Available online at www.sciencedirect.
com
ScienceDirect
The neuroscience of empathy
Abigail A Marsh
In the fifty years following the first mention of empathy in the
neuroscience literature, significant gains in understanding the
neural basis of empathy have been made. Converging strands
of evidence support simulation-based models of empathy,
such that, for example, empathy for pain recruits networks
involved in the first-hand experience of pain. Similarly, empathy
for other distinct sensory and affective states (e.g. tactile
pleasure, fear) leverages the networks involved in the firsthand
experience of those states. Such empathic simulations are not
unique to humans but can be observed across social species.
Both emotional empathy and mentalizing (cognitive empathy)
may promote empathic concern or compassion, an outcome
variable of particular interest to researchers and practitioners.
Although individuals vary in their baseline empathic capacities
and proclivities, empathy and concern can be modulated by
interpersonal and contextual variables and with training.
Address
Department of Psychology, Georgetown University, United States
Corresponding author: .Marsh, Abigail A (aam72@georgetown.edu)
Current Opinion in Behavioral Sciences 2018, 19:110–115
This review comes from a themed issue on Emotion–memory
interactions
Edited by Mara Mather and Michael Fanselow
For a complete overview see the Issue and the Editorial
Available online 10th January 2018
https://doi.org/10.1016/j.cobeha.2017.12.016
2352-1546/ã 2017 Elsevier Ltd. All rights reserved.
The neuroscience of empathy
The word empathy first appeared in the neuroscience
literature fifty years ago in a 1967 article by Paul MacLean
[1], who defined it as ‘the capacity to identify one’s own
feelings and needs with those of another person.’ He
viewed empathy as the basis of caring for and desiring to
help others, and therefore a topic of critical importance for
solving pressing problems of the modern era, including
interpersonal callousness and aggression. He concluded
with a plea that physicians and scientists of the future not
neglect the study of empathy and the brain. One can only
imagine the pleasure that the profusion of neuroscience
research on empathy in recent decades would have
brought him.
Current Opinion in Behavioral Sciences 2018, 19:110–115
It might have brought him surprises as well. MacLean
described empathy as a holistic and cognitively complex
phenomenon that was primarily the endowment of
humans, and he hypothesized that it was heavily reliant
on the prefrontal cortex. Recent empirical evidence sug-
gests otherwise. Empathy instead appears to reflect mul-
tiple dissociable processes, many of which rely on ancient,
subcortical structures that function similarly to promote
empathy across a variety of social species.
Evidence for simulation-based accounts of
empathy
MacLean’s definition of empathy anticipated empirical
findings that empathically representing others’ sensory
and emotional states may require leveraging the networks
involved in experiencing those states firsthand. Perhaps
the clearest such example is observed in the literature on
empathy for pain. Early evidence for empathic simulation
emerged nearly two decades ago, when co-activation in a
single neuron in dorsal anterior cingulate cortex was
observed both during experienced and observed pain
[2]. Since that time, a large body of functional neuroim-
aging research has demonstrated that experiencing pain
firsthand and observing or inferring others’ pain are both
associated with increased activation in this and other
cortical and subcortical structures collectively described
as the ‘pain matrix’ [3], a network of second-order pain
processing regions that include somatosensory cortex,
posterior insula, and periaqueductal gray as well as inte-
grative regions involved in affective and motivational
aspect of pain, such as the mid-anterior cingulate cortex
(which includes the cingulate and paracingulate gyri) and
the anterior insula [4,5]. Direct sensory exposure to
others’ pain is not necessary for empathic activity in this
network; reading about others’ pain, for example, results
in comparable patterns of activity [6]. That overlapping
patterns of activity are recruited during experienced and
observed or imagined pain echoes MacLean’s conjectures
about the neural basis of empathy.
However, until recently, a major limitation of this inter-
pretation has been the correlational nature of most inves-
tigations of empathic neural responses. Large integrative
regions such as the mid-anterior cingulate cortex and
anterior insula subserve a wide array of processes not
directly related to pain, and each measured voxel within
them incorporates activation in thousands of neurons, so
fMRI-based observations of shared activation patterns
cannot conclusively demonstrate true empathic simula-
tion. But recent experimental approaches to understand-
ing empathy for pain lend further support to simulationist
models. Multi-voxel pattern analysis confirms common
www.sciencedirect.com

neural coding of experienced and empathic pain in anterior
insula and mid-anterior cingulate cortex [4]. And placebo
analgesia reduces — to comparable degrees — reports of
pain during electrical stimulation and descriptions of
others’ pain during identical stimulation (whereas
responses during non-pain control conditions are unaf-
fected) [7,8,9]. These subjective changes are accompa-
nied by nearly identical reductions in activity in mid-
anterior cingulate cortex and bilateral anterior insula dur-
ing experienced and empathic pain. More, placebo effects
for both experienced and empathic pain are blocked by the
opioid antagonist naltrexone, indicating that both pro-
cesses are subserved by comparable neurochemical pro-
cesses [7]. Empathy for pain can also be reduced by
administration of non-opioid analgesics, including acet-
aminophen and oxytocin [10,11], or increased using real-
time fMRI-based neurofeedback. Using this approach to
increase activity in anterior insula and functional coupling
between this region and frontal cortex results in increased
reported empathy for pain in the absence of changes in
general arousal [12]. Collectively, these findings provide
strong support for the causal role of empathic simulation in
representing others’ internal states.
In terms of the specific functions served by regions within
the pain matrix, the anterior insula appears to play an
interpretive role, as lesions to this region interfere with
the ability to simply perceive and recognize others’
pain [13]. But debates about the specific role of the
Figure 1
dmACC
mPFC
precuneus
VS
sgACC
amygdala PAG
Emotional empathy (negative emotions)
Emotional empathy (positive emotions)
Overview of key regions associated with four forms of empathy. Emotional empathy for negative emotion is associated with activity in dorsal mid-
anterior cingulate cortex (dmACC) and amygdala; emotional empathy for positive emotion is associate with activity in medial prefrontal cortex
(mPFC); cognitive empathy (mentalizing) is associated with activity in mPFC, temporo-parietal junction (TPJ) and precuneus; and empathic
concern is associated with activity in amygdala, ventral striatum (VS), subgenual anterior cingulate cortex (sgACC), and periaqueductal gray (PAG).
www.sciencedirect.com
The neuroscience of empathy .Marsh 111
mid-anterior cingulate cortex in pain are ongoing.
Although it has been suggested that activity in this region
encodes pain specifically [14], other theoretical models
argue instead that it is primarily involved in allocating
control [15] or in threat appraisal [16], such that neither
subjective nor empathic pain can be accurately described
as localized to this region [17]. But the mid-anterior
cingulate cortex is clearly causally implicated in affective
and motivational aspects of first-hand pain. Ablations or
deep-brain stimulation of this region provide significant
pain relief, which appears to reflect reductions in the
motivational significance of pain [18–20]. It may therefore
be the case that these procedures also decrease the
motivational significance of others’ pain.
Empathy as multiple distinct processes
Empathic pain is the most robustly supported form of
empathy in the neuroscience literature. But it should not
be viewed as synonymous with the construct of empathy,
which represents a collection of dissociable processes [21]
(Figure 1). Indeed, simulation-based accounts of empa-
thy require that empathic pain be subserved by different
processes than empathy for other sensory and affective
states, as empathy for any given state would rely on the
recruitment of systems that support the first-hand expe-
rience of that state.
Thus, for example, although empathy for physical pain
recruits some overlapping neurocircuitry as empathy for
TPJ
AI
Cognitive empathy
Empathic concern
Current Opinion in Behavioral Sciences
Current Opinion in Behavioral Sciences 2018, 19:110–115
112 Emotion–memory interactions
social pain (for example, rejection by peers), these forms
of empathy can be dissociated, with one study finding that
empathic social pain recruits greater activation in a wide
array of visual, sensory processing, and emotional regions
of the cortex than empathic physical pain [22]. These
differences could be interpreted as reflecting in part the
differing cognitive demands involved in processing social
rejection versus electrical stimulation-induced pain. But
empathic responses to distinct states are dissociable even
when processing demands are similar [23]. For example,
whereas empathic responses to painful touch recruit
fronto-insular cortex, responses to pleasant touch recruit
medial orbitofrontal cortex, which is involved in integrat-
ing multimodal information about reward [24]. Studies of
responses to facial affect also consistently demonstrate
that distinct networks respond to faces that differ only in
their emotional expressions. The amygdala is most
responsive to expressions of fear and the anterior insula
to disgust [25], in keeping with findings from electrical
stimulation studies that implicate regions within these
structures in the experiences of fear and disgust, respec-
tively [26–28]. And whereas written narratives about
physical pain elicit activity in the canonical pain matrix,
narratives about emotional suffering (for example, a target
who learns that a loved one is seriously ill) yields activity
in cortical regions involved in representing others’ cogni-
tive states (or mentalizing), including bilateral tempor-
oparietal junction, precuneus, and medial prefrontal cor-
tex, as well as amygdala [6].
The critical distinction between emotional empathy and
mentalizing (also called cognitive empathy or Theory of
Mind) has been clearly demonstrated using a novel para-
digm called the EmpaTOM [12,29]. During this task,
participants view videos in which a target individual
describes either neutral or emotionally negative autobio-
graphical events (such as being struck by an intimate
partner) with accompanying nonverbal displays. After-
ward, participants report their own emotional response to
the video and answer questions about the target’s beliefs.
In this paradigm, viewing emotionally negative videos
yields increased activation in anterior insula and mid-
cingulate gyrus, whereas answering questions requiring
mental state inferences recruits activity in the mentaliz-
ing network, including temporoparietal junction, precu-
neus, and medial prefrontal cortex. These regions were
also recruited while empathizing, however, the peaks of
activity were dissociable across the two empathy tasks.
The authors speculate that distinct target locales within
these regions may support differentiation of own versus
others’ affective versus cognitive states. Arguing against
the possibility that these networks represent subcompo-
nents of a single overarching social cognition network,
empathizing and mentalizing capacities appear to be
independent and predict distinct behavioral outcomes
[30]; empathizing may even inhibit mentalizing in
response to strong negative emotions [30,31].
Current Opinion in Behavioral Sciences 2018, 19:110–115
Empathy promotes prosocial motivations and
behavior
A limitation of the EmpaTOM is that its negative affect
videos convey various specific states, including pain, fear,
and sadness, which prevents identification of unique
networks that may support empathy for these distinct
states. However, the task is designed to distinguish
between the two forms of empathy that are, historically,
most often conflated — including by MacLean — which
are emotional empathy and empathic concern.
Empathic concern (also called care or compassion) is the
prosocial motivational state that promotes caring and
altruistic helping [32]. It is synonymous neither with
emotional empathy nor cognitive empathy, although
either process may induce it [29,33]. For example,
individuals who engage in extraordinary acts of altruism
show heightened emotional empathy for others’ fear [34].
And trial-by-trial empathic distress in an empathic pain
paradigm predicts increases in later altruistic responding
[35]. Although recruitment of either emotional or cogni-
tive empathy networks can yield altruistic motivation,
people differ in their tendency to recruit these networks
[29,36].
Concerned, prosocial motivation itself is supported by a
network that incorporates regions underlying goal-
directed approach, such as the ventral tegmental area,
subgenual anterior cingulate cortex, and striatum
[35,37,38]. A structural imaging study of neurodegener-
ative disorders linked declines in concern specifically to
reductions of gray matter volume in these structures,
including caudate, nucleus accumbens, and orbitofrontal
cortex, whereas declines in emotional empathy were
linked to reduced gray matter in regions that included
temporal cortex and amygdala [39]. In addition, both of
forms of empathy were linked to alterations in anterior
insula. These findings support the causal involvement of
the goal-directed approach network in concern, and con-
firm that it is dissociable from networks that support
empathy.
One as-yet unaddressed puzzle is how is the negative
social inputs encoded during emotional empathy for
others’ suffering are transformed into the positive proso-
cial motivation that drives care and altruism in response.
This transformation cannot be solely reliant on higher-
level cognition, since clear evidence of emotional empa-
thy and prosocial motivation can be observed in non-
human animals, including rats and mice [40,41]. More
likely, caring motivation relies on ancient subcortical
mechanisms that support parental care, including the
amygdala and stria terminalis [42]. The amygdala in
particular may be essential for triggering the motivation
to care in response to distress through its interactions with
the striatum and subgenual anterior cingulate cortex, as
well as the periaqueductal gray and hypothalamus
www.sciencedirect.com

[35,43–45]. Notably, the amygdala is the region in which
structure and functioning are most consistently linked to
stable individual differences in care and concern
[6,32,46].
Key variables moderating empathy
These considerations are useful for addressing critical
questions posed by MacLean as well as researchers today:
Why are some people more concerned about others’
welfare than others, and more concerned about some
people than others? And how can empathy be modulated?
The evidence for stable individual differences in empa-
thy is clear [47]. Sometimes structural bases for these
deficits can be identified, as in the case of neurodegener-
ative disorders or lesions that disrupt empathy [39,48,49].
In healthy adults, differences in the structure of the insula
and somatosensory cortex may also correspond to varia-
tion in facets of empathy [50–52]. In extreme cases,
malformation or dysfunction in the systems that support
empathy may lead to disorders that are associated with
severe, pervasive deficits in the capacity for various forms
of empathy [21]. For example, alexithymia is character-
ized by deficits in emotional empathy and is linked to
reduced amygdala volume [53], whereas autism is char-
acterized primarily by reduced cognitive empathy and
is associated with reduced medial prefrontal activation
during empathy inductions [54]. And psychopathy,
the hallmarks of which are reduced emotional empathy
and empathic concern (callousness), is consistently
linked to structural malformations and hypoactivation
in the amydgala [46,55] as well as aberrant empathic
pain responding in anterior insula and mid-cingulate
cortex [56].
Distinguishing empathic capacities from empathic pro-
clivities is essential [56–59]. When underlying networks
are significantly degraded, facets of empathy may become
impossible. But for the majority of the population, varia-
tion in empathy may reflect proclivities as much as
capacities. Empathy can, for example, be enhanced by
social and interpersonal variables like similarity or shared
group membership of the target and observer [60,61,62],
or by the target’s high social status or moral status [63–65].
Situational variables can shape empathy as well; empathy
may increase under acute stress [31], or following behav-
ioral synchrony [66]. The temporoparietal junction is
thought to be a key region responsible for modulating
emotional and cognitive empathy in response to situa-
tional variables. Activity in this region corresponds to
variation in a number of empathy-linked outcomes
[6,29], and electromagnetically disrupting the area
results in reduced perceived intensity of others’ pain,
dampened neural responses to pain [67], and reduced
generosity for socially distant others [68]. Broader cultural
factors are also relevant. Empathic biases for ingroup
members may be reduced by individualistic cultural
www.sciencedirect.com
The neuroscience of empathy .Marsh 113
values, which, relative to collectivist values, de-empha-
size strict boundaries between groups [69].
Methods for actively increasing empathy for outgroup
members include increasing positive social contact
between groups [70,71] and engaging in practices that
encourage consideration of distant others’ well-being,
including compassion or loving-kindness meditation
and reading emotionally engrossing literature [72,73].
Compassion-based meditation has been shown to alter
the function [72] and structure of key empathic brain
regions — for example, increasing cortical thickness in
anterior insula [74] — suggesting a mechanism by which
training may yield changes in empathic outcomes. In a
sense, empathy may be skill like any other: one that
improves with practice.
This, too, was intuited by MacLean, who surmised that
perhaps the key to enhancing empathy and concern is
being ‘exposed to human suffering . . . [and] the respon-
sibility of ministering to it.’ (p. 380).
Conflict of interest statement
Nothing declared.
Acknowledgements
Thank you to Miriam Crinion for her assistance in reviewing the relevant
literature. This research was supported in part by John Templeton
Foundation Grant 47861 to the author.
References and recommended reading
Papers of particular interest, published within the period of review,
have been highlighted as:
 of special interest
 of outstanding interest
1. MacLean PD: The brain in relation to empathy and medical
education. J Nerv Ment Dis 1967, 144:374-382.
2. Hutchison WD et al.: Pain-related neurons in the human
cingulate cortex. Nat Neurosci 1999, 2:403-405.
3. Lamm C, Decety J, Singer T: Meta-analytic evidence for
common and distinct neural networks associated with directly
experienced pain and empathy for pain. NeuroImage 2011,
54:2492-2502.
4. Corradi-Dell’Acqua C et al.: Cross-modal representations of
first-hand and vicarious pain, disgust and fairness in insular
and cingulate cortex. Nat Commun 2016, 7:10904.
5. Zaki J et al.: The anatomy of suffering: understanding the
relationship between nociceptive and empathic pain. Trends
Cogn Sci 2016, 20:249-259.
6. Bruneau EG, Jacoby N, Saxe R: Empathic control through
coordinated interaction of amygdala, theory of mind and
extended pain matrix brain regions. NeuroImage 2015, 114:
105-119.
7. Rütgen M et al.: Placebo analgesia and its opioidergic
 regulation suggest that empathy for pain is grounded in self
pain. Proc Natl Acad Sci U S A 2015, 112:E5638-E5646.
This study demonstrates that placebo analgesia reduces not only the
experience of pain, but empathy for pain, effects that accompanied by
common patterns of activity in the neural pain matrix, and reversed by the
opioid antagonist naltrexone. These findings strongly support similar
causal mechanisms underlying experienced and empathic pain.
Current Opinion in Behavioral Sciences 2018, 19:110–115
114 Emotion–memory interactions
8. Rütgen M et al.: Reduction of empathy for pain by placebo
analgesia suggests functional equivalence of empathy and
first-hand emotion experience. J Neurosci 2015, 35:8938-8947.
9. Rütgen M et al.: Psychopharmacological modulation of event-
related potentials suggests that first-hand pain and empathy
for pain rely on similar opioidergic processes.
Neuropsychologia 2017, S0028-3932:30152-30155.
10. Mischkowski D, Crocker J, Way BM: From painkiller to empathy
killer: acetaminophen (paracetamol) reduces empathy for
pain. Soc Cogn Affect Neurosci 2016, 11:1345-1353.
11. Bos PA et al.: Oxytocin reduces neural activity in the pain
circuitry when seeing pain in others. NeuroImage 2015,
113:217-224.
12. Yao S et al.: Voluntary control of anterior insula and its
functional connections is feedback-independent and
increases pain empathy. NeuroImage 2016, 130:230-240.
13. Gu X et al.: Anterior insular cortex is necessary for empathetic
pain perception. Brain 2012, 135:2726-2735.
14. Lieberman MD, Eisenberger NI: The dorsal anterior cingulate
cortex is selective for pain: results from large-scale reverse
inference. Proc Natl Acad Sci U S A 2015, 112:15250-15255.
15. Shenhav A, Cohen JD, Botvinick MM: Dorsal anterior cingulate
cortex and the value of control. Nat Neurosci 2016, 19:
1286-1291.
16. Kalisch R, Gerlicher AM: Making a mountain out of a molehill: on
the role of the rostral dorsal anterior cingulate and
dorsomedial prefrontal cortex in conscious threat appraisal,
catastrophizing, and worrying. Neurosci Biobehav Rev 2014,
42:1-8.
17. Wager TD et al.: Pain in the ACC. Proc Natl Acad Sci U S A 2016,
113:E2474-E2475.
18. Agarwal N et al.: Anterior cingulotomy for intractable pain.
Interdiscip Neurosurg 2016, 6:80-83.
19. Sharim J, Pouratian N: Anterior cingulotomy for the treatment of
chronic intractable pain: a systematic review. Pain Physician
2016, 19:537-550.
20. Russo JF, Sheth SA: Deep brain stimulation of the dorsal
anterior cingulate cortex for the treatment of chronic
neuropathic pain. Neurosurg Focus 2015, 38:E11.
21. Bird G, Viding E: The self to other model of empathy: providing a
new framework for understanding empathy impairments in
psychopathy, autism, and alexithymia. Neurosci Biobehav Rev
2014, 47:520-532.
22. Novembre G, Zanon M, Silani G: Empathy for social exclusion
involves the sensory-discriminative component of pain: a
within-subject fMRI study. Soc Cogn Affect Neurosci 2015,
10:153-164.
23. Lamm C, Silani G, Singer T: Distinct neural networks underlying
 empathy for pleasant and unpleasant touch. Cortex 2015,
70:79-89.
The results of this study demonstrate that empathy for distinct sensory
experiences (pleasurable versus painful touch) is supported by distinct
neural networks, supporting the idea that empathy is not a single phe-
nomenon, but multiple related phenomena.
24. Levy DJ, Glimcher PW: The root of all value: a neural common
currency for choice. Curr Opin Neurobiol 2012, 22:1027-1038.
25. Fusar-Poli P et al.: Functional atlas of emotional faces
processing: a voxel-based meta-analysis of 105 functional
magnetic resonance imaging studies. J Psychiatry Neurosci
2009, 34:418-432.
26. Caruana F et al.: Emotional and social behaviors elicited by
electrical stimulation of the insula in the macaque monkey.
Curr Biol 2011, 21:195-199.
27. Davis M: The role of the amygdala in fear and anxiety. Annu Rev
Neurosci 1992, 15:353-375.
28. Penfield W, Faulk MEJ: The insula; further observations on its
function. Brain 1955, 78:445-470.
Current Opinion in Behavioral Sciences 2018, 19:110–115
29. Tusche A et al.: Decoding the charitable brain: empathy,
 perspective taking, and attention shifts differentially predict
altruistic giving. J Neurosci 2016, 36:4719-4732.
This study demonstrates that distinct routes can drive variation in chari-
table giving: anterior insula-linked empathy and temporoparietal junction-
linked mentalizing, reinforcing both the distinctions between these phe-
nomena and their relevance to prosocial motivation.
30. Kanske P et al.: Are strong empathizers better mentalizers?
Evidence for independence and interaction between the
routes of social cognition. Soc Cogn Affect Neurosci 2016,
11:1383-1392.
31. Tomova L et al.: Increased neural responses to empathy for
pain might explain how acute stress increases prosociality.
Soc Cogn Affect Neurosci 2017, 12:401-408.
32. Marsh AA: Neural, cognitive, and evolutionary foundations of
human altruism. Wiley Interdiscip Rev Cogn Sci 2016, 7:59-71.
33. Kanske P et al.: Dissecting the social brain: introducing the
EmpaToM to reveal distinct neural networks and brain–
behavior relations for empathy and Theory of Mind.
NeuroImage 2015, 122:6-19.
34. Marsh AA et al.: Neural and cognitive characteristics of
extraordinary altruists. Proc Natl Acad Sci U S A 2014,
111:15036-15041.
35. FeldmanHall O et al.: Empathic concern drives costly altruism.
 NeuroImage 2015, 105:347-356.
Self-reported empathic concern was associated with costly altruism and
with increased activity in reward-expectancy regions; in addition, trial-by-
trial distress during others’ pain during the task independently predicted
costly altruism, linking empathy and altruistic motivation.
36. Hein G et al.: The brain’s functional network architecture
reveals human motives. Science 2016, 351:1074-1078.
37. Genevsky A, Knutson B: Neural affective mechanisms predict
market-level microlending. Psychol Sci 2015, 26:1411-1422.
38. Singer T, Klimecki OM: Empathy and compassion. Curr Biol
2014, 24:R875-R878.
39. Shdo SM et al.: Deconstructing empathy: neuroanatomical
 dissociations between affect sharing and prosocial motivation
using a patient lesion model. Neuropsychologia 2017.
The results of this structural neuroimaging study comparing healthy
controls to adults with several neurodegenerative disorders help to
disambiguate structures that support emotional empathy (including affect
recognition and sharing) from empathic concern.
40. Burkett JP et al.: Oxytocin-dependent consolation behavior in
rodents. Science 2016, 351:375-378.
41. Martin LJ et al.: Reducing social stress elicits emotional
contagion of pain in mouse and human strangers. Curr Biol
2015, 25:326-332.
42. Preston SD: The origins of altruism in offspring care. Psychol
Bull 2013, 139:1305-1341.
43. Lockwood PL et al.: Neurocomputational mechanisms of
prosocial learning and links to empathy. Proc Natl Acad Sci U S
A 2016, 113:9763-9768.
44. Decety J: The neural pathways, development and functions of
empathy. Curr Opin Behav Sci 2015, 3:1-6.
45. Brethel-Haurwitz et al.: Amygdala-midbrain connectivity
indicates a role for the mammalian parental care system in
human altruism. Proc Biol Sci 2017, 284:1865.
46. Blair RJ, Leibenluft E, Pine DS: Conduct disorder and callous-
unemotional traits in youth. N Engl J Med 2014, 371:2207-2216.
47. Williams JH et al.: Perceiving and expressing feelings through
actions in relation to individual differences in empathic traits:
the Action and Feelings Questionnaire (AFQ). Cogn Affect
Behav Neurosci 2016, 16:248-260.
48. Oishi K et al.: Critical role of the right uncinate fasciculus in
emotional empathy. Ann Neurol 2014, 77:68-74.
49. Johnstone B et al.: Functional and structural indices of
empathy: evidence for self-orientation as a
www.sciencedirect.com

neuropsychological foundation of empathy. Neuropsychology
2015, 29:463-472.
50. Allen M et al.: Insula and somatosensory cortical myelination
and iron markers underlie individual differences in empathy.
Sci Rep 2017, 7:43316.
51. Eres R et al.: Individual differences in local gray matter density
are associated with differences in affective and cognitive
empathy. NeuroImage 2015, 117:305-310.
52. Patil I et al.: Neuroanatomical basis of concern-based altruism
in virtual environment. Neuropsychologia 2017, S0028-3932
30061-30061.
53. Goerlich-Dobre KS et al.: The left amygdala: a shared substrate
of alexithymia and empathy. NeuroImage 2015, 122:20-32.
54. Fan YT et al.: Empathic arousal and social understanding in
individuals with autism: evidence from fMRI and ERP
measurements. Soc Cogn Affect Neurosci 2014, 9:1203-1213.
55. Vieira JB et al.: Psychopathic traits are associated with cortical
and subcortical volume alterations in healthy individuals. Soc
Cogn Affect Neurosci 2015, 10:1693-1704.
56. Seara-Cardoso A et al.: Neural responses to others’ pain vary
with psychopathic traits in healthy adult males. Cogn Affect
Behav Neurosci 2015, 15:578-588.
57. Olsson A et al.: Vicarious fear learning depends on empathic
appraisals and trait empathy. Psychol Sci 2016, 27:25-33.
58. Meffert H et al.: Reduced spontaneous but relatively normal
deliberate vicarious representations in psychopathy. Brain
2013, 136:2550-2562.
59. Zaki J: Moving beyond stereotypes of empathy. Trends Cogn
Sci 2016.
60. Bucchioni G et al.: Do we feel the same empathy for loved and
hated peers. PLoS ONE 2015, 10:e0125871.
 J et al.: The selfless mind: how prefrontal
61. Majdandic
involvement in mentalizing with similar and dissimilar others
shapes empathy and prosocial behavior. Cognition 2016,
157:24-38.
62. Levy J et al.: Adolescents growing up amidst intractable
 conflict attenuate brain response to pain of outgroup. Proc
Natl Acad Sci U S A 2016, 113:13696-13701.
The results of this magnetoencephalography study of Jewish-Israeli and
Arab-Palestinian adolescents suggest that empathic responses to the
www.sciencedirect.com
The neuroscience of empathy .Marsh 115
pain of members of an ingroup and a threatening outgroup are initially
identical, but later modulated by top-down processes that reduce
empathic responses to the outgroup.
63. Zheng L et al.: Perceived reputation of others modulates
empathic neural responses. Exp Brain Res 2016, 234:125-132.
64. Cui F, Ma N, Luo YJ: Moral judgment modulates neural
responses to the perception of other’s pain: an ERP study. Sci
Rep 2016, 6:20851.
65. Feng C et al.: Social hierarchy modulates neural responses of
empathy for pain. Soc Cogn Affect Neurosci 2016, 11:485-495.
66. Koehne S et al.: Perceived interpersonal synchrony increases
empathy: insights from autism spectrum disorder. Cognition
2016, 146:8-15.
67. Coll MP, Tremblay MB, Jackson PL: The effect of tDCS over the
right temporo-parietal junction on pain empathy.
Neuropsychologia 2017.
68. Soutschek A et al.: Brain stimulation reveals crucial role of
overcoming self-centeredness in self-control. Sci Adv 2016, 2:
e1600992.
69. Wang C et al.: Challenging emotional prejudice by changing
self-concept: priming independent self-construal reduces
racial in-group bias in neural responses to other’s pain. Soc
Cogn Affect Neurosci 2015, 10:1195-1201.
70. Cao Y et al.: Racial bias in neural response to others’ pain is
reduced with other-race contact. Cortex 2015, 70:68-78.
71. Hein G et al.: How learning shapes the empathic brain. Proc Natl
 Acad Sci U S A 2016, 113:80-85.
This study demonstrated a mechanism by which reduced empathy for
outgroup members can be reduced: receiving costly help from outgroup
members yielded a prediction error signal in the insula that (after only a
few trials) predicted subsequent increases in empathy for the outgroup.
72. Klimecki OM et al.: Functional neural plasticity and associated
changes in positive affect after compassion training. Cereb
Cortex 2013, 23:1552-1561.
73. Oatley K: Fiction: simulation of social worlds. Trends Cogn Sci
2016, 20:618-628.
74. Valk SL et al.: Structural plasticity of the social brain:
differential change after socio-affective and cognitive mental
training. Sci Adv 2017, 3:e1700489.
Current Opinion in Behavioral Sciences 2018, 19:110–115