopamine reactivity is mixed (See Chapters 2 and
3), with some studies reporting heightened cortisol and dopamine reactivity, and others
reporting blunted cortisol and dopamine reactivity. Thus, while there is a large body of
evidence linking exposure to adverse life-experiences to variation in cortisol and
dopamine reactivity, the direction of the change remains unclear. In this thesis, I
investigated whether differences in the severity of adverse life-experiences across studies
contributes to the mixed findings. To help resolve the mixed findings, I used narrow
definitions of trauma versus adversity to operationalize the terms “adversity” and
“trauma”. Specifically, trauma was defined using Criterion A of trauma-related disorders
using DSM-5 (e.g., war-related trauma, sexual abuse), and adversity was defined as
negative life events that did not fit the criteria for traumatic events (e.g., low SES, work-
related stress). In Chapter 2, I used p-curve to determine whether trauma and adversity
were differentially associated with variation in HPA axis reactivity. In Chapter 3, I
reviewed the animal literature to determine how stress, the HPA axis, and the mesolimbic
dopamine system form a physiological link; in addition, I reviewed the human PET
imaging literature to investigate whether trauma and adversity were differentially
associated with variation in dopamine reactivity. Finally, in Chapter 4, I conducted an
empirical study to assess whether individual differences in exposure to moderate levels of
adversity were related to variation in BOLD responses within regions of the mesolimbic
dopamine pathway. In addition, I examined whether differences in exposure to adversity
were related to reward-based learning and decision-making at the behavioural level.
5.2.2 Summary of findings

In Chapter 2, using p-curve meta-analytic methods, I found that the literature
reporting associations between trauma and blunted cortisol reactivity contained evidential
211
value; in contrast, the literature reporting associations between adversity and heightened
cortisol reactivity contained evidential value. The literature reporting that trauma was
associated with heightened cortisol reactivity was inconclusive; in this case, the p-curve
was too noisy to determine whether a very small effect exists or not. Finally, the literature
demonstrating that adversity was associated with blunted cortisol reactivity did not
contain evidential value, and in this case, we could not rule out p-hacking as an
explanation for the reported findings (Hosseini-Kamkar, Lowe, & Morton, 2021). Taken
together then, Chapter 2, provides evidence for an inverted U-shaped relationship
between severity of adversity and HPA axis reactivity, and highlights the importance of
distinguishing between trauma and adversity (Hosseini-Kamkar et al., 2021).
In Chapter 3, I conducted a systematic review of the animal models to examine

biological mechanisms that might explain the link between adversity, the HPA axis, and
the mesolimbic dopamine pathway. Indeed, across different studies and paradigms, a
physiological link between stress exposure, glucocorticoid release, and dopamine
reactivity is established. Overall, the evidence in animals suggests that stress exposure
and subsequent glucocorticoid release plays a substantial role in stress-induced
vulnerability to addictions (Campbell & Carroll, 2001; Goeders & Clampitt, 2002; Piazza
et al., 1991; Piazza et al., 1996). Importantly, in animal models, acute stressors are
associated with elevated dopamine release within the mesolimbic pathway (Abercrombie
et al., 1989; Valentini et al., 2011; See Douma & de Kloet, 2020; Holly & Miczek, 2016
for reviews). In contrast, chronic stressors appear to be related to blunted dopamine levels
(Gambarana et al., 1999; Miczek et al., 2011; Shimamoto et al., 2011; Mangiavacchi et
al., 2001; Pascucci et al., 2007; Imperato et al., 1992; Puglisi-Alegra et al., 1991; Enman
et al., 2015). In addition to the systematic review of the animal literature, I also
systematically examined the human PET imaging literature to assess whether trauma and
adversity were related to differing patterns of dopamine reactivity. Preliminary evidence
from this analysis demonstrated that exposure to moderate levels of adversity were
associated with a hyperdopaminergic profile, whereas exposure to severe trauma was
associated with a hypodopaminergic profile (See Chapter 3).
212
Finally, in Chapter 4, an empirical study was conducted to examine the

association between exposure to moderate levels of adversity and dopamine-related
behaviours, including impulsivity and reward-based learning and decision-making.
Moreover, I conducted an fMRI study to examine whether adversity was associated with
differences in BOLD responses in regions within the mesolimbic dopamine pathway
(Kamkar, Lewis, van den Bos, & Morton, 2017). The results demonstrated that greater
exposure to moderate levels of adversity were associated with heightened impulsivity and
learning from rewards, notably, loss-related learning was not associated with adversity.
Moreover, the fMRI results demonstrated that the association between adversity and
reward-related learning was partially mediated by ventral striatal reactivity to rewards
(Kamkar et al., 2017).
5.2.3 Adversity versus trauma

Across all three chapters, a consistent pattern of findings suggests that moderate
levels of adversity are associated with heightened HPA axis and dopamine reactivity,
whereas trauma is associated with blunted HPA axis and dopamine reactivity.
Importantly, when assessing the literature on adverse-life experiences broadly the results
of findings reporting associations between adverse life-experiences and HPA axis and
dopamine reactivity were highly conflicting (See Chapters 2 and 3). Indeed, the lack of
consensus regarding definitions of adversity and trauma may have contributed to the
mixed findings reported to date (Bunea et al., 2017; Fogelman & Canli, 2018; Krupnik,
2019). The term “adverse life-experiences” is typically used as a broad term and can
encompass a wide variety of stressors ranging from severely traumatic (e.g., sexual
abuse, severe violence) to moderate stressors (e.g., work-related stress, and changing
schools). Krupnik (2019) suggests that “adversity” and “trauma” should be differentiated
based on the severity of the stressor and their tendency to overwhelm self-regulatory
processes. In this thesis, trauma was defined based on Criterion A of DSM-5 trauma-
related disorders and includes events that can predispose individuals to developing
trauma-related disorders (e.g., PTSD). In contrast, adversity was defined as negative life-
experiences that did not fit the criteria for the development of trauma-related disorders.
This method allowed us to clarify some of the mixed reports in the literature and resulted
213
in a consistent pattern of findings across studies—with moderate adversity being

associated with elevated cortisol and dopamine levels, and severe trauma associated with
blunted levels.
5.3 Integration of findings with available theory

5.3.1 The Biological Sensitivity to Context (BSC) model
The biological sensitivity to context (BSC) model predicts a U-shaped relationship
between severity of adversity and stress reactivity—including HPA axis reactivity.
According to this model then, heightened cortisol reactivity is expected under conditions
of low and extreme adversity (Boyce & Ellis, 2005). The apparent paradox of heightened
stress reactivity under both protected environments and extremely threatening
environments is said to reflect heightened sensitivity to resources under protected
environments, and heightened vigilance to threats under extremely adverse environments
(Boyce & Ellis, 2005). According to the BSC model then, heightened cortisol reactivity
would be expected after exposure to traumatic life events, in contrast, blunted cortisol
reactivity would be predicted after exposure to moderate levels of adversity. In contrast to
the predictions of the BSC model, the findings reported in Chapter 2, provide evidence
for an inverted U-shaped relationship. Based on the findings of the p-curve meta-analyses
in Chapter 2, extremely adverse life-experiences (i.e., trauma) are associated with
blunted cortisol reactivity, while moderately adverse experiences are associated with
heightened cortisol reactivity (Hosseini-Kamkar et al., 2021). One possible explanation
for the lack of support for the BSC model may be that HPA axis reactivity is merely one
small aspect of “biological reactivity”, perhaps biological reactivity measured via other
methods including heart rate variability, respiratory sinus arrhythmia (RSA), or
preejection period (PEP) would obtain results in support of the BSC model. Moreover,
perhaps the range of experiences that the BSC makes predictions about are within
normative experiences; that is, perhaps adversity as defined according to the BSC ranges
from protected environments to moderately adverse environments, but not traumatic
environments. Nevertheless, the findings of Chapter 2 do not provide direct support for
the BSC model and in fact, provide evidence for the opposite pattern of findings as would
be expected based on the BSC.
214
5.3.2 Theories on the plasticity of the mesolimbic dopamine

pathway
Gatzke-Kopp (2011) proposed that in addition to the HPA axis, the mesolimbic
dopamine pathway demonstrates significant sensitivity to environmental influences;
however, the empirical evidence in relation to the directionality of dopaminergic change
remains inconsistent—with evidence present for both hyperdopaminergic and
hypodopaminergic profiles after exposure to adversity (Gatzke-Kopp, 2011). That said,
Gatzke-Kopp predicted that more severe and chronic stressors would be associated with a
hypodopaminergic profile favoring choices that are least costly in terms of effort
expenditure. Relatedly, Cabib and Puglisi-Allegra (2012) proposed that heightened
dopamine levels would be related to active coping strategies (e.g., escape, fight), whereas
blunted dopamine levels would be related to passive coping strategies (e.g., helplessness).
Evidence from animal models presented in Chapter 3 is consistent with these models in
that acute stressors were associated with elevated dopamine levels, while chronic
stressors were related to blunted dopamine levels. Moreover, if the assumption is made
that chronic stressors are also more severe, then the findings from human PET imaging
studies also provide evidence in support of these models; as demonstrated in Chapter 3,
in humans, moderate levels of adversity were related to heightened dopamine levels,
whereas trauma was associated with blunted dopamine levels.
5.3.3 Stress-response Adaptations to Variability in Early-life

Experiences (SAVE)
As mentioned previously, a consistent pattern of findings emerged across studies
and demonstrated that moderate levels of adversity were associated with heightened
cortisol and dopamine reactivity, while severe trauma was related to blunted cortisol and
dopamine reactivity. In terms of dopamine reactivity, Cabib and Puglisi-Allegra (2012)
proposed that heightened dopamine levels support active coping strategies in the face of
events appraised as stressors, while blunted dopamine levels support passive coping
strategies in the face of situations appraised as uncontrollable or unescapable. Given the
established link between the HPA axis, the mesolimbic dopamine pathway, and the
theoretical model proposed by Cabib and Puglisi-Allegra (2012), I propose the Stress-
215
response Adaptations to Variability in Early-life Experiences, or SAVE model. The SAVE

model predicts an inverted U-shaped relationship between severity of adversity and HPA
axis and dopamine reactivity. The SAVE model is based on the biological role that
glucocorticoid release plays in glucose metabolism and the tendency of the body to
“save” resources unless it is efficient and necessary to use them. Under stressful
conditions, glucocorticoid release facilitates the mobilization of energy resources in the
form of glucose to initiate “fight or flight” responses (Del Rey et al., 2008). Under
extremely traumatic situations during which “fight or flight” responses may not be
feasible, according to the SAVE model, it would not make sense to utilize energy
resources. In contrast, under moderately adverse situations, where “fight or flight” is
feasible with a good chance of success, it is adaptive for the body to secrete
glucocorticoids and mobilize energy resources. Indeed, the findings of this thesis are
consistent with the predictions made under the SAVE model, individuals exposed to
moderate levels of adversity had heightened cortisol levels (Chapter 2), heightened
dopamine levels (Chapter 3) and heightened ventral striatal BOLD responses to a reward-
based learning task (Chapter 4). In contrast, exposure to severe trauma was related to
blunted cortisol levels (See Chapter 2) and blunted dopamine levels (See Chapter 3).
5.4 Implications
5.4.1 Health consequences related to cortisol and dopamine
reactivity
Allostatic load refers to the potentially negative physiological consequences of
prolonged exposure to stressors via wear and tear on the body (McEwen & Stellar, 1993;
McEwen, 1998; McEwen & Wingfield, 2003). The primary mediators of allostasis are
said to be the HPA axis, catecholamines, and cytokines. While in the short term, the
function of the mediators of allostasis have protective effects for the organism, their
prolonged activation can have negative health consequences (McEwen & Wingfield,
2003). For example, chronically heightened levels of glucocorticoids impede insulin
function, and the combination of heightened glucocorticoids and insulin results in
increased fat storage and the formation of atherosclerotic plaques in coronary arteries
(McEwen & Wingfield, 2003). Moreover, according to allostatis load models, heightened
216
stress and cortisol reactivity have negative consequences for cognitive function (e.g.,
depression and Alzheimer’s disease), metabolism (e.g., diabetes and obesity),
cardiovascular function (e.g., endothelial cell damage and atherosclerosis), and immune
function (e.g., immune suppression; Juster, McEwen, & Lupien, 2010). Therefore,
according to allostatic load models, chronically elevated stress-responses, including
heightened cortisol reactivity, are related to the development of serious pathophysiology
(McEwen & Wingfield, 2003). Blunted cortisol levels have also been linked to adverse
health consequences. For example, blunted cortisol levels are not only associated with
PTSD, but have also been linked to chronic fatigue syndrome, fibromyalgia, rheumatoid
arthritis, and asthma (Heim, Ehlert, & Hellhammer, 2000).
Similarly, both heightened and blunted dopamine levels have been associated
with attention deficit hyperactivity disorder (ADHD; Gainetdinov & Caron, 2000; Giros
et al., 1996; Miller et al., 2012; van den Buuse, Linthorst, Versteeg, & de Jong, 1991;
Yousfi-Alaoui, Hospital, Garcia-Sanz, Badia, & Clos, 2001). Likewise, a
hypodopaminergic state (Diana, 2011; Mellis, Spiga, & Diana, 2005) and a
hyperdopamiergic state (Dagher & Robbins, 2009; Marinelli & White, 2000; Ritz &
Kuhar, 1993) have been implicated in susceptibility to the development of substance
abuse problems. While a direct assessment of the health consequences relating to cortisol
and dopamine levels were beyond the scope of this present thesis, based on available
literature, it appears that both heightened and blunted cortisol and dopamine levels are
associated negative health consequences. Thus, it seems counterintuitive to refer to the
plasticity of the HPA axis and mesolimbic dopamine pathway as “adaptive” responses.
What must be taken into consideration however, is that while there may indeed be
downstream health-consequences related to cortisol and dopamine reactivity, the
immediate response of the organism in the face of the stressor may be an adaptive
response in that it increases the evolutionary fitness of the individual (See Section 5.4.2).
5.4.2 Adaptive advantages of high-risk behaviours

Impulsivity, risk-taking, consumption of high-fat high-sugar foods, and risky
sexual behaviours are dopamine-mediated behaviours that may have serious negative
health-consequences in the long-term (Duffy et al., 2018; Felitti et al., 1998; Garrido,
217
Weiler, & Taussig, 2018; Stevens et al., 2011). However, while these behaviours may
have consequences for health and illness, they may also increase the reproductive fitness
of the organism. For example, life-history theory posits that after exposure to early-life
adversity, females may mature early to increase reproductive fitness. While this early
sexual maturation has health risks for the individual including the risk of sexually
transmitted infections, it increases reproductive fitness by ensuring that the individual has
an opportunity to pass on her genes (Belsky & Pleuss, 2013; Ellis, 2004). Similarly, while
impulsivity can pose a risk for the development of addictions and substance abuse
disorders, it may facilitate migration from unstable and adverse environments (Chen et
al., 1999).
The consumption of high-fat high-sugar foods may have serious risks for
cardiovascular health, however, it may be adaptive to consume these foods in resource-
deprived environments. Thus, while both elevated and blunted levels of cortisol and
dopamine have been linked to adverse health consequences, the plasticity of the HPA
axis and mesolimbic dopamine pathway is still “adaptive” in that it increases the
evolutionary fitness of the organism. For instance, according to the SAVE model, after
exposure to moderate levels of adversity, the organism perceives stressors as controllable
and escapable, and therefore, it is adaptive to release glucocorticoids, dopamine, and
energy resources to overcome the stressor (See Chapter 3). In contrast, after exposure to
severely traumatic adversity, the organism learns that stressors are likely uncontrollable
and inescapable, and thus, there is little utility to waste energy resources in a futile
attempt to escape the stressor—in this case, the individual exhibits a form of
“helplessness” in the face of acute stressors.
5.5 Limitations and future directions

An investigation of the developmental timing of exposure to adversity and trauma
was beyond the scope of this present thesis. This constitutes a major limitation in that any
discussion of developmental plasticity warrants a consideration of sensitive periods of
plasticity across development. Indeed, developmental psychologists assert that
susceptibility to environmental factors is greatest during the early years of development,
as the nervous systems exhibits the greatest plasticity early in life (Belsky & Pleuss,
218
2013; Blair & Raver, 2012; Ganzel & Morris, 2011). Some biologists even claim that
plasticity only occurs during sensitive periods early in development (Bateson, 2001);
according to this view, exposure to adversity during sensitive periods can permanently
alter physiological processes (Shonkoff et al., 2009). Moreover, much of the theory that
served as the foundation of this thesis emphasizes the need to take developmental timing
as well as the severity of the insult into consideration (Gatzke-Kopp, 2011; Krupnik,
2019). For instance, Krupnik (2019) recommends that adversity and trauma should be
differentiated based on both the severity of the insult, as well as their tendency to
overwhelm self-regulatory processes. Therefore, this model suggests that children—with
under-developed self-regulatory abilities—are most vulnerable to becoming traumatized
(Krupnik, 2019). Accordingly, this model takes both the severity of stressors as well
sensitive periods in development (i.e., periods during which self-regulation is under-
developed), into consideration. In this thesis however, trauma and adversity were
differentiated solely based on the severity of the insult and their tendency to predispose
individuals to developing trauma-related disorders (Criterion A of DSM-5 trauma-related
disorders). According to Krupnik (2019) however, even a moderately adverse life-event
can be “traumatic” to children if it overwhelms their self-regulatory processes. Thus,
future research must take both the severity of adversity as well as the timing of exposure
into consideration when addressing the impact of adversity versus trauma on the brain
and behaviour.
Although a major strength of this thesis was that we were able to tease apart the
impact of moderate adversity and severe trauma on cortisol and dopamine reactivity, a
limitation was that we used DSM-5 criteria to define traumatic events. Using DSM-5 to
define trauma has the obvious benefit of allowing for a narrow operationalization of
trauma; however, it introduces the confound of stress-related psychopathology into the
analyses. For example, many individuals that experience traumatic life events (as defined
by DSM-5 criteria) also meet the criteria for the development of PTSD and other stress-
related disorders. Therefore, it remains unclear whether exposure to traumatic events
themselves are related to blunted cortisol and dopamine reactivity, or whether stress-
related psychopathology is driving the effect. We were able to partially account for this
limitation in Chapter 2, by removing studies that explicitly reported psychopathology in
219
their samples (See Supplementary Materials); however, it is still possible that other
studies included in the analyses did not measure or report psychopathology. Future
studies should attempt to tease apart exposure to trauma versus stress-related
psychopathology by comparing cortisol and dopamine reactivity in individuals exposed
to trauma with and without stress-related disorders.
5.6 Final Remarks

The overarching goal of this thesis was to use a variety of review, meta-analytic,
and empirical methods to assess how exposure to adverse life-experiences are related to
the reactivity of the HPA axis and the mesolimbic dopamine pathway. A substantial body
of evidence exists linking adverse life-experiences to cortisol and dopamine reactivity;
however, the findings of the literature are conflicting in that some studies report
heightened reactivity after exposure to adversity, while others report blunted reactivity.
Using clear definitions of trauma and adversity based on DSM-5 criteria helped resolve
the apparent inconsistencies in the mixed findings to date. Specifically, the results
demonstrated that severe trauma is associated with blunted cortisol and dopamine
reactivity, while moderate adversity is related to heightened cortisol and dopamine
reactivity. Moreover, based on a review of the animal literature, a physiological link is
established between exposure to stress, glucocorticoid release, and dopamine reactivity—
thereby providing a mechanism that explains how different stressors are related to both
cortisol and dopamine reactivity. The results of Chapter 2 and 3 provide evidence for an
inverted U-shaped model that explains the link between severity of adversity and cortisol
and dopamine reactivity (SAVE model). Consistent with the predictions of the SAVE
model, Chapter 4 provides empirical evidence that demonstrates that exposure to
moderate levels of adversity is associated with dopamine-mediated behaviours—
including impulsivity and reward-based learning. Finally, the association between
adversity and reward-based learning was partially mediated by differences in the
functional reactivity of regions within the mesolimbic dopamine pathway. Overall, the
primary contribution of this thesis was to help resolve conflicting findings in a large body
of literature by using clear definitions of moderate adversity versus severe trauma.
220
5.7 References
Abercrombie, E. D., Keefe, K. A., DiFrischia, D. S., & Zigmond, M. J. (1989).

Differential Effect of Stress on In Vivo Dopamine Release in Striatum, Nucleus
Accumbens, and Medial Frontal Cortex. Journal of Neurochemistry, 52(5), 1655–
Anda, R. F., Felitti, V. J., Bremner, J. D., Walker, J. D., Whitfield, Ch., Perry, B. D.,
Dube, Sh. R., & Giles, W. H. (2006). The enduring effects of abuse and related
adverse experiences in childhood: A convergence of evidence from neurobiology
and epidemiology. European Archives of Psychiatry and Clinical Neuroscience,
256(3), 174–186. https://doi.org/10.1007/s00406-005-0624-4
Bateson, P. (2001). Fetal experience and good adult design a. International journal of
epidemiology, 30(5), 928-934.
Belsky, J., & Pluess, M. (2013). Beyond risk, resilience, and dysregulation: Phenotypic
plasticity and human development. Development and Psychopathology, 25(4 Pt
2), 1243–1261. https://doi.org/10.1017/s095457941300059x
Blair, C., & Raver, C. C. (2012). Individual development and evolution: experiential
canalization of self-regulation. Developmental psychology, 48(3), 647.
https://doi.org/10.1017/S0954579405050145
7(12), 1274. https://doi.org/10.1038/s41398-017-0032-3
221

Chapman, D. P., Whitfield, C. L., Felitti, V. J., Dube, S. R., Edwards, V. J., & Anda, R.
F. (2004). Adverse childhood experiences and the risk of depressive disorders in
adulthood. Journal of Affective Disorders, 82(2), 217–225.
Chen, C., Burton, M., Greenberger, E., & Dmitrieva, J. (1999). Population migration and
the variation of dopamine D4 receptor (DRD4) allele frequencies around the
globe. Evolution and Human Behavior, 20(5), 309-324.
Dagher, A., & Robbins, T. W. (2009). Personality, addiction, dopamine: insights from
Parkinson's disease. Neuron, 61(4), 502-510.
Diana, M. (2011). The dopamine hypothesis of drug addiction and its potential
therapeutic value. Frontiers in psychiatry, 2, 64.
Dube, S. R., Felitti, V. J., Dong, M., Chapman, D. P., Giles, W. H., & Anda, R. F. (2003).
Childhood Abuse, Neglect, and Household Dysfunction and the Risk of Illicit
Drug Use: The Adverse Childhood Experiences Study. Pediatrics, 111(3), 564–
572. https://doi.org/10.1542/peds.111.3.564
Edwards, V. J., Holden, G. W., Felitti, V. J., & Anda, R. F. (2003). Relationship Between
Multiple Forms of Childhood Maltreatment and Adult Mental Health in
Community Respondents: Results From the Adverse Childhood Experiences
Study. American Journal of Psychiatry, 160(8), 1453–1460.
https://doi.org/10.1176/appi.ajp.160.8.1453
Enman, N. M., Arthur, K., Ward, S. J., Perrine, S. A., & Unterwald, E. M. (2015).
Anhedonia, reduced cocaine reward, and dopamine dysfunction in a rat model of
Posttraumatic Stress Disorder. Biological Psychiatry, 78(12), 871–879.
222
https://doi.org/10.1016/S0749-3797(98)00017-8
Gainetdinov, R. R., & Caron, M. G. (2000). An animal model of attention deficit
hyperactivity disorder. Molecular medicine today, 6(1), 43.
Gambarana, C., Masi, F., Tagliamonte, A., Scheggi, S., Ghiglieri, O., & Montis, M. G. D.
(1999). A Chronic Stress that Impairs Reactivity in Rats Also Decreases
Dopaminergic Transmission in the Nucleus Accumbens: A Microdialysis Study.
Journal of Neurochemistry, 72(5), 2039–2046. https://doi.org/10.1046/j.1471-
4159.1999.0722039.x
Ganzel, B. L., & Morris, P. A. (2011). Allostasis and the developing human brain:
Explicit consideration of implicit models. Development and
psychopathology, 23(4), 955-974.
Giros, B., Jones, S. R., Wightman, R. M., Caron, M. G., & Jaber, M. (1996).
Hyperlocomotion and indifference to cocaine and amphetamine in mice lacking
the dopamine transporter. Nature, 379(6566), 606-612. doi: 10.1038/379606a0
https://doi.org/10.1007/s00213-002-1007-4
223
Heim, C., Ehlert, U., & Hellhammer, D. H. (2000). The potential role of hypocortisolism
in the pathophysiology of stress-related bodily
disorders. Psychoneuroendocrinology, 25(1), 1-35.
https://doi.org/10.1007/s00213-015-4151-3
Imperato, A., Angelucci, L., Casolini, P., Zocchi, A., & Puglisi-Allegra, S. (1992).
Repeated stressful experiences differently affect limbic dopamine release during
and following stress. Brain Research, 577(2), 194–199.
https://doi.org/10.1016/0006-8993(92)90274-D
Juster, R. P., McEwen, B. S., & Lupien, S. J. (2010). Allostatic load biomarkers of
chronic stress and impact on health and cognition. Neuroscience & Biobehavioral
Reviews, 35(1), 2-16.
Kamkar, N. H., Lewis, D. J., van den Bos, W., & Morton, J. B. (2017). Ventral striatal
activity links adversity and reward processing in children. Developmental
Cognitive Neuroscience, 26, 20–27. https://doi.org/10.1016/j.dcn.2017.04.002
Krupnik, V., (2019). Trauma or adversity? Traumatology: An International Journal,
25(4), 256–261. http://dx.doi.org.proxy1.lib.uwo.ca/10.1037/trm0000169
1662.
Mangiavacchi, S., Masi, F., Scheggi, S., Leggio, B., Montis, M. G. D., & Gambarana, C.
(2001). Long-term behavioral and neurochemical effects of chronic stress
224
exposure in rats. Journal of Neurochemistry, 79(6), 1113–1121.

https://doi.org/10.1046/j.1471-4159.2001.00665.x
Marinelli, M., & White, F. J. (2000). Enhanced vulnerability to cocaine self-
administration is associated with elevated impulse activity of midbrain dopamine
neurons. Journal of Neuroscience, 20(23), 8876-8885.
McEwen, B. S. (1998). Stress, Adaptation, and Disease: Allostasis and Allostatic Load.
Annals of the New York Academy of Sciences, 840(1), 33–44.
https://doi.org/10.1111/j.1749-6632.1998.tb09546.x
McEwen, B. S. (2008). Central effects of stress hormones in health and disease:
Understanding the protective and damaging effects of stress and stress mediators.
European Journal of Pharmacology, 583(2–3), 174–185.
https://doi.org/10.1016/j.ejphar.2007.11.071
McEwen, B. S., & Stellar, E. (1993). Stress and the Individual: Mechanisms Leading to
Disease. Archives of Internal Medicine, 153(18), 2093–2101.
https://doi.org/10.1001/archinte.1993.00410180039004
McEwen, B. S., & Wingfield, J. C. (2003). The concept of allostasis in biology and
biomedicine. Hormones and Behavior, 43(1), 2–15.
https://doi.org/10.1016/S0018-506X(02)00024-7
Meaney, M. J. (2010). Epigenetics and the Biological Definition of Gene × Environment
Interactions. Child Development, 81(1), 41–79. https://doi.org/10.1111/j.1467-
8624.2009.01381.x
Melis, M., Spiga, S., & Diana, M. (2005). The dopamine hypothesis of drug addiction:
hypodopaminergic state. Int Rev Neurobiol, 63(10), 101-54.
Miczek, K. A., Nikulina, E. M., Shimamoto, A., & Covington, H. E. (2011). Escalated or
Suppressed Cocaine Reward, Tegmental BDNF, and Accumbal Dopamine
Caused by Episodic versus Continuous Social Stress in Rats. Journal of
Neuroscience, 31(27), 9848–9857. https://doi.org/10.1523/JNEUROSCI.0637-
11.2011
Pascucci, T., Ventura, R., Latagliata, E. C., Cabib, S., & Puglisi-Allegra, S. (2007). The
Medial Prefrontal Cortex Determines the Accumbens Dopamine Response to
225
Stress through the Opposing Influences of Norepinephrine and Dopamine.

Cerebral Cortex, 17(12), 2796–2804. https://doi.org/10.1093/cercor/bhm008
Piazza, P. V., Maccari, S., Deminiere, J. M., Le Moal, M., Mormede, P., & Simon, H.
(1991). Corticosterone levels determine individual vulnerability to amphetamine
self-administration. Proceedings of the National Academy of Sciences, 88(6),
2088–2092. https://doi.org/10.1073/pnas.88.6.2088
Piazza, P. V., Rouge-Pont, F., Deroche, V., Maccari, S., Simon, H., & Le Moal, M.
(1996). Glucocorticoids have state-dependent stimulant effects on the
mesencephalic dopaminergic transmission. Proceedings of the National Academy
of Sciences, 93(16), 8716–8720. https://doi.org/10.1073/pnas.93.16.8716
Pluess, M. (2015). Individual Differences in Environmental Sensitivity. Child
Development Perspectives, 9(3), 138–143. https://doi.org/10.1111/cdep.12120
Puglisi-Allegra, S., Imperato, A., Angelucci, L., & Cabib, S. (1991). Acute stress induces
time-dependent responses in dopamine mesolimbic system. Brain Research,
554(1), 217–222. https://doi.org/10.1016/0006-8993(91)90192-X
Ritz, M. C., and M. J. Kuhar. "Psychostimulant drugs and a dopamine hypothesis
regarding addiction: update on recent research." In Biochemical Society
Symposium, vol. 59, pp. 51-64. 1993.
Shimamoto, A., DeBold, J. F., Holly, E. N., & Miczek, K. A. (2011). Blunted accumbal
dopamine response to cocaine following chronic social stress in female rats:
Exploring a link between depression and drug abuse. Psychopharmacology,
218(1), 271–279. https://doi.org/10.1007/s00213-011-2364-7
Shonkoff, J. P., Boyce, W. T., & McEwen, B. S. (2009). Neuroscience, Molecular
Biology, and the Childhood Roots of Health Disparities: Building a New
Framework for Health Promotion and Disease Prevention. JAMA, 301(21), 2252–
2259. https://doi.org/10.1001/jama.2009.754
Slavich, G. M., Irwin, M. R., & Link to external site, this link will open in a new
window. (2014). Monr. Psychological Bulletin, 140(3), 774–815.
Stamps, J. A. (2016). Individual differences in behavioural plasticities. Biological
Reviews, 91(2), 534–567. https://doi.org/10.1111/brv.12186
226
van den Buuse, M., Linthorst, A. C., Versteeg, D. H., & de Jong, W. (1991). Role of
brain dopamine systems in the development of hypertension in the spontaneously
hypertensive rat. Clinical and experimental hypertension, 13(5), 653-659.
Yousfi-Alaoui, M. A., Hospital, S., Garcia-Sanz, A., Badia, A., & Clos, M. V. (2001).
Presynaptic Modulation of K+-Evoked [3H]Dopamine Release in Striatal and
Frontal Cortical Synaptosomes of Normotensive and Spontaneous-Hypertensive
Rats. Neurochemical research, 26(12), 1271-1275. doi:
10.1023/A:1014237215200
.
227
Appendices
Appendix A: Supplementary Results for Chapter 2
Cortisol hypo-reactivity in healthy participants– removed PTSD/MDD Patients
From the original 48 studies included, 4 were removed because the sample included
participants with PTSD/MDD. Therefore, 44 studies were entered that that reported
results showing that a history of traumatic life experiences was related to cortisol hypo-
reactivity in healthy patients. We entered these results into p-curve to determine whether
this set of literature contains evidential value. In this case, the two conditions that
determine whether a p-curve is significantly right-skewed were met; therefore, as shown
in Figure x, the p-curve for this set of results was significantly right-skewed (half: z = -
5.5, p < .00001; full: z = -6.53, p < .00001). As shown in Figure x, these results do not
change if we remove the highest p-values and the conclusion drawn from the p-curve
would remain unaltered.
Figure 3. Cortisol hypo-reactivity p-curve. The blue line shows the distribution of p-values from the data (44 statistically
228
significant results). For example, 43% of the significant p values were below .01. The red dotted line shows the expected
distribution of p values if there were no effect. The green dashed line shows the distribution of p values if an effect existed
with studies powered at 33%. N = 44 p values, p-curve is significantly right-skewed.
Figure 4. Dropping the lowest and highest p values on the significance of right-skewness of the full p-curve (top row),
half p-curve (middle row), and the test for flatness relative to 33% power (bottom row). The red solid line demonstrates
the significance threshold of p = .05. The filled circle demonstrates the results reported in the text, the open circle
demonstrates the results relative to the significance level if (k) number of p-values were dropped.
An additional test was conducted to remove all studies that had used any patient
samples (e.g., anorexia, chronic fatigue syndrome etc.), therefore, this analysis included
studies with only healthy participants. Forty results were entered into p-curve and once
229
again, this set of results contained evidential value as the p-curve was significantly right-
skewed (half: z = -6.19, p < .00001; full: z = -6.47, p < .00001). As shown in Figure x,
these results do not change if we remove the highest p-values and the conclusion drawn
from the p-curve would remain unaltered.
230
Cortisol hyper-reactivity p-curve in healthy participants
We removed 6 studies that included patients with PTSD/MDD from the original
22 studies. The results of the p-curve analysis did not change. Neither of the conditions of
p-curve were met, and the p-curve was not significantly right-skewed (half: z = -0.25, p =
.3996; full: z = -1.48, p = .0692). P-curve indicates that evidential value is inadequate or
absent if the 33% power test is p< .05 for the full p-curve or if both the half p-curve and
binomial 33% power test are p <.1. Here, neither condition was met, the test for 33%
power was not significant (half: z = -2.63, p = .9957; full: z = -0.93, p = .1766; Binomial
test p = .5238). Therefore, like the results obtained with all data entered, the p-curve for
the subset of studies that were entered (excluding those with patients with PTSD/MDD)
was inconclusive.
231
Figure 5. Cortisol hyper-reactivity p-curve. The blue line shows the distribution of p-values from the data (16 statistically
significant results). For example, 19% of the significant p values were below .01. The red dotted line shows the expected
distribution of p values if there were no effect. The green dashed line shows the distribution of p values if an effect existed
with studies powered at 33%. N = 16 p values, p-curve is not significantly right-skewed.
232
Figure 6. Dropping the lowest and highest p values on the significance of right-skewness of the full p-curve (top row),
half p-curve (middle row), and the test for flatness relative to 33% power (bottom row). The red solid line demonstrates
the significance threshold of p = .05. The filled circle demonstrates the results reported in the text, the open circle
demonstrates the results relative to the significance level if (k) number of p-values were dropped.
An additional test was conducted to remove all studies that had used any patient
samples (e.g., social anxiety disorder, fibromyalgia etc.), therefore, this analysis included
studies with only healthy participants. Thirteen results were entered into p-curve and
once again, the conclusions previously drawn were not altered. This set of results did not
contain evidential value as the p-curve was not significantly right-skewed (half: z = -0.33,
p = .3713; full: z = -1.33, p = .091). The test for 33% power was also not significant (half:
z = -2.43, p = .9924; full: z = -0.85, p = .1985; Binomial test p = .5239). As shown in
233
Figure x, these results do not change if we remove the highest p-values and the
conclusion drawn from the p-curve would remain unaltered.
234
Trauma-Cortisol Hypo-Reactivity Literature
Random-Effects Model
k Hedges g Tau2 I2 H2 Q
48 0.159 0.202 (SE = 0.051) 88.42% 8.64 Q(47) = 330.81, p <

.001
235
Trauma-Cortisol Hyper-Reactivity Literature
22 0.426 0.022 (SE = 0.018) 43.29% 1.76 Q(21) = 37.302, p

=.016
Adversity-Cortisol Hypo-Reactivity Literature
16 0.027 0.116 (SE = 0.057) 79.19% 4.80 Q(15) = 59.940, p <

.001
Adversity-Cortisol Hyper-Reactivity Literature
15 0.380 0.033 (SE = 0.026) 52.60% 2.11 Q(14) = 30.038, p =

.008
236
Appendix B: Supplementary Table for Chapter 3
Appendix B-1: Summary of all studies reporting associations between adverse life-experiences and dopaminergic function
Original Quoted text from Sample Type of Adversity/ Reactivity Measure Measure of Brain Regions Quoted text from Hyper/ Conclusion quoted from
Paper original paper Characteristics Trauma Dopamine (DA) original paper Hypo original text
indicating with statistical DA
question/ results
prediction of
interest
Bloomfield et Given the findings We recruited two TRAUMA: Psychosocial stress [18F]-DOPA Striatum and sub- Dopamine synthesis Hypo Our main finding is that chronic
al., 2019 of dopaminergic groups of healthy Psychosocial task - Montreal regions capacity was exposure to psychosocial stressors is
dysfunction volunteers, one Adversity High Imaging Stress Task significantly reduced associated with significantly
associated with exposed to multiple Adversity - HA (MIST) -The in HA compared with reduced striatal dopamine synthesis
childhood risk factors (exposed (exposed) group rationale for LA (t32= 2.27, capacity, particularly in the limbic
maltreatment high adversity group, included at least one conducting the PET p=0.03). (ventral) striatum. In addition, we
presented above, ‘HA’)) and one not childhood stressor scan on the same found evidence that striatal
we hypothesized exposed (unexposed and at least two adult day as the stress dopamine synthesis capacity
that healthy low adversity group, stressors. Measured task was to reduce correlated with both the
humans with a high ‘LA’), from throughout via Childhood Trauma the variance in the physiological and subjective
cumulative expo- the UK via public Questionnaire (CTQ) time between the responses to an acute psychosocial
sure to psychosocial advertisement, and the Childhood measures. stressor. Chronic stress exposure is
stressors would newspaper Experiences of Care associated with a dissociation
have altered striatal advertisement and and Abuse (CECA). between physiological and
dopamine synthesis, national media Childhood stressors: psychological acute stress responses
compared to engagement. in the form of an attenuated stress-
humans with a low Responding induced increase in blood pressure
exposure individuals were then alongside a potentiated stress-
screened via induced subjective response. These
237
telephone. LA findings support our hypothesis that

‘controls’ were high cumulative exposure to
individually matched psychosocial adversity would be
to the HA group on associated with altered dopamine
the basis of age (+ /- 5 synthesis capacity
years) and sex.
Dahoun et al We hypothesized Healthy participants TRAUMA- assessed Dex-amphetamine- [11C]-PHNO PET data Ventral Striatum Unexpectedly, we Hypo Childhood trauma and ventral
2019 that the interaction (n=24) with the self- induced dopamine acquisition. D3 preferring found a negative striatal dopamine release interact to
and/or mediation completed CTQ 25. release D2/D3 agonist . correlation between influence the emergence of positive
between childhood The CTQ is a childhood trauma psychotic symptoms following
trauma and retrospective self- load and ventral dexamphetamine. This is not
dopamine release completed striatal dopamine consistent with a simple dopamine
capacity would questionnaire release (Pearson’s sensitization model for the link
predict covering the r=−0.488, p=0.016) between childhood trauma and
dexamphetamine- following five (Supplementary Fig. psychosis risk, but could, instead,
induced positive domains of childhood 1) suggest that childhood trauma
psychotic trauma: sexual abuse, moderates the cognitive response
symptoms, in light physical abuse, to dopamine release to make
of the sensitization emotional abuse, psychotic experiences more likely.
of the mesostriatal physical neglect and
dopamine system emotional neglect.
described above.
Sebold et al We assessed the The final sample ADVERSITY: Life None [18F]-fallypride (FP) Striatum- The The linear model Hyper Our main finding was a significant
2019 effect of stressful consisted of 20 Stress -subjects striatum is revealed that the correlation between striatal
life events as an alcohol-dependent completed several functionally divided main effect of stress dopamine D2/D3 receptor
indicator of chronic patients and 19 HC questionnaires, into three parts: the was not significant availability and stressful life events
stress exposure on matched for gender, including the Social sensorimotor (F=0.339, p=.5), in detoxified patients with AD but
striatal dopamine age, handedness, Readjustment Rating striatum (involved in whereas the inter- not in HC.
D2 receptor nicotine consumption, Scale, where subjects locomotion), the action between
availability and education have to indicate limbic striatum group and stress
measured via whether any of 43 (ventral striatum,
238
positron emissions potentially stressful involved in drive and was significant

tomography (PET) life events occurred motivation) and the (F=4.77, p=.035).
with the radiotracer within the last 12 associative striatum
[18F]-fallypride (FP) months. Each life [central striatum,
in detoxified AD event is associated involved in cognition
patients and with a specific
matched healthy amount of life change
controls (HC). units (LCUs), ranging
from 11 LCUs for
“minor violation of
the law” to 100 LCUs
for “death of a
spouse”
Schifani et al The current study Forty-two participants ADVERSITY: Recent Psychosocial stress 11C-FLB457 (measure of dlPFC and mPFC (both In CHR, but not in Hypo These findings provide first direct
2018 aimed to examine (84 scans) were Life Events task - MIST: In extrastriatal D2 receptors). hemispheres) and healthy volunteers evidence of a disrupted
stress-induced PFC included in this study, questionnaire (RLE), brief, subjects cerebellum as or schizophrenia, prefrontocortical dopamine-stress
dopamine response comprising 14 and the Trier perform mental comparison region chronic stress (total regulation in schizophrenia. Our
in patients with patients with Inventory of the arithmetic TICS score) was results suggest that PFC dopamine
schizophrenia and schizophrenia, 14 Assessment of presented on a significantly release in response to stress in
those at clinical high individuals at CHR and Chronic Stress (TICS). computer screen, negatively healthy volunteers and CHR, but not
risk (CHR) for 14 matched healthy which also displays associated with in schizophrenia, is associated with
psychosis using a volunteers. However, information about BPND specifically in salivary cortisol response, implying
validated two-scan two healthy the total number of mPFC (Fig. 5A;r= abnormal PFC stress regulation in
paradigm with11C- volunteers had to be errors, expected 0.62,P=0.018). schizophrenia. In addition,
FLB457 PET. We excluded from the aver-age number of Furthermore, in CHR individuals at CHR with higher
hypothesized a analysis be-cause of errors, time spent the number of distress and anxiety had lower
reduction in mPFC excessive head on the current stressful life events dopamine release in mPFC and
and dlPFC motion that could not problem, and (RLE) was negatively salivary cortisol response following
dopamine release in be corrected. All performance associated with the the stress challenge, associations
response to the patients were feedback for each BPND in mPFC (Fig. that were absent in the
stress challenge antipsychotic-free, problem (correct, 5B; r = 0.63,P=0.015) schizophrenia group.
corroborated by clean incorrect, time out).
239
urine drug screens, All subjects

with eight of them completed six
also being blocks of
antipsychotic-naive arithmetic, each
approximately 6min
in length, while
lying in the scanner.
Bloomfield et al We therefore We compared TRAUMA: Psychosocial stress Dopamine synthesis capacity Striatum The High Adversity Hypo Long-term exposure to psychosocial
2017 sought to examine dopamine synthesis Psychosocial task - MIST was indexed as the influx group had reduced adversity and stressors is associated
whether long-term capacity in n = 17 Adversity High rate constant using 3,4- dopamine synthesis with reduced striatal dopamine
exposure to human participants Adversity - HA dihydroxy-6-[18F]-fluoro-l- capacity in the synthesis capacity. Our finding of a
psychosocial with a high (exposed) group phenylalanine ([18F]-DOPA) striatum (effect size negative relationship between
adversity was cumulative exposure included at least one positron emission d = 0.80, t = 2.27,p = depressive symptoms and limbic
associated with to psycho-social childhood stressor tomography (PET) 0.03) and its limbic dopamine synthesis capacity
alterations in adversity (“High and at least two adult (effect size d = 0.95, suggests that long-term exposure to
striatal dopamine Adversity ”group) with stressors. Measured t = 2.64, p = psychosocial adversity may increase
synthesis capacity n = 17 participants via Childhood Trauma 0.01)and associative the risk of mental illness by
as well as with low cumulative Questionnaire (CTQ) subdivisions (effect dampening the dopamine system.
psychological and psychosocial adversity and the Childhood size d = 0.81, t = Likewise, our finding that in
physiological exposure (“low Experiences of Care 2.28, p =0.03) participants with a high exposure to
responses to acute Adversity” group) who and Abuse (CECA). compared with Low psychosocial adversity acute stress
psychosocial stress. were age- and sex- Childhood stressors Adversity results in blunted physio-logical yet
matched. potentiated subjective responses to
acute psychosocial stress suggests
that chronic exposure to
psychosocial adversity disrupts
these mechanisms.
Egerton et al The Canadian study CANADA STUDY: The ADVERSITY- Psychosocial stress CANADIAN STUDY: [11C]- Striatum CANADIAN STUDY: Hyper These results indicate that striatal
2017 investigated immigrant group (N = Immigration task - MIST (+)-PHNO PET Image Immigrants DA function is elevated in both
whether the 26) included first (N = Acquisition. All 56 subjects demonstrated immigrants and their children,
240
induction of stress 9) and second- completed 2 PET scans (n = elevated striatal DA including those at risk for psychosis
by a validated generation 112 PET scans) at least a release in response or with schizophrenia, confirming its
laboratory immigrants (N = 8) to week apart at the same time to stress compared relevance for psychotic disorders.
psychosocial task Canada. The of the day, one while per- to nonmigrant The elevation in DA in immigrants
(Montreal Imaging nonimmigrant group forming the SMCT (sensory Canadians (F = 8.08; was present with relatively large
Stress Task, MIST) (N = 31) had been in motor control task) and one df = 1, 52; P = .006; effect size in both the Canada and
elicits more DA Canada for at least 3 while undergoing the MIST partial eta2 = 0.13). UK studies, which were performed
release in generations. UK (stress task). Stress-induced UK STUDY: Striatal in independent samples using 2
immigrants (first STUDY: The DA release was measured DA synthesis complementary approaches to
and second immigrant group (N = using [11C]-(+)-PHNO capacity was imaging presynaptic DA function.
generation) as 31) included first (N = positron emission elevated in the
compared to the 13) and second- tomography (PET), through immigrant
host population. generation (N = 18) quantification of the compared to the
The UK study immigrants to the competition between nonimmigrant
sought to confirm United Kingdom. The endogenous DA and [11C]- group (F = 4.95; df =
the association nonimmigrant group (+)- PHNO for D2/3 receptor 1, 73; P = .03, partial
between (N = 44) had been in binding in the striatum. UK eta2 = 0.06, Figure
immigration (1rst the United Kingdom STUDY: [18F]- DOPA 2).
and second for at least 3
generation) and generations. The
elevated striatal DA immigrant and
function in a nonimmigrant groups
different cohort. were comparable for
demographic
Kasanova et al The current study, The sample consisted TRAUMA: Childhood Psychosocial stress [18F]Fallypride - (D2/D3 mPFC and vmPFC Within the control Hyper We examined the association
2016 therefore, aimed to of 12 HV (healthy trauma was task - MIST receptor antagonist) group, a significant between childhood trauma and
investigate the volunteers) and 12 measured using displacement positive association stress-induced prefrontal DA activity
effect of childhood unmedicated NAPD Childhood Experience emerged between of healthy individuals and patients
adversity on (non-affective of Care and Abuse the spatial extent of with psychotic disorder using
DAergic stress psychotic disorder) (CECA-Q), a validated, stress-induced tracer [18F]fallypride PET. We observed a
processing in frontal matched on age, retrospective displacement in the significant difference in the
cortical areas in gender and education questionnaire to mPFC and early association between childhood
241
non-medicated described in detail assess childhood childhood trauma (b trauma score and spatial extent of
patients with non- previously. All NAPD trauma in early = 7.23, t(11) = 3.06, stress-induced prefrontal DA activity
affective psychotic were currently off childhood spanning p = .016; Figs 2 and in each group; In healthy subjects,
disorder (NAPD) as antipsychotic from 0 to 11 years of 3) and late severity of childhood trauma was
well as in healthy medication (AP) for age, and late childhood trauma associated with more extensive
volunteers (HV) in longer than one year, childhood scores (b = 5.47, stress-related DA activity in mPFC.
order to further did not currently use encompassing years t(11) = 2.54, p = This effect was especially
elucidate the mood stabilizers, 12 through 17. For .035; Fig 2). pronounced in relation to early
DAergic antidepressants or the purpose of this childhood trauma, and largely
contribution to both benzodiazepines. study, a composite driven by DA activity in the ventral
vulnerability as well score was created for portion of mPFC. Contrarily, in the
as resilience to each time period patient group, there was no
psychosis. using 15 association between childhood
dichotomous (‘yes’ = trauma and the spatial extent of
1 and ‘no’ = 0) items stress-related DA activity in this
informing about region, and this was the case for its
family arrangements, ventral and dorsal portions, as well
parental loss, physical as for early and late childhood
and sexual abuse, trauma.
neglect and bullying.
Egerton et al The aim of this The sample TRAUMA: Childhood None Presynaptic dopamine Associative striatum: Dopamine function Hyper These findings provide evidence
2014 study was to consisted of 47 adversity was synthesis capacity (Kicer) in The associative was elevated in that exposure to childhood
examine how individuals who met assessed using the the associative striatum was striatum comprises participants who had psychosocial stressors that increase
exposure to operationalized Childhood Experience estimated using 18F-DOPA the dorsal caudate experienced severe risk of developing psychosis in later
childhood adversity criteria for Ultra of Care and Abuse positron emission and precomissural sexual or physical life is associated with elevated
impacts on brain High Risk of Questionnaire [4]. tomography. The impact of putamen. abuse in childhood striatal dopamine function in early
dopamine function psychosis (UHR) (age Events (Table 1) were exposures on dopamine compared to those adulthood. The elevation was
in young adults. We 23.6 ± 4.6 years, rated as present function was determined who had not T63 = significant following exposure to
hypothesised that 57% male) and 20 (exposure) or absent using independent samples 2.92; P=0.005), and severe (sexual or physical) abuse, or
A) a history of healthy volunteers (no exposure). * t-tests exposure versus non- in those who had to multiple family arrangements,
childhood adversity (Control, age 23.8 ± Events included exposure). experienced which may be a marker for
would be associated 4.3 years, 60% male) . multiple family unknown stressors, including abuse.
242
with increased physical and sexual arrangements

striatal dopamine abuse compared to those
function in who had not T57 =
adulthood; B) this 2.80; P=0.007). Both
relationship would findings were above
be particularly the corrected level
evident in people at for statistical
high risk for significance
psychosis. P=0.0125).
Mizrahi et al Psychosocial stress Seventeen non- ADVERSITY: Psychosocial stress We measured dopamine The whole striatum, A statistical model Hyper This result is, to our knowledge, the
2014 (e.g.marginalization, immigrants (9 healthy Immigration task - MIST release in the striatum using associative striatum (including clinical first report of stress-induced
social defeat, etc) volunteers(HV), 4 a previously described (AST), limbic striatum vulnerability (HV, dopaminergic sensitization in
may sensitize the patients with positron emission (LST) and CHR, SCZ) and immigrants.
brain dopaminergic schizophrenia (SCZ) tomography (PET) sensorimotor immigration as
system, changing its and 4 subjects at displacement paradigm striatum (SMST), with predictors)
reactivity and clinical high risk(CHR) involving scans under stress cerebellum as a explained 39% of the
increasing the of developing the and control conditions with reference region variance in stress-
amount of disease) and fifteen the tracer [11C]-(+)-PHNO. induced dopamine
dopamine released first and second Binding potential relative to release in the whole
during stressful generation immigrant non-displaceable ligand striatum (F=5.98,
situations. Since subjects (3 HV, 5 SCZ (BPND) was quantified using df=3,28 p=0.002),
psychosis has been and 7 CHR), aged 18- the simplified reference with a significant
linked to altered 35 years and matched tissue model. vulnerability group
striatal dopamine for age, and gender effect (F=3.49,
release in response df=2,28 p=0.04)
to both confirming previous
amphetamine and findings (Mizrahi et
stress, this suggests al., 2012), and a
a pathophysiological significant
model in which immigration effect
elevated stress- (F=5.97, df=1,28
induced dopamine p=0.02). In the
243
in migrants entire striatum,

predisposes this displacement in
population to immigrant subjects
schizophrenia was significantly
higher compared to
the non-immigrant
subjects (8.06% and
0.96%, respectively.
Oswald et al Our primary Twenty-eight healthy TRAUMA: Early Psychostimulant Binding potential (BPND) of Striatum A positive Hyper To our knowledge, this study
2014 hypotheses were male (n = 19) and Trauma Inventory Reactivity - AMPH [11C]raclopride was relationship was represents the first examination of
that childhood female (n = 9) Short Form—The 27- (0.3 mg/Kg). obtained by the reference found between ETI relationships between childhood
trauma would be participants, aged 18– item ETISR-SF is a tissue graphical analysis scores and VS DA adversity and mesolimbic DA
positively 29 years, were self-report measure (RTGA) (Logan et al. 1996) release in the responses to AMPH in humans.
associated with 1) recruited from the that assesses for striatum subdivisions. sample as a whole Results indicated that the number of
VS AMPH- induced Baltimore traumatic events that Then, intrasynaptic DA (Table 2), suggesting ACEs that participants experienced
DA release, 2) metropolitan area by occurred before the release, which represents that individuals who was positively associated with their
pleasant drug newspaper age of 18 years in the displacement of experience a greater DA responses to AMPH, as well as
effects, and 3) advertisements, fliers, each of four domains: [11C]raclopride by number of adverse with levels of perceived stress that
current levels of and internet postings. general trauma (11 endogenous DA was events in childhood they experienced as adults.
perceived stress. low- risk) participants items), physical obtained. have greater VS DA
for analysis. punishment (5 responses to AMPH
more sensitive to items), emotional (r(26) = 0.3895).
persistent effects of abuse (5 items), and
later adversities on sexual abuse (6
the HPA-axis. items).
Oswald et al The purpose of the Twenty healthy TRAUMA: Trauma Psychostimulant Subjects underwent two Ventral striatum - left Total ETI scores Hyper This report provides the first
2011 present study was adults (M = 14), ages Inventory (ETISR-SF), Reactivity – AMPH consecutive 90-minute PET (LVS) and right (RVS) correlated with DA evidence of relationships between
to determine 18 – 29 years, which evaluates four studies with [11C]raclopride subdivisions. release in the VS (r early trauma and stimulant-induced
whether childhood completed diagnostic types of trauma (RAC). The first scan was (18)= 0.49; p = 0.03) DA release in humans. Although
trauma is associated screening and a (general, physical, preceded by intravenous and RVS (r = 0.49; p causal relationships cannot be
with dopaminergic battery of self- saline; the second by 0.3 = 0.03); scores on definitively established on the basis
244
or subjective effects report measures of emotional, and mg/kg AMPH. Dopamine the emotional of this correlational study, findings
of amphetamine personality, mood, sexual). (DA) release was defined as trauma subscale support notions that early childhood
(AMPH) in humans. anxiety, and stress. percent change in RAC BP correlated with DA trauma may increase vulnerability
On the basis of the between scans release in the VS (r = for drug abuse by altering the
rodent studies, as 0.45; p = 0.047) and function of dopaminergic incentive-
well as prior LVS (r = 0.65; p = motivational neurocircuitry in the
evidence from our 0.002). The only brain.
group and others associations
which has shown observed between
that greater early trauma and
responsivity to baseline D2 receptor
acute stressors is availability were on
associated with the physical trauma
enhanced responses subscale (r = .45; p =
to .049 for both VS and
psychostimulants, RVS).
we hypothesized
that subjects who
reported greater
exposure to adverse
events as a child
would show
increased sensitivity
to AMPH.
Casey et al Animal models Twenty healthy ADVERSITY - family Psychostimulant 11C]raclopride Dorsal putamen Regression maps Hyper The results indicate that early family
2010 suggest that these stimulant drug naïve functioning measured Reactivity - d- identified bilateral functioning influences dopamine
early life stressors subjects (age via Family amphetamine (0.3 clusters of responses to a first dose of d-
can lead to 20.5±2.2, 11 male). Environment Scale. mg/kg, p.o.) vs. correlation in the amphetamine. Since striatal
increased drug- Two of these placebo scan dorsal putamen for dopamine is thought to influence
induced dopamine subscales, Moral- the Moral-Religious drug taking behaviour, this may
responses in Religious Emphasis subscale (Figure 1: identify a mechanism by which early
adulthood. To test and Family Cohesion, Left: 312 mm3 R= -
245
this hypothesis in predict vulnerability 0.71, p<0.05. Right: life difficulties affect vulnerability to
humans we to addiction. 304 mm3 R= -0.72, addiction.
assessed whether p<0.05). Subjects
individual who rated their
differences in early families as being less
family functioning concerned with
would predict morality had larger
differences in dopaminergic
amphetamine- responses to the test
induced striatal dose of
dopamine release. amphetamine.
Pruessner et al We hypothesized 10 subjects (1 woman ADVERSITY: Psychosocial stress [11C]raclopride binding Striatum We found a Hyper The psychosocial stressor caused a
2004 that low self- and 9men; 21–30 Adversity - Parental task - Psychological potential (BP) was significant effect of significant release of dopamine in
reported early life years of age) were Care measured via stress was induced calculated using a simplified the perceived the ventral striatum as indicated
parental care would selected as Parental Bonding using a mental reference tissue maternal care on by a reduction in [11C]raclopride
be associated with representative of Index (PBI Scale): arithmetic task on compartmental model stress-induced binding potential in the stress
higher cortisol and individuals in the top consists of four sub- the basis of a changes in versus resting condition in
dopamine levels in 25% (scores 37) and scales (mother care, previous [11C]raclopride BP in subjects reporting low parental
response to the bottom 10% (scores mother psychosocial stress the ventral striatum. care.
psychosocial stress 20) of the distribution overprotection, paradigm Post hoc analysis
task. on the PBI maternal father care, and developed to revealed a
care scale (population father over- investigate significant decrease
mean score, 31.65 protection) hormonal stress in [11C]raclopride BP
7.6; range, 12– 42) comprised of 12 responses. during the stress
and invited to questions each. condition in the low,
participate in the PET but not high,
study. maternal care
group (F(1,8) =
7.78;p = 0.02) (Fig.
2).
246
Appendix B-2: EMBASE Search Terms

EMBASE
Search Concept Search Line Search Strategy Number
Adversity 1 Trauma.mp. 372638

2 Adversity.mp. or childhood adversity/ 10745
3 child abuse"/ or Maltreatment.mp. 34894
4 PTSD.mp. or posttraumatic stress disorder/ 67026
5 life stress/ or chronic stress/ or emotional stress/ or family stress/ 1355678
or stress/ or Perceived Stress Scale/ or chronic unpredictable
stress/ or early life stress/ or Stress.mp.
6 Abuse.mp. or physical abuse/ or child sexual abuse/ or emotional 277067
abuse/ or child abuse survivor/ or sexual abuse/ or child abuse/ or
childhood sexual abuse survivor/
7 1 or 2 or 3 or 4 or 5 or 6 1935843
Dopaminergic 8 Dopamine.mp. or dopamine receptor/ or dopamine 2 receptor/ or 233428

Function dopamine brain level/ or dopamine release/ or dopamine 1
receptor/ or dopamine/ or dopamine transporter/
PET Studies 9 carbon 11/ or positron emission tomography/ or PET Studies mp. 145538
10 raclopride c 11/ or Raclopride.mp. or raclopride/ 4792

11 6 flurodopa f 18/ or fluorine 18/ or FDOPA.mp. 18578
12 9 or 10 or 11 156707
247
Total 13 7 and 8 and 12 1056

LIMIT limit 13 to (human and English language and yr=’2001 – Current’) 693
Appendix B-3: MEDLINE Search Terms
MEDLINE

Adversity 1 Trauma.mp. 275602
2 Adversity.mp. or Child Abuse/ or Stress, Psychological/ 150570
3 "Adult Survivors of Child Abuse"/ or Maltreatment.mp. 9826
4 PTSD.mp. or Stress Disorders, Post-Traumatic/ 42059
5 Stress.mp. or Stress, Psychological/ 960276
6 Spouse Abuse/ or Child Abuse/ or "Adult Survivors of Child 174714
Abuse"/ or Abuse.mp. or Child Abuse, Sexual/ or Physical
Abuse/
7 1 or 2 or 3 or 4 or 5 or 6 1366729
Dopaminergic Function 8 Receptors, Dopamine D1/ or Dopamine/ or Receptors, 162828
Dopamine D2/ or Dopamine.mp.
9 Dopamine Transporter.mp. 6653
10 8 or 9 162828
PET Studies 11 Positron-Emission Tomography/ or PET Studies.mp. 56063
12 Raclopride.mp. or Raclopride/ 2764
13 FDOPA.mp. 555
14 11 or 12 or 13 58429
Total 15 7 and 10 and 14 308
248
limit 15 to (human and English language and yr=’2001 – 161

LIMIT Current’)
Appendix B-4: Web of Science Search Terms
WEB OF SCIENCE

Adversity 1 (TS = (trauma*OR advers* OR maltreatment OR PTSD OR abuse 2022624
OR stress*)) AND LANGUAGE: (English) AND DOCUMENT TYPES:
(Articles) Timespan = 2001-2021
Dopaminergic Function 2 (TS = (dopamine OR D1 receptor* OR D2 receptor* OR 99923

dopamine transporter*)) AND LANGUAGE: (English) AND
DOCUMENT TYPES: (Articles) Timespan = 2001-2021
PET Studies 3 (TS = (PET stud* OR Raclopride OR FDOPA)) AND LANGUAGE: 61890
(English) AND DOCUMENT TYPES: (Articles) Timespan = 2001-
2021
Total 4 1 and 2 and 3 380
249
Appendix C: Supplementary Materials for Chapter 4

Delay Discounting
Because children exhibit choice “jitter” around indifference points, we developed an objective means of identifying
indifference points based on logistic regression and the fitting of a logistic sigmoid function to choice data. For each delay period, the
choice of delayed vs. immediate reward was defined categorically as 0 and 1, respectively. All choices at a particular delay, 𝑥 (𝑖) , were
then fit with the generalized logistic sigmoid function, ℎ𝜃 , modified by fitted coefficients 𝜃 (Equation 1).
Equation 1
1
ℎ𝜃 (𝑥 (𝑖) ) = 𝑇 𝑥 (𝑖)
1 + 𝑒 −𝜃
The values, 𝜃 𝑇 , were determined through the minimization of a regularized cost function that reflects the degree of fit between the
logistic function and the participant data (Equation 2).
Equation 2
𝑚 𝑛
1 1
𝐽(𝜃) = − ∑ [𝑦 (𝑖) log (ℎ𝜃 (𝑥 (𝑖) )) + (1 − 𝑦 (𝑖) ) log (1 − ℎ𝜃 (𝑥 (𝑖) ))] + 𝜆 ∑ 𝜃𝑗2
𝑚 2
𝑖=1 𝑗=1
250
Because the logistic function can be interpreted as indicating the probability of selecting an immediate reward of a given value,
an indifference point at a particular delay was defined as the point at which the probability of selecting the immediate reward was
equal to 0.5. Having determined indifference points in this way, each participant’s four indifference points were modeled by means of
a hyperbolic function (Equation 3; Reed et al. 2012), where V is the subjective value of the larger delayed reward amount (A) at a
given delay (D). The parameter k denotes the discounting rate observed (i.e., the steepness of the curve or how fast the subjective
value drops as a function of the delay).
Equation 3
𝐴
𝑉=
1 + (𝑘D)
Computational Modelling
To assess subcomponents of feedback-based learning, a reinforcement learning (RL) model was fit to each participant’s
behavioral data (Sutton & Barto, 1998). These models have been successfully applied to describe the behavior of teenagers and
children (van den Bos & McClure, 2013). The RL model uses the prediction error (δ) to update the decisions weights (w) associated
with each choice option (e.g., A, B, C, or D in experiment 1). The prediction error compares the actual outcome with the expected
outcome. Whenever feedback is better (worse) than expected, the model will generate a positive (negative) prediction error, which is
251
used to increase (decrease) the weight of the decision associated with the chosen option (e.g., option A). The impact of the prediction
error on the updating is scaled by the learning parameter alpha (0<α<1). Based on previous research, we assumed that gains and losses
are asymmetrically updated (Kahnt et al., 2009; van den Bos & McClure, 2013), thus we modeled the behavior with two independent
learning rates; (αpos) for positive feedback and (αneg) for negative feedback.
To model trial-by-trial choices, we used the soft-max choice rule to compute the probability (P) of choosing the most
rewarding target of a presented pair of choices (e.g. probability of choosing A of the AB pair experiment 1) on trial (t) as the
difference in the decision weights in each trial (wt) associated with each option, passed through a sigmoid function (Montague,
Hyman, & Cohen, 2004). For example, when stimulus pair AB is presented, the probability of choosing A is determined by:
Equation 4
1
𝑃(𝐴)𝑡 =
1+ 𝑒 −𝛽∙(𝑤 (𝐴)𝑡−𝑤 (𝐵)𝑡)
where the β parameter is a free parameter that indicates the sensitivity of the subject to the differences in decision-weights. The lower
the β parameter, the more random the choices appear. After each decision, the prediction error (δ) is calculated as the difference
between the outcome received (r = 1 for positive feedback and 0 for negative feedback) and the decision weight (wt) for the chosen
stimulus:
252
Equation 5
𝛿𝑡 = 𝑟𝑡 − 𝑤(chosen_stimulus)
Subsequently, the decision weights are updated according to:
Equation 6
𝑤𝑡+1 = 𝑤𝑡 + 𝜆 × 𝛼𝑜𝑢𝑡𝑐𝑜𝑚𝑒 × 𝛿𝑡
where 𝜆 is 1 for the chosen and 0 for the unchosen stimulus, αoutcome is a set of learning rates for positive (αpos) and negative feedback
(αneg), which scale the effect of the prediction error on the future decision weights and thus subsequent decisions. For example, a high
learning rate for positive feedback but a low learning rate for negative feedback indicates that positive feedback has a high impact on
future behavior, whereas negative feedback will hardly change future behavior. Importantly, at the beginning of the experiment the
decision weights (w) for each option was always set to the same value (0.5 for the PST, and 0 for the Instrumental Learning Task, in
both cases this is the mean of the expected outcomes), indicating that at the beginning of the experiment the participants had the same
expectation of rewards for each option. The learning rates and sensitivity parameter were individually estimated by fitting the model
predictions to participants’ actual decisions. Finally, note that for the Instrumental Learning Task, we fit gain and loss trials separately
given that the zero outcome in the gain and loss context has a different meaning. That is, in the gain context the zero outcome is seen
253
as a missed gain and in the loss context it means avoiding a loss. Thus our modeling is based on the (uncontroversial) assumption that
a missed gain is not the same thing as an actual loss, nor is avoiding a loss the same as gaining a point. As a result, we labeled the
learning rates as follows; αgain and αmiss in the gain context and αloss and αavoid in the loss context.
We used a robust combination of grid-search and maximum likelihood estimation using the Nelder–Mead simplex algorithm
implemented in the optim function in R to estimate the model parameters for each participant. Each point on the grid served as a
starting position for the minimization function, which is then used to find the parameters that maximize the log likelihood for an
individual. The grid point that produced the maximum over all starting positions was selected as starting point for finding the final
solution.
For quality assessment, the dual learning rate model was compared to several other models; random choice, single learning rate
and win-stay lose-shift (WSLS). The single learning rate is just as described above, but with a single learning rate for both type of
outcomes. The random choice model would choose each option with 50% probability (no learning). Finally, the WSLS also suggests
people use a simple choice rule without any learning.
For model selection purposes we computed the Bayesian information criterion (BIC), where lower BIC values indicate better
fit. These model comparisons indicate that, on the whole, the dual learning rate model best described the behavior on both experiment
1 (PST) and 2 (Instrumental Learning Task) (see Table S1). Finally, if subject level fits of the dual learning rate model were not
significantly different from the non-learning models these subjects were excluded from further analyses of parameter estimates;
254
parameters of the learning model could not be reliably estimated in the absence of learning. This led to the exclusions of 12 subjects in
experiment from the PST dataset, and 1 in the Instrumental Learning dataset.
Finally, results for the Kolmogorov-Smirnov test indicated that the distributions of all of the model parameters deviate
significantly from a normal distribution (all p’s < .001), so for subsequent analyses all parameters were first log transformed and then
scaled.
Tables
Table S1.
Model comparisons for the Probabilistic Selection Task and the Instrumental Learning Task.
single
alpha dual alpha WSLS
PST BIC 4378 4292* 18687
Instrumental
Learning
Task –gain BIC 1832 1824* 25035
Instrumental
Learning
Task –loss BIC 1854 1846* 26048
255
Curriculum Vitae
Niki Hosseini-Kamkar
EDUCATION
PhD. Cognitive, Developmental, and Brain Sciences, Western University, London, Present
Ontario
Thesis: The impact of early-life adversity and trauma on the mesolimbic dopamine
pathway
Thesis Advisory Committee: Dr. Daniel Ansari and Dr. Jessica Grahn
Supervisor: Dr. J. Bruce Morton
MSc. Developmental Psychology, Western University, London, Ontario 2013 -2015

Thesis: Neural and Behavioural Responses to Rewards and Losses in Early
Development: A Functional Magnetic Resonance Imaging Study
Thesis Advisory Committee: Dr. Riley Hinson, Dr. Martin Kavaliers, Dr. Raj
Rajakumar
Supervisor: Dr. J. Bruce Morton
BA Psychology, Western University, London, Ontario 2008 - 2012

Specialization: Developmental Cognitive Neuroscience
256
RESEARCH EXPERIENCE
Grants and Awards
NSERC Postgraduate Scholarship-Doctoral PGS-D, $63,000 2016-2019
Doctoral Excellence Research Award (DERA), $18,000 2016-2018
SONA Trainee Award, $250 May 2016

Received for the highest rated abstract at the PhD level.
Student Organizational Grant - Social Science Student Council, $1500 Dec 2015
Funding in support of organizing the “Inspiring Young Women in Science Conference”
Marilyn (Pack) McClelland Award in Psychology, $550 May 2015

Awarded to a full-time graduate student conducting research related to children.
Selection was based on academic achievement, research productivity and quality of
research publications.
Western Graduate Research Scholarship , $10,000/year 2013 - 2019
Western Open Access Fund , $2,270 August 2014

Funding provided for Frontiers open access publishing fee.
Fanshawe College Award of Academic Excellence, $750

Awarded to full-time students who were on the Dean’s Honour Roll Dec 2007
Publications
257
Hosseini-Kamkar, N., Lowe, C., & Morton, J. B. (2021). The differential calibration of the HPA axis as a function of trauma versus adversity: A
systematic review and p-curve meta-analyses. Neuroscience and Biobehavioral Reviews (accepted). Preprint: https://psyarxiv.com/qnyr8/
Kamkar, N., Hosseini-Kamkar, N., & Ansari D. (2018). What counts as success? And how do we get there?. Frontiers for young minds, 6(12),
492-511.
Hosseini-Kamkar, N., & Morton J.B. (2017). CanDiD: A Framework for Linking Executive Function and Education. Frontiers in Education.
Hosseini-Kamkar, N., Lewis, D.J., van den Bos, W., & Morton J.B. (2017). Ventral striatal activity links adversity and reward processing in
children. Developmental Cognitive Neuroscience.
Hosseini-Kamkar, N. & Morton, J.B. (2014). Sex differences in self-regulation: An evolutionary perspective. Frontiers in Neuroscience.
Nicholson, A.A., Hosseini-Kamkar, N. , Fallowfield, H., & Morton J.B.(2015). The neural correlates of cognitive behavioural self-regulation in
early development. Western Undergraduate Psychology Journal, 3(1).
Hosseini-Kamkar, N. (2014).The A1 allele of the Taq1 A polymorphism in association with addiction: A review. Western Undergraduate
Psychology Journal, 1(1), Article 10
Presentations
Hosseini- Kamkar, N., and Morton, J.B. (Jan 2021). The Long-Term Embedding of Traumatic Life Experiences on the HPA-axis: A p-curve Meta-
Analysis. Poster Presentation at SfN Global Connectome, virtual event.
H. Kamkar, N., Lewis, D.J., van den Bos W., Morton, J.B. (May 2017). The Association between Adversity and Reward Processing is Mediated by
Ventral Striatal Sensitivity to Rewards. Oral Presentation at The Child Health Symposium, London, Ontario.
258
H. Kamkar, N. (December 2016). Integrating Immigrants into Canadian Society: Psychological and Cultural Considerations. Oral Presentation
at Pathways to Prosperity, Ottawa, Ontario.
Hosseini-Kamkar, N., & Morton, J.B. (September 2016). The Mesolimbic Dopamine Pathway Is Sensitive To Early Life Adversity. Poster
presented at The FLUX Congress, St. Louis, Missouri.
Hosseini-Kamkar, N., & Morton, J.B. (May 2016). The Mesolimbic Dopamine Pathway Is Sensitive To Early Life Adversity. Oral Presentation at
Southern Ontario Neuroscience Association Annual Conference, Waterloo, Ontario.
Hosseini-Kamkar, N., & Morton, J.B. (May 2016). The Mesocorticolimbic Dopamine Pathway Exhibits A Phenotypic Plasticity To The Experience
Of Early Life Adversity. Poster presented at 10th Annual Canadian Association for Neuroscience Meeting, Toronto, Ontario.
Hosseini-Kamkar, N., & Morton, J.B. (February 2016). The Neural Correlates Of Cognitive And Behavioral Self-Regulation In Preadolescents: A
Multidisciplinary Approach. Poster presented at The 44th Annual Meeting of The International Neuropsychological Society, Boston,
Massachusetts.
Hosseini-Kamkar, N., El-Baba, M., & Morton, J.B. (February 2016). The Mesocorticolimbic Dopamine Pathway Exhibits A Phenotypic Plasticity
To The Experience Of Early Life Adversity. Poster presented at Lake Ontario Visionary Establishment 45th Annual Conference, Niagara Falls,
Ontario.
Hosseini-Kamkar, N. & Kamkar, N. (November 2015). Memory, Self-Control Prosopagnosia and Savants: An Overview of What Neuroscientists
and Psychologists Can Study. Oral Presentation at Lord Dorchester Secondary School, London, Ontario.
Hosseini-Kamkar, N. (October 2015). What is Graduate School? An Overview of Graduate Studies in Psychology and Neuroscience. Oral
Presentation at Mother Teresa Catholic Secondary School, London, Ontario.
Hosseini-Kamkar, N. (September 2015). Why Neuroscience is Interesting: Neuroscience Case Studies. Oral Presentation at Saint André Bassette
Catholic Secondary School, London, Ontario.
259
Hosseini-Kamkar, N. (May 2015). Closing the Gap Between Neuroscience and Education: Proposal for the Parents Reaching Out Regional Grant.
Oral Presentation at TVPIC, London, Ontario.
Hosseini-Kamkar, N. (November 2014). Why Neuroscience is Interesting: Neuroscience Case Studies. Oral Presentation at Holy Cross Catholic
Secondary School, Strathroy, Ontario.
Fallowfield, H., Hosseini-Kamkar, N., & Morton, J.B. (2014). Individual Differences in Cognitive and Behavioural Self-Regulation: A Multi-Level
Approach for Assessing Genetic and Environmental Influences Early in Development. Poster presented at Minds on Minds Symposium,
London, Ontario.
TEACHING EXPERIENCE
CAMH – Temerty Centre for Therapeutic Brain Intervention, Toronto Ontario June 2018 –
Project Coordinator: Designed and created lecture material and practical manuals Dec 2019
for physicians, nurses, and technicians. Coordinated between CAMH Education and
CAMH Temerty Centre to ensure all aspects of the course from registration to delivery
were conducted in a professional and effective manner. Obtained course
accreditation and certification from the University of Toronto, The Royal College of
Physicians and Surgeons of Canada, The Canadian Psychological Association, The
American Medical Association, and The Canadian Association for Medical Radiation
Technologists. Worked with diverse bodies and organizations to conduct and manage
a large and continuous project for educating family physicians, psychiatrists, nurses,
and technicians in the use of repetitive transcranial magnetic stimulation (rTMS) in
treating psychiatric illnesses.
260
Research in Developmental Cognitive Neuroscience, Western University Sept-Dec 2017

Sessional Lecturer: Designed and created weekly lectures and laboratory materials
focusing on the analysis of functional magnetic resonance imaging (fMRI) data using
SPM. Discussed ethical and technical issues in conducting research with
developmental populations. Evaluated course material including exams,
presentations, and final papers.
PSYC 1027 - Human Relations , Fanshawe College Sep 2016 - Apr

Part-time Lecturer: Created all course material and delivered weekly lectures to 2017
approximately 25 students. Evaluated students on all examinations and coursework.
Jan-Apr 2016
PSYC 1011 - Interpersonal Psychology , Fanshawe College
Part-time Lecturer: Created all course material and delivered weekly lectures to
Jan-Apr 2016
PSYC 1067 - A Culture of Addictions , Fanshawe College
PSYC 5011 - Industrial Psychology, Fanshawe College Jan-Apr 2016

PSYC 3490 -Special Topics: The Development of the Social Brain, Western Jan-Apr 2015
University
261
Teaching Assistant and Guest Lecturer: Assisted the Professor in grading one-half of
all assignments and coursework. Served as a Guest Lecturer to teach about domain-
general versus domain-specialized accounts of Theory of Mind.
PSYC 1042B - Exceptional Children: Behavioural Disorders , Western University Jan-Apr 2014
Teaching Assistant: Assisted the Professor in the management of students. Graded
and evaluated students on all examinations and coursework. Delivered all relevant
course material and grades via Sakai.
SPM8 Lecture and Tutorial, Western University 2013 - 2014

Instructor: Designed and presented numerous tutorials to train laboratory members
and colleagues on the use of SPM8 for the preprocessing and analysis of fMRI data.
Supervised Andrew Nicholson (an Honours student) in the design of an SPM8 manual
for the Developmental Neuroimaging Laboratory.

Early Experience and The Functional Calibration of The Stress-Response Systems

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Early Experience and The Functional Calibration of The Stress-Response Systems

Uploaded by

Copyright:

Available Formats

Western University

Electronic Thesis and Dissertation Repository

Early experience and the functional calibration of the stress-

Supervisor: Morton, J Bruce, The University of Western Ontario

Follow this and additional works at: https://ir.lib.uwo.ca/etd

Part of the Cognitive Neuroscience Commons

Individuals exposed to adversities in childhood are at a greater risk of developing

In Chapter 3, a systematic review of the human Positron Emission Tomography (PET)

Do adverse life-experiences leave a lasting imprint on the brain? Likewise, does

Table of Contents .............................................................................................................. vii

List of Tables (where applicable) ..................................................................................... xii

List of Figures (where applicable) ................................................................................... xiii

List of Appendices (where applicable) ........................................................................... xvii

1 General Introduction ...................................................................................................... 1

1.1 The long-term impact of early-life adversity .......................................................... 1

1.2 Adaptive phenotypic plasticity ............................................................................... 2

1.2.1 Individual differences in phenotypic plasticity ........................................... 3

1.3 The role of the central nervous system ................................................................... 4

1.3.1 The plasticity of the HPA axis .................................................................... 5

1.3.2 The plasticity of the dopamine system........................................................ 6

1.4 Distinguishing adversity and trauma ...................................................................... 8

1.5 Aim of current thesis and summary of findings...................................................... 9

1.5.1 Chapter 2 – The differential calibration of the HPA axis as a function of

1.5.2 Chapter 3 – Stress, glucocorticoids, and the mesolimbic dopamine

1.5.3 Chapter 4 – Ventral striatal reactivity links adversity and reward

1.6 Adaptive plasticity: Stress-response Adaptations to Variability in Early-life

1.7 References ............................................................................................................. 15

2.1 Introduction ........................................................................................................... 22

2.1.1 The stress response system ....................................................................... 23

2.2.2 The General Adaptation Syndrome (G-A-S) Model ................................. 27

2.2.3 Adaptationist perspectives ........................................................................ 28

2.4 Sources of variability and confounding factors .................................................... 31

2.4.1 Cortisol measurements .............................................................................. 31

2.4.2 Trauma versus adversity ........................................................................... 33

2.5 Current studies: p-curve meta-analyses ................................................................ 35

2.6.1 Study 1: Data sources and study selection ................................................ 37

2.6.2 Study 2: Data sources and study selection ................................................ 43

2.6.3 P-curve data abstraction ............................................................................ 48

2.7 Results ................................................................................................................... 48

2.7.1 Study 1.1: Traumatic experiences and cortisol hyper-reactivity p-curve . 48

2.7.2 Study 1.2: Traumatic experiences and cortisol hypo-reactivity p-curve .. 52

2.8 Discussion ............................................................................................................. 59

2.9 References ........................................................................................................... 121

Chapter 3 ......................................................................................................................... 141

3 Stress, glucocorticoids, and the mesolimbic dopamine pathway: How early-life

3.1 Introduction ......................................................................................................... 141

3.1.1 The mesolimbic dopamine pathway: A brief review .............................. 142

3.1.2 Stress and drugs: Animal studies ............................................................ 143

3.1.3 Stress and drugs: Human studies ............................................................ 145

3.1.5 Mechanisms of action: The role of glucocorticoids ................................ 147

3.1.6 Stress and dopamine: Animal literature .................................................. 149

3.1.8 Theoretical frameworks and models ....................................................... 152

3.1.9 Stress-response Adaptations to Variability in Early-life experiences

3.1.10 Current study and hypotheses ................................................................. 154

3.2 Methods............................................................................................................... 155

3.2.1 Data sources and study selection ............................................................ 155

3.2.2 Screening and inclusion .......................................................................... 155

3.3 Results ................................................................................................................. 158

3.4 Discussion ........................................................................................................... 161

3.1 Conclusion .......................................................................................................... 164

3.2 References ........................................................................................................... 166

Chapter 4 ......................................................................................................................... 182

4.2 Methods............................................................................................................... 185