See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/6258160

Goal Shifts Following Reminders of Mortality: Reconciling Posttraumatic

Growth and Terror Management Theory

Article  in  Personality and Social Psychology Bulletin · September 2007

DOI: 10.1177/0146167207303015 · Source: PubMed

CITATIONS READS

82 998

5 authors, including:

Suzanne C Segerstrom Daniel R Evans

University of Kentucky University of Washington Seattle
154 PUBLICATIONS   10,712 CITATIONS    23 PUBLICATIONS   873 CITATIONS   

SEE PROFILE SEE PROFILE

Margaret E Kemeny
University of California, San Francisco
117 PUBLICATIONS   15,657 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Margaret E Kemeny on 29 May 2014.

The user has requested enhancement of the downloaded file.

 Pers Soc Psychol Bull OnlineFirst, published on June 19, 2007 as doi:10.1177/0146167207303015
Goal Shifts Following Reminders of Mortality:
Reconciling Posttraumatic Growth and Terror
Management Theory
Emily L. B. Lykins
Suzanne C. Segerstrom
Alyssa J. Averill
Daniel R. Evans
University of Kentucky
Margaret E. Kemeny
University of California, San Francisco
Research findings within posttraumatic growth (PTG)
and terror management theory (TMT) currently appear
contradictory. Following confrontations with mortality,
PTG research demonstrates intrinsic goal shifts, whereas
TMT suggests extrinsic shifts. The current studies exam-
ine factors contributing to these inconsistent results.
Study 1 demonstrates that perceived death threat is
associated with PTG effects. Study 2 illuminates the
importance of duration of death processing. Study 3
demonstrates that existing goal values and duration and
type of processing all interact in determining ultimate
goal structure, with a match between level of goals and
processing producing the most psychologically advanta-
geous outcomes. Although previous research suggests
that short-term confrontations with death may lead to
defensiveness, the current studies suggest that encoun-
tering death over a longer period or in a manner consis-
tent with goal structure may lead individuals to
transcend defensiveness and maintain intrinsic goals or
become more intrinsically oriented.
Keywords: posttraumatic growth; terror management; intrinsic
and extrinsic goals; earthquake; terrorism
I t is generally agreed that the foreknowledge of death is a
uniquely human characteristic. However, two theories—
posttraumatic growth (PTG) and terror management
theory (TMT)—disagree as to the consequences of this
foreknowledge. PTG findings suggest that facing a life
Copyright 2007 by Society for Personality and Social Psychology, Inc..
crisis, especially one’s own mortality or that of close
others, can lead to “positive psychological change expe-
rienced as a result of the struggle with highly challeng-
ing life circumstances,” a phenomenon known as PTG
(Calhoun & Tedeschi, 1999, 2001; Tedeschi &
Calhoun, 2004, p. 1). Positive changes typically include
a greater appreciation for life, a changed sense of prior-
ities, more intimate relationships with others, a greater
sense of personal strength, recognition of new possibili-
ties for one’s life, and spiritual development (Tedeschi &
Calhoun, 1996). PTG has been demonstrated to occur
in a variety of different circumstances (Tedeschi &
Calhoun, 2004), including cancer (Collins, Taylor, &
Skokan, 1990), bone marrow transplantation (Andry-
kowski, Brady, & Hunt, 1993), heart attack (Affleck,
Tennen, Croog, & Levine, 1987), HIV infection
(Bower, Kemeny, Taylor, & Fahey, 1998), transporta-
tion accident (Joseph, Williams, & Yule, 1993), house
fire (Thompson, 1985), combat (Elder & Clipp, 1989),
Authors’ Note: The authors acknowledge the late George F. Solomon,
principal investigator for Study 1. They thank Peter Grohr for his
assistance with data collection for Study 1 and Andrea Elliot, Natalie
Fantini, Joseph Mayne, Abbey Roach, and Aaron Zink for their assis-
tance with data collection for Study 2. Correspondence concerning
this article should be addressed to Dr. Suzanne C. Segerstrom,
Department of Psychology, 115 Kastle Hall, University of Kentucky,
Lexington, KY 40506-0044; e-mail: scsege0@uky.edu.
PSPB, Vol. XX No. X, Month XXXX xx-xx
DOI: 10.1177/0146167207303015
© by the Society for Personality and Social Psychology, Inc.
2 PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN
sexual assault and abuse (Burt & Katz, 1987), bereave-
ment (Calhoun & Tedeschi, 1989-1990), and divorce
(Wallerstein, 1986), among others. Many, though not
all, circumstances promoting PTG include reminders of
or the potential for personal mortality.
PTG can be characterized as a shift toward intrinsic
goals, which are oriented toward building meaningful
and lasting resources and satisfying essential human
needs for autonomy, relatedness, competence, or
growth (e.g., build closer relationships, help the world
be a better place) and away from extrinsic goals, which
are oriented toward building only transient resources
and satisfying superficial needs (e.g., be attractive, make
more money) (Kasser, 2002; Kasser & Ryan, 1996).
On the other hand, experimental reminders of death
lead to extrinsic shifts in tests of TMT (Arndt, Solomon,
Kasser, & Sheldon, 2004; Dechesne et al., 2003; Kasser
& Sheldon, 2000). TMT posits that humans possess an
instinctive drive for self-preservation but also have an
understanding of the inevitability of death, which can
produce existential anxiety (Greenberg, Pyszczynski, &
Solomon, 1986; S. Solomon, Greenberg, & Pyszczynski,
1991). Humans manage this anxiety by living in accord
with a cultural worldview in which their lives are imbued
with worldview-consistent meaning, permanence, and
stability (Pyszczynski, Greenberg, & Solomon, 1999).
The mortality salience (MS) hypothesis of TMT suggests
that reminding people of their mortality (MS induction)
should increase their need for the structure and meaning
provided by their personal worldview and their need for
self-esteem. Experimental evidence suggests that induc-
tion of MS promotes worldview defense in the form of
greater emphasis placed on personally and culturally val-
ued, but extrinsic, goals such as physical attractiveness,
materialism, and wealth (Arndt et al., 2004; Kasser &
Sheldon, 2000; Routledge, Arndt, & Goldenberg, 2004).
PTG and TMT suggest that reminders of mortality lead
to intrinsic and extrinsic goal value, respectively. This arti-
cle is an attempt to resolve this theoretical paradox. It is
likely that one or more major experiential or method-
ological differences in these two research traditions
account for the different results obtained. PTG research
typically examines naturalistic challenges, including con-
frontations with death (i.e., cancer, heart attacks, HIV dis-
ease) that are typically uncontrollable and occur over
weeks, months, or years, whereas TMT research typically
experimentally manipulates death reminders (e.g., presen-
tations of death-related words) on a single occasion. Thus,
the current research will focus on examining what types of
experiences may prompt TMT- or PTG- consistent out-
comes, which will help develop a better understanding of
the mechanisms underlying the different consequences of
confronting death, which are associated with significantly
different psychological outcomes.
The first difference between these traditions is the
emphasis on mortality per se. As PTG has been demon-
strated in circumstances both including and not includ-
ing direct reminders of mortality (e.g., cancer vs.
divorce), an important initial question is what role
death threat or reminders of mortality play in PTG.
Reminders of mortality in the TMT tradition play a
causal role in demonstrated effects, as these reminders
are under the experimental control of the investigator.
However, in naturalistic studies of PTG, reminders of
mortality can coexist with changes in resources,
lifestyle, or energy level. The determination of what role
perceived mortality plays in PTG may be an important
first step in understanding the differences between the
PTG and TMT literatures.
A second difference between the PTG and TMT tra-
ditions is the duration of mortality reminders. The nat-
urally occurring confrontations with death in the PTG
literature tend to be on a longer term (typically days to
years) than the experimental confrontations of death in
TMT (typically seconds to minutes). Currently, the
effect of length of mortality reminders on goal value has
not yet been investigated empirically.
A third potential difference between the PTG and
TMT traditions is the type of processing associated with
mortality reminders in the two literatures. Cozzolino,
Staples, Meyers, and Samboceti (2004) compared a TMT
manipulation, which typically elicits more defensive pro-
cessing such as distraction and optimistically biased
appraisals (Pyszczynksi et al., 1999), with a death-reflection
manipulation in which participants focused on their own
death in a manner consistent with PTG (Bower et al.,
1998), including an imagined real death, a life review,
and the opportunity to take the perspective of others.
Cozzolino et al. found that those who were initially
extrinsically oriented and engaged in death reflection
took fewer raffle tickets in return for their participation
than if they processed death more generally and
abstractly in the mortality-salience condition. These find-
ings suggest that type of processing of death-related stim-
uli affects behavior associated with selfishness and greed,
which in turn suggests changes in goal structure.
However, although Cozzolino et al. investigated the
influence of type of death processing, they did so on a
cross-sectional basis, which did not allow for an exami-
nation of the influence of duration of processing.
The present studies used both naturalistic and exper-
imental paradigms to investigate the influences of mor-
tality threats and reminders, the passage of time, and
type of processing on intrinsic and extrinsic goal value.
In general, we hypothesized that mortality threats and
reminders are associated with more intrinsic goal value,
but only when these threats and reminders are prolonged
and when thoughts about them are reflective.
 Lykins et al. / GOAL SHIFTS FOLLOWING REMINDERS OF MORTALITY 3

STUDY 1 Procedure
Participants were approached by an investigator at their
The influence of perceived death threat on goal shifts
workplace. The purposes of the study, the procedure, and
consistent with PTG was investigated in the context of
the planned follow-up waves were explained. To ensure a
the Northridge earthquake, which struck the Los
sample of people without severe mental health problems,
Angeles area in the early morning hours of January 17,
participants were excluded if, previous to the earthquake,
1994, and measured 6.7 on the Richter scale. The expe-
they had had a psychiatric hospitalization or were taking
rience of the earthquake could justifiably have reminded
psychiatric medication or were currently experiencing dis-
affected individuals of their own mortality: Fifty-seven
tress so severe that they were not able to function in their
people died as a direct result of the earthquake, includ-
normal activities. Participants were given questionnaires to
ing 16 crushed in a single apartment building that col-
be completed and returned in the mail. Some participants
lapsed. More people died as an indirect result of the
had been transferred to a nearby VAMC at Time 2 and
earthquake; for example, cardiac deaths in the area
were followed up at that site. Time 1 data were collected
doubled on the day of the earthquake (Kloner, Leor,
11-24 days after the earthquake; Time 2 data were col-
Poole, & Perritt, 1997). Thousands of people were
lected 53-67 days after the earthquake.
injured (Coffey, 1994).
In a sample of hospital employees who experienced
the Northridge earthquake in one of the Los Angeles Measures
areas most heavily damaged by the earthquake, we
Demographic information. Participants were asked
hypothesized that surviving the earthquake would lead
to report their gender, age, and ethnic background.
to intrinsic goal shifts consistent with PTG literature, as
their experience would include many long-term
Life Goals Inventory. At Time 2, participants were
reminders of their own mortality. We further hypothe-
asked to indicate how important 16 specific goals were to
sized that the degree of mortality threat experienced
them before and after the earthquake on the Life Goals
during the earthquake would be associated with the
Inventory (Kemeny, 1995). Responses were on a 7-point
degree of intrinsic goal shift, such that greater threat
scale with endpoints 1 = not at all and 7 = extremely. The
would lead to larger shifts.
16 goals were divided into two categories: intrinsic (9
items; e.g., cultivate close friendships, give and receive
Method love, develop myself as a person, do creative work; α =
.89) and extrinsic (7 items; e.g., be attractive, achieve
Participants acknowledgement and praise, have means or possessions,
career advancement; α = .83). Following Kasser and Ryan
Participants were employees of the Sepulveda
(1996), difference scores were calculated by subtracting
Veterans Affairs Medical Center (VAMC) in clinical,
the extrinsic scale mean from the intrinsic scale mean so
administrative, and support services. The Sepulveda
positive values indicate the relative importance of intrinsic
VAMC is located less than 2 km from the epicenter of
over extrinsic goals. The magnitude of the difference score
the Northridge earthquake and sustained severe dam-
relays information as to how much intrinsic goals are
age. Most employees lived near the hospital in the San
prioritized over extrinsic goals (Kasser & Ryan, 1996).
Fernando Valley, and most experienced damage to
their homes and possessions in addition to the damage
Mortality threat. At Time 1, participants were asked,
to their workplace and local infrastructure (roads, elec-
“At any time during the earthquake did you think you
tric, gas, and water; G. F. Solomon, Segerstrom, Grohr,
might die?” and responded on a 7-point Likert-type
Kemeny, & Fahey, 1997). The participants for this
scale with endpoints 1 = not at all and 7 = absolutely.
study were 74 employees who completed question-
Participants reported a wide range of mortality threat.
naires at two waves after the earthquake. Mean age
Mean response to this item was 3.0 (SD = 2.3), with a
was 45 years (range = 24 to 66). The sample was 63%
range from 1 to 7. Approximately one third of the sam-
female and 30% minority. These participants were a
ple responded at or above the scale midpoint.
subset of 113 participants who participated at Time 1.
Those who continued at Time 2 tended to be slightly
older than those who did not continue (M = 38 years), Results
t(111) = 1.77, p < .10. However, there were no differ-
ences in gender or ethnicity or in any of the relevant Mortality Threat and Changes in Goal Value
study variables between those who continued and those Changes in goal value are shown in Table 1. There
who did not. was no effect of the earthquake on goal value generally,
4 PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN
TABLE 1: Reported Changes in Goal Importance Before and After the Northridge Earthquake
Before Earthquake
Mean importance 4.93 (0.99)
Intrinsic mean 5.08 (1.19)
Extrinsic mean 4.63 (1.11)
Mean difference score 0.45 (0.81)
NOTE: Standard deviations in parentheses.
so pre-event and current mean goal value reported at
Time 2 were almost identical before and after the earth-
quake, F(1, 73) = 0.05, ns, η = .03. However, intrinsic
goals were reported to be more valued after the earth-
quake than before, F(1, 73) = 6.48, p < .05, η = .29, and
extrinsic goals were reported to be less valued, F(1,73)
= 8.52, p < .05, η = .32. Consequently, there was a sig-
nificant perception of goal shift, as represented by the
intrinsic-extrinsic difference, from before to after the
earthquake, F(1, 73) = 24.53, p < .05, η = .50.
Reports of a change in goal value from before the
earthquake to 8 weeks after the earthquake, that is, per-
ceived goal shift at Time 2, was significantly related to
the perception that one’s life was at risk during the earth-
quake as reported at Time 1. Thoughts of more certain
death during the earthquake predicted a greater perceived
goal shift (r = .38, p < .001). People who feared most for
their lives during the earthquake were later most likely to
endorse increased value placed on intrinsic goals (r = .24,
p < .04) and, to a lesser degree, decreased value placed on
extrinsic goals (r = –.19, p < .11).
Discussion
Following the Northridge earthquake, hospital
employees reported shifts in their personal goals away
from extrinsic goals such as “please people, be liked,
popular” and “have means or possessions” and toward
intrinsic goals such as “maintain close relationships with
relatives” and “accomplish things in life.” In addition,
more certain thoughts of dying during the earthquake,
recalled in the days afterward, predicted greater intrin-
sic goal shifts 6 weeks later. Thus, a reminder of one’s
death on the magnitude of surviving an earthquake
appears to lead to intrinsic goal shifts. That greater per-
ceived mortality threat during the earthquake predicted
greater perceived intrinsic goal shifts over 8 weeks sug-
gests that mortality threat does play a role in PTG.
Although this finding does not dispute the additional
role of changes in other areas such as resources,
lifestyle, or energy level, this study does demonstrate
that the magnitude of PTG outcomes is influenced by
the magnitude of mortality threat. Thus, the differences
between the PTG and TMT literatures cannot simply be
After Earthquake
4.92 (0.89)
5.31 (0.97)
4.41 (1.07)
0.90 (0.97)
explained by a lack of influence of mortality threat in
PTG outcomes.
The question of whether people can accurately recall
their previous goal values after they have experienced a
traumatic event must be addressed. It is possible that
after mortality threat, people devalue their prior intrin-
sic goals in order to perceive goal shift where none has
actually occurred. One remedy would be to assess goals
before and after a traumatic event; however, most such
events are difficult to predict and therefore to study in a
prospective design. Another remedy is to compare goal
shifts occurring after mortality threat to those occurring
after a benign event for which people should not be
motivated to devalue their prior selves, as changes in the
mortality threat group can be separated out from goal
shifts normally reported over time. This approach was
employed in Study 2, in which the influence of duration
of mortality threat was also investigated.
STUDY 2
The influence of duration of mortality reminders was
investigated by examining perceived goal shifts occur-
ring over a 6-week period in a sample of midwestern
undergraduate students who either vicariously experi-
enced the terrorist attacks of September 11, 2001, or
were completing a normal semester 3 years later.
Mortality reminders associated with the terrorist
attacks of 9/11, which killed almost 3,000 people, were
experienced widely, albeit indirectly. Sixty-nine percent
of American households were estimated to have
watched prime-time television on the day of the attacks,
when most programming focused on the terrorist
attacks (Kissell, 2002), and American adults were esti-
mated to have watched approximately 8.1 hr of televi-
sion coverage of the attacks by 3 to 5 days after the
attacks (Schuster et al., 2001). This coverage included
scenes of airplanes crashing into the World Trade
Center, the wreckage of the planes that crashed into the
Pentagon and the Pennsylvania countryside, and the
extreme fear and distress experienced by those surviving
the attacks. The extensive media coverage of the terror-
ist attacks following 9/11, as well as messages about the
 Lykins et al. / GOAL SHIFTS FOLLOWING REMINDERS OF MORTALITY
continued threat of terrorism, likely served to remind
many Americans of their own mortality.
Preliminary evidence seems to support the experience
of increased intrinsic goal value and PTG in Americans
following the terrorist attacks of September 11, 2001.
Americans reported higher levels of character strengths
such as spirituality, hope, kindness, leadership, love,
gratitude, and teamwork after the attacks of 9/11 than
before (Peterson & Seligman, 2003) and reported the
initiation of prosocial or altruistic behaviors, intensifi-
cation of the quest for value or meaning and relational
investment, and increased feelings of patriotism (Yum
& Schnuck-Hamlin, 2005).
Study 2 allows for an examination of the influence of
duration of mortality reminders and additionally offers
broader generalizability to life events. Because these
participants experienced the 9/11 attacks vicariously,
they did not experience mortality threat to the degree
that the earthquake sample did. Such reminders are
more likely to occur to the average person than are
direct experiences, so evidence of goal shifts following
indirect confrontations with mortality are more applic-
able to daily life.
We hypothesized, this time in a between-groups
design, that a sample that had recently experienced the
9/11 terrorist attacks (the 2001 sample) would report
greater intrinsic goal shifts than a sample that had not
experienced a recent attack (the 2004 sample). We
further hypothesized that these effects would be more
evident with passage of time since the attacks.
Method
Participants
The participants were 225 undergraduate students at
a large midwestern university who participated for par-
tial course credit. At the first of two time points (Time 1),
110 students participated in 2001, and 115 students par-
ticipated in 2004. In 2001, the initial sample size
decreased from N = 110 at Time 1 to n = 92 at Time 2
(16% attrition). In 2004, the initial sample size decreased
from N = 115 to n = 95 (17% attrition). In 2001, there
were no significant differences in age or gender between
the individuals who completed the study and those who
dropped out. However, a higher proportion of minorities
dropped out of the study, χ2(1) = 11.16, p = .001. In
2004, there were no significant differences in age, gender,
or minority status between those who did and those who
did not complete the study.
The demographic characteristics of the final samples
in 2001 and 2004 are as follows: In 2001, the final sample
of 92 participants was 82% female and 96% Caucasian,
with an average age of 18.41 years (range = 18 to 25). In
5
2004, the final sample of 95 participants was 92%
female and 91% Caucasian, with a mean age of 19.04
years (range = 18 to 44). The samples from 2001 and
2004 with complete data had no significant age or
minority participation differences, though the 2004
sample did have a lower proportion of males than the
2001 sample, χ2(1) = 4.58, p < .05.
Procedure
Participants completed questionnaires in small groups
at two time points. In 2001, the first assessment took
place during the week of October 15, 2001, approxi-
mately 5 weeks after the September 11 attacks. The sec-
ond assessment took place 6 weeks later, during the
week of November 26, 2001, which was approximately
11 weeks after the attacks. In 2004, the first assessment
took place during the week of October 18, 2004,
approximately 8 weeks after the initiation of the semes-
ter. The second assessment took place 6 weeks later,
during the week of November 29, 2004, approximately
14 weeks after the semester’s initiation. All measures
were administered at both time points.
Measures
Demographic information. Participants were asked
to report their gender, age, and ethnic background.
Life Goals Inventory. At Time 1 and Time 2, partic-
ipants were asked to complete the Life Goals Inventory
(Kemeny, 1995), in which they were asked to indicate
how important the same 16 goals used in Study 1 were
to them both currently and before a designated time
point. In 2001, that time point was “before the terror-
ist attacks of September 11, 2001.” In 2004, that time
point was “before the beginning of the current semes-
ter.” Alpha reliabilities were acceptable for all time
frames at all time points (intrinsic, .79 to .84.; extrin-
sic, .73 to .79).
Data Analysis
To use a sample with a roughly equivalent experience
with the 9/11 terrorist attacks, individuals who knew a
close friend, relative, or acquaintance who was injured
or killed (n = 22, with 11 knowing someone who was
killed) were excluded from all analyses. Repeated-
measure ANOVAs were conducted for the intrinsic—
extrinsic difference scores, using the Time 1 pre-event
retrospective, Time 1 current, and Time 2 current val-
ues for each participant. Group (2001 vs. 2004 sample)
was included as a between-subjects factor.
6 PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN
Results
Validity of Retrospective Reports of Goal Value
We employed two comparisons to assess the validity
of retrospective reports of goal value prior to a threaten-
ing event. First, we calculated the differences between ret-
rospective reports of pre-event goal value at two time
points, 5 and 11 weeks after the event in 2001 and 8 and
14 weeks after the event in 2004. If motivational factors
affect retrospective reports, then there may be changes in
motivation and retrospective report between those two
time points. Conversely, lack of change would tend to
support accurate reporting. Test-retest reliability was
high, with correlations between retrospective reports at
Time 1 and Time 2 of .73 for intrinsic goals, p < .001,
and of .73 for extrinsic goals, p < .001. Second, we com-
pared retrospective reports by the sample asked to report
on goal shifts after the terrorist attacks to those by the
sample asked to report on goal shifts after starting col-
lege. If motivational factors affect retrospective reports,
the former sample should devalue their pre-event goals
relative to the latter. A repeated-measures ANOVA with
goal type (within; extrinsic, intrinsic), time of retrospec-
tive report (within; Time 1 pre-event, Time 2 pre-event),
and group (between; 2001, 2004) was conducted to
examine these motivational factors. Overall goal values
were consistent across time, F(1, 149) = 1.16, ns, η = .09,
did not vary across time by group, F(1, 149) = 1.66, ns,
η = .10, and did not vary by group and goal type across
time, F(1, 149) = 2.31, ns, η = .12. These analyses demon-
strate that the retrospective reports of pre-event goal
value were not affected by motivational factors, as the
reported pre-event goal values were consistent at Times 1
and 2 and did not vary by group or group by goal value
at Times 1 and 2. Thus, for all analyses, the Time 1 pre-
event goal values will be used as the “before” value, due
to its temporal proximity to the pre-event time frame.
Changes in Goal Value
From an initial equivalent pre-event difference score,
t(149) = –1.58, ns, difference scores became more
intrinsic over time in the combined sample, F(2, 296) =
20.24, p < .001, η = .35. However, this was qualified by
a significant group by time interaction, F(2, 296) =
11.18, p < .001, η = .26, in which the 2001 group
showed more intrinsic goal shift (difference scores of
0.61 pre-event, 1.93 at Time 1, and 1.72 at Time 2)
than did the 2004 group (difference scores of 1.09 pre-
event, 1.40 at Time 1, and 1.08 at Time 2). Figure 1
shows the relative value of goals for the 2001 and 2004
groups: The group that was indirectly exposed to mor-
tality threat reported a larger shift toward intrinsic
goals than did the control group. The difference scores
2.5
Mean Difference Score
2
1.5
1
0.5
0
Before Time 1 Time 2
Time Point
2001 2004
Figure 1 Mean Difference Scores as a Function of Group and Time
Point in Study 2
NOTE: Higher scores represent more value attached to intrinsic over
extrinsic goals.
of the groups did not significantly differ at baseline (see
above) or at Time 1, t(149) = 1.84, ns, though they sig-
nificantly differed at Time 2, t(148) = 2.50, p = .01.
Discussion
A sample recently exposed to the 2001 terrorist
attacks perceived greater intrinsic goal shift over time as
compared with a parallel control sample assessed in
2004. Although the experimentally controlled, short-
term confrontations with death in TMT may trigger
defensive processing, this study suggests that confronting
more real-life and personal reminders of one’s mortality
may eventually promote psychological growth, as demon-
strated in goal shifts, though it may not promote such
growth immediately, as group differences were demon-
strated at Time 2 but not at Time 1.
Studies 1 and 2 demonstrate that confronting more
real-life and personal reminders or threats toward per-
sonal mortality can produce perceived intrinsic goal
shifts consistent with PTG, suggest that degree of mor-
tality threat is associated with degree of perceived
intrinsic goal shift, and suggest that duration of pro-
cessing does play an important role in goal shifts.
However, these studies used traditional PTG paradigms
in which people who were not randomly assigned to
degree of mortality threat retrospectively reported
changes in their goals. Furthermore, the influence of
type of death-related processing has not yet been
addressed. Thus, Study 3 used an experimental para-
digm to examine the joint influences of duration of pro-
cessing, type of processing, and pre-existing goal values
 Lykins et al. / GOAL SHIFTS FOLLOWING REMINDERS OF MORTALITY
in determining the effects of reminders of mortality on
perceived goal shifts.
STUDY 3
A midwestern student sample reported on current goal
values both before and after completing 6 days of pro-
cessing in one of three topic conditions: death reflection,
mortality salience, or control (dental pain). The death-
reflection manipulation was designed to capture the key
elements of near-death experience, contains elements
related to PTG, and may lead to shifts toward intrinsic
goals (Cozzolino et al., 2004), whereas the mortality-
salience manipulation was consistent with typical MS
inductions that have resulted in shifts toward extrinsic
goals (Routledge et al., 2004). We hypothesized that type
of processing would influence perceived goal values and
goal shifts such that on Day 1, the participants in the
death-reflection condition would report increased intrin-
sic goal value and lower extrinsic goal value than those in
the mortality-salience condition, with those in the control
condition reporting values between the two death-related
groups. In addition, we hypothesized that duration of
processing and type of processing would interact over the
6-day processing period. Those participants in the death-
reflection condition would experience early intrinsic goal
shift whereas those in the mortality-salience condition
would report early extrinsic goal shift, but as a function
of repeated mortality reminders, those in the mortality-
salience condition would also report intrinsic goal shift
by Day 6. Those in the control condition would report
minimal or no shift in goal values. Finally, we predicted
that Day 1 goal values would interact with duration of
processing and type of processing such that relatively
intrinsic individuals would experience less positive
change than relatively extrinsic individuals over the pro-
cessing period because they already value intrinsic goals
(Cozzolino et al., 2004).
Method
Participants
The participants were 149 undergraduate students at
a large midwestern university who participated for par-
tial course credit. Participants were eligible for the study
if they were 18 years of age and were able to read and
write in English. The initial sample size decreased from
N = 149 at Day 1 to n = 103 at Day 6 (31% attrition).
There were no significant differences in age, gender, or
minority status between the individuals who completed
the study and those who dropped out or were terminated
from the study. In the sample of individuals completing
7
the full study, mean age was 19.1 years (range = 18 to 29).
The sample was 80% female and 10% minority.
Procedure
Three groups of participants were randomly assigned
to one of three conditions, each of which involved 6
total days of participation. In the first sessions, each
group of participants was seated in a common room,
received information about the study, and signed an
informed consent form. Participants then spent 10 min
engaged in a processing task that consisted of answering
open-ended essay questions that comprised the experi-
mental manipulation. The participants also completed
the demographic questionnaire, the Life Goals Inventory,
and additional filler questionnaires.
Members of the death-reflection group were pre-
sented a vivid half-page scenario in which they were
asked to imagine themselves caught in a building fire,
frantically attempting to escape, but eventually suc-
cumbing to smoke and fire. They were then asked to
describe the thoughts and emotions they felt while
imagining the scenario, discuss how they would handle
the final moments, describe the life they had led up to
that point, and discuss how they think their family and
friends would react if the event actually occurred. The
mortality-salience group members were asked to simply
describe the thoughts, feelings, and emotions that the
thought of their own death aroused in them, as well as
describing what they think will happen to them physi-
cally as they die. The dental-pain condition participants
were asked a parallel set of questions related to dental
pain, describing the thoughts, feelings, and emotions
that the thought of dental pain aroused in them, as well
as describing what they think will happen to them phys-
ically as they experience dental pain.
On the next 4 consecutive days (Days 2 to 5), partici-
pants completed their respective 10-min processing task,
submitting their responses daily through the Internet. On
Days 2 and 4, the death-processing group was asked to
write about their thoughts and emotions and how they
would handle the final moments; the mortality-salience
group was asked to write about the thoughts, feelings,
and emotions that the thought of their own death
aroused in them; and the dental-pain group was asked to
write about the thoughts, feelings, and emotions that the
thought of dental pain aroused in them. On Days 3 and
5, the death-processing group was asked to write about
the life they had led up to the point of their death and
how their family and friends would react; the mortality-
salience group was asked to write about what will hap-
pen to them physically after death; and the dental-pain
group was asked to write about what will happen to
them physically as they experience dental pain.
8 PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN

TABLE 2: Content Differences Among Groups: Day 1

Content Category Death Reflection Mortality Salience Dental Pain p (F) <

Positive affect .09 (.16)a .09 (.14)a .03 (.08)a .13

Negative affect .47 (.19)a .39 (.19)a .17 (.19)a .08
Physical reality .06 (.13)a .55 (.18)b .62 (.25)b .001
Physical reaction .04 (.11)a .05 (.13)a .42 (.32)b .001
Life reflection .41 (.16)a .05 (.11)b .00 (.00)b .001
Life goals .10 (.16)a .16 (.21)b .00 (.00)a .001
Life regrets .13 (.18)a .03 (.08)b .00 (.00)b .001
Thoughts of others .55 (.13)a .25 (.20)b .01 (.07)c .001
Selfish thoughts of others .03 (.08)a .06 (.12)a .00 (.00)a .03
Religion .17 (.18)a .21 (.21)a .00 (.00)b .001
Personal mortality .05 (.11)a .28 (.16)b .01 (.08)a .001
Universal mortality .01 (.04)a .11 (.18)b .00 (.00)a .001

NOTE: Means not sharing subscripts are significantly different after Bonferroni correction (α = .004).

On the final day of the study (Day 6), participants
completed their final 10-min processing task, using the
same prompts as on Day 1. They also again completed
the Life Goals Inventory and additional filler question-
naires. After submitting their answers, participants were
presented with online debriefing information. Student
participants received research credits commensurate
with the number of days they completed, but they were
dropped from the study if they failed to submit their
answers by midnight on any active study day.
.78 to .99. Following discussion of disagreements, individ-
ual raters then coded the remaining responses. Following
Cozzolino et al., all content codings were transformed into
proportions (content category coding matches/total content
coding matches) for each participant to account for indi-
vidual differences in the length of responses. Distributions
of these proportions were skewed and therefore were
square-root transformed to improve normality.
Results

Measures Content Differences Among Groups

Demographic information. Participants were asked
to report their gender, age, and ethnic background.
Life Goals Inventory. On Days 1 and 6, participants
were asked to complete the Life Goals Inventory
(Kemeny, 1995), in which they were asked to indicate
how important the same 16 goals used in Studies 1 and
2 were to them currently. Alpha reliabilities were
acceptable for all time frames at all time points (intrin-
sic, .67 on Day 1 and .83 on Day 6; extrinsic, .75 on
Day 1 and .74 on Day 6).
Content Coding
Based on Cozzolino et al. (2004), responses to the pro-
cessing tasks on Days 1 and 6 were coded using 12 cate-
gories: positive affect; negative affect; physical reality of
dying or dental pain (e.g., pain); physical reactions to
thinking about death or dental pain (e.g., increased heart
rate, “stomach dropped”); life reflections or summary; life
goals; life regrets; thoughts of others; selfish thoughts of
others; religious references; personal mortality (e.g., “I
know I am going to a better place”); and universal mortal-
ity (“Death is natural or inevitable”). Initially, two raters
coded each of 45 responses, yielding alpha reliabilities of
One-way ANOVAs with Bonferroni correction (α =
.004) revealed significant differences among the three
experimental groups in all categories of response content
except positive affect, negative affect, and selfish thoughts
of others. Table 2 shows the group means of the square-
root transformed variables on Day 1. To summarize, the
dental-pain group had the most references to physical
reality and physical reactions; both death-topic groups
had more references to religion than the dental-pain
group; the mortality-salience group had more references
to mortality, both personal and universal, and life goals
than the other two groups; and the death-reflection group
had more references to life reflection, life regrets, and
thoughts of others than the other two groups. These pat-
terns were maintained at Day 6, with the exceptions that
the death-reflection group had more references to life
goals than the dental-pain group, and the difference in life
regrets between death reflection and mortality salience no
longer reached Bonferroni-corrected significance.
Goal Differences Among Groups
At Day 1, for the total sample, there were no signifi-
cant group differences in intrinsic or extrinsic goal value
or in goal value difference scores. For the sample that
 Lykins et al. / GOAL SHIFTS FOLLOWING REMINDERS OF MORTALITY
completed through Day 6, there were no significant dif-
ferences on Day 1 between groups in their extrinsic goal
value or in the difference scores. The groups did differ
significantly in their intrinsic goal value, F(2, 102) = 3.37,
p < .05, η = .24, with the dental-pain group members
reporting higher value attached to intrinsic goals (M =
53.52) than either death-processing group (death reflec-
tion M = 50.09; mortality salience M = 50.55).
On Day 6, there were no significant differences
between groups in their goal value difference score
after controlling for Day 1 goal value, F(2, 89) = 0.04,
ns, η = .03. However, there was an interaction between
group and initial goal value, F(2, 89) = 3.92, p < .05,
η = .28. Figure 2 shows the results from regression
models run for each group, with low initial values
equal to –1 SD and high initial values equal to +1 SD.
First, there was evidence for regression toward the
mean; those with high initial values tended to decrease,
whereas those with low initial values tended to
increase. However, this tendency differed significantly
among groups. Participants with high initial values
decreased the least in the death-reflection group and
most in the mortality-salience group; participants with
low initial values increased the most in the mortality-
salience group and least in the death-reflection group.
The impression that the differences in goal change were
driven by the two death-topic groups is confirmed by a
significant interaction between initial goal value and
these two groups, F(1, 61) = 7.21, p < .05, η = .33; the
interaction was not significant when restricted to either
death-topic group and the dental-pain group.
Analysis of intrinsic and extrinsic goals separately
suggested that this effect was due to their combined
influence on the difference score, because the initial
value by group interaction was not significant for either
one: intrinsic, F(2, 89) = 1.43, ns, η = .17; extrinsic, F(2,
89) = 2.03, ns, η = .20.
Initial Goals and Content Changes
The possibility that content changes mediated these
differences was explored in the two death-topic groups.
After Bonferroni correction (α = .004) and controlling
for initial content, there were several main effects of
group at Day 6 such that the death-reflection group was
lower in physical reality and higher in life reflection and
thoughts of others. There were no significant interac-
tions between initial content and group; the largest
effect was for life reflection (p < .01). However, con-
trolling for life reflection at Day 1 and Day 6 did not
affect the initial goal value by group interaction, F(1,
59) = 7.64, p < .05, η = .35. Therefore, this particular
content category did not account for the effect of group
and initial goal value on later goal value.
9
6
Goal Change
4
2
0
−2
−4
−6
Death Mortality Dental pain
reflection salience
Experimental Group
High initial value (+1 SD)
Low initial value (−1 SD)
Figure 2 Predicted Changes in Goals From Day 1 to Day 6 in
Study 3 for Participants High and Low in Initial Goal
Emphasis by Experimental Group
Discussion
Duration and type of processing and pre-existing goal
values are all important to the type of outcomes observed
or perceived following a prolonged confrontation with
reminders of personal mortality. Although no reliable
group differences were found in difference scores at Day 1
or in goal shifts occurring from Day 1 to Day 6, individu-
als who reported more intrinsic goals at Day 1 appeared
most able to maintain their intrinsic goal orientation
across time in the death-reflection condition, whereas indi-
viduals with higher extrinsic goals at Day 1 appeared to
experience the most intrinsic goal shifts in the mortality-
salience condition. Thus, some combinations of pre-existing
goal values and processing type are more psychologically
advantageous across time t
han others.
Death reflection led to the most intrinsic goal value for
individuals who were already intrinsically oriented and
the least for those who were already extrinsically oriented.
Following confrontations with mortality, processing
focused on the quality of one’s life to date (e.g., reflection,
regrets) may reinforce the value of intrinsic goals for those
who already value them but may also reinforce extrinsic
goal values for those who already value them. In contrast,
rehearsal of one’s own theory of what happens to oneself,
one’s body, or humans generally following death, which was
the content of the processing conducted by the mortality-
salience group, may actually lead to more intrinsically ori-
ented goals over time for individuals who possess more
extrinsic goals initially.
The experimental groups maintained their content dif-
ferences over time, suggesting that extrinsically oriented
individuals are not simply changing the nature of their
processing over time to produce intrinsic goal shifts. This
suggests that the psychological outcomes of prolonged
processing of death reminders may be influenced by the
10 PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN
“match” between type of goal structure (deep for intrinsic,
shallow for extrinsic) and type of processing done (deep
for death reflection, shallow for mortality salience).
Unlike Cozzolino et al. (2004), we did not find effects
of condition in our cross-sectional data analysis, and the
longitudinal patterns suggested here are almost com-
pletely opposite. However, we examined perceived goal
values at two time points, whereas Cozzolino et al. exam-
ined a behavioral outcome. It is possible that behavioral
changes dissociate from goal values, at least after acute
experimental manipulation, or that altruistic behavior is
unrelated to generally intrinsic and extrinsic goal values
as measured by the Life Goals Inventory. Furthermore,
the content analysis patterns are generally consistent with
Cozzolino et al. but do not match exactly. Thus, future
work needs to examine how outcome type and content
type affect the results of death reflection.
GENERAL DISCUSSION
PTG and terror management findings appear contra-
dictory, with PTG suggesting that confrontations with
death lead to intrinsic goal shift and TMT suggesting
that they lead to extrinsic goal shift. The present studies
demonstrate that differences in duration of those con-
frontations and the type of processing that occurs after-
ward are important factors in determining the type of
goal change that results. Studies of PTG typically
include long-term reminders of death, which appear to
promote maintenance of intrinsic goals or intrinsic goal
shifts. Alternately, studies in TMT typically include very
short-term reminders of death, which here led to no dis-
cernable group differences but in other studies have
elicited behavior consistent with extrinsic goal values
such as greed and vanity (Cozzolino et al., 2004;
Routledge et al., 2004). Although individual goal valu-
ation and type of processing influence goal outcomes,
the main factor influencing differences between PTG
and TMT results may be the influence of time, as longer
duration of processing was related to outcomes consis-
tent with PTG across all studies. The current research
suggests that when people encounter death over a
longer period of time or in a manner consistent with
their goal structure, they move to transcend their defen-
siveness, maintain or become more intrinsically ori-
ented, and may end up healthier in the long term.
Research on the correlates of intrinsic and extrinsic
goals demonstrates that the relative centrality of intrinsic
goals is positively related to well-being, adjustment, and
self-actualization and is negatively related to anxiety. In
contrast, the relative focus on extrinsic goals is related
to lower levels of happiness, higher susceptibility to psy-
chological disorders in adolescents, and less loving and
more conflicted relationships with friends and romantic
partners (Kasser, 2002; Kasser & Ryan, 1996;
Schmuck, Kasser, & Ryan, 2000; Sheldon, Ryan, Deci,
& Kasser, 2004). In addition, individuals valuing
extrinsic over intrinsic goals may adopt contingencies of
self-worth, make more frequent social comparisons,
and diminish time available for intrinsic goals (Sheldon
et al., 2004). Thus, maintaining an intrinsic goal orien-
tation (as intrinsically oriented individuals did best in
the death-reflection condition) or making an intrinsic
shift in goal valuation (as did the extrinsically oriented
individuals in the mortality-salience condition) across
time will likely maintain or lead to new, qualitatively
positive effects. As intrinsic goal shifts appear to lead to
the building of important resources (i.e., close relation-
ships, feelings of competence), it is difficult to imagine
that these effects would be transient. Thus, long-term
changes in goal value appear possible as a link between
confrontations with death and PTG.
Of course, Study 3 suggests that psychologically
advantageous goal shifts or the maintenance of health-
ier goal values will only occur following processing that
is matched to an individual’s goal value structure over
time, which suggests that the current findings may be
used in an intervention framework. Encouraging intrin-
sically oriented individuals who are confronting death
to reflect on life, process life regrets, and think of others
may promote PTG through continued valuation of
intrinsic goals. Alternately, encouraging individuals
with more extrinsic goals to process their theory of
what happens following death as well as current life
goals may help promote PTG over time through intrin-
sic goal shifts. However, a mismatch between level of
processing and goal values may lead to a devaluation of
intrinsic goals, or extrinsic shifts. Very traumatic events
or reminders of one’s mortality create tremendous psy-
chosocial needs. Thus, it is quite important to ensure
that clinicians are helping clients face death in a manner
that will have the most psychologically beneficial out-
comes, and the current research suggests that goal val-
ues and type of processing must be considered before
beginning an intervention, though death-reflection pro-
cessing may be feasible for all individuals, initially or
once their goals begin to become more intrinsically ori-
ented. The promotion of PTG is a reasonable therapeu-
tic goal; for example, it mediates the physiological
benefit of cognitive-behavioral therapy in breast cancer
patients (Cruess et al., 2000; McGregor et al., 2004).
Although this series of studies resolves much of the
PTG versus TMT paradox, there are important ques-
tions remaining. In the absence of information on pro-
cessing type in Studies 1 and 2, the three-way
interaction of processing type, initial goal value, and
duration could only be examined in Study 3. In
 Lykins et al. / GOAL SHIFTS FOLLOWING REMINDERS OF MORTALITY
addition, work examining the influence of all three fac-
tors in naturally occurring confrontations with death is
necessary, as the only study examining the influence of
all three factors simultaneously was an experimental
study. However, the current studies do point to the
influences of all three factors and interactions between
the factors that may have clinical importance in long-
term psychological outcomes. The literature would ben-
efit from future studies examining such questions as
whether different types of mortality reminders automat-
ically trigger different types of processing, whether
people with different goal-value structures are naturally
drawn to different types of processing, why a match
between goal values and processing type is more advan-
tageous than a mismatch, and how long goal shifts fol-
lowing mortality reminders are maintained. Further
work using both experimental and naturalistic methods
may help illuminate why duration of processing appears
to lead to beneficial outcomes in naturally occurring
confrontations with death such as those examined in
Studies 1 and 2.
Confrontations with mortality are a fact of life. With
the extensive access to news outlets in our society, the
goal to stay informed about world events can induce
frequent reminders of mortality. However, the current
research suggests that reminders of our own mortality
can lead to beneficial outcomes such as intrinsic goal
shifts over time, with the caveat that an individual’s
goal-value structure and the type of processing may
need to be matched with one another. Furthermore,
when reminders of mortality reach the quality of
trauma, as in life-threatening disease or disaster, intrin-
sic goal-structure maintenance or intrinsic shifts may
produce PTG by building up personal and relational
resources that the individual can draw upon in times
of need. Intrinsic goals may allow for growth, instead
of immobility or defensiveness, to occur following a
confrontation with the inevitability of mortality.
REFERENCES
Affleck, G., Tennen, H., Croog, S., & Levine, S. (1987). Causal attri-
bution, perceived control, and recovery from a heart attack.
Journal of Social and Clinical Psychology, 5, 339-355.
Andrykowski, M. A., Brady, M. G., & Hunt, J. W. (1993). Positive
psychological adjustment in potential bone marrow transplant
recipients: Cancer as a psychosocial transition. Psycho-Oncology,
2, 261-276.
Arndt, J., Solomon, S., Kasser, T., & Sheldon, K. M. (2004). The urge
to splurge revisited: Further reflections on applying terror man-
agement theory to materialism and consumer behavior. Journal of
Consumer Psychology, 14, 225-229.
Bower, J. E., Kemeny, M. E., Taylor, S. E., & Fahey, J. L. (1998).
Cognitive processing, discovery of meaning, CD 4 decline, and
AIDS-related mortality among bereaved HIV-seropositive men.
Journal of Consulting and Clinical Psychology, 66, 979-986.
11
Burt, M. R., & Katz, B. L. (1987). Dimensions of recovery from rape:
Focus on growth outcomes. Journal of Interpersonal Violence, 2,
57-81.
Calhoun, L. G., & Tedeschi, R. G. (1989-1990). Positive aspects of
critical life problems: Recollections of grief. Omega, 20, 265-272.
Calhoun, L. G., & Tedeschi, R. G. (1999). Facilitating posttraumatic
growth: A clinician’s guide. Mahwah, NJ: Lawrence Erlbaum.
Calhoun, L. G., & Tedeschi, R. G. (2001). Posttraumatic growth: The
positive lessons of loss. In R. A. Neimeyer (Ed.), Meaning recon-
struction and the experience of loss (pp. 157-172). Washington,
DC: American Psychological Association.
Coffey, Shelby, III (Ed.). (1994). 4:31–Images of the 1994 Los
Angeles Earthquake. Los Angeles: Los Angeles Times.
Collins, R. L., Taylor, S. E., & Skokan, L. A. (1990). A better world
or a shattered vision? Changes in life perspectives following vic-
timization. Social Cognition, 8, 263-285.
Cozzolino, P. J., Staples, A. D., Meyers, L. S., & Samboceti, J. (2004).
Greed, death, and values: From terror management to transcen-
dence management theory. Personality and Social Psychology
Bulletin, 30, 278-292.
Cruess, D. G., Antoni, M. H., McGregor, B. A., Kilbourn, K. M.,
Boyers, A. E., Alferi, S. M., et al. (2000). Cognitive-behavioral
stress management reduces serum cortisol by enhancing benefit
finding among women being treated for early stage breast cancer.
Psychosomatic Medicine, 62, 304-308.
Dechesne, M., Pyszczynski, T., Arndt, J., Ransom, S., Sheldon, K. M.,
van Knippenberg, A., et al. (2003). Literal and symbolic immor-
tality: The effect of evidence of literal immortality on self-esteem
striving in response to mortality salience. Journal of Personality
and Social Psychology, 84, 722-737.
Elder, G. H., Jr., & Clipp, E. C. (1989). Combat experience and emo-
tional health: Impairment and resilience in later life. Journal of
Personality, 57, 311-341.
Greenberg, J., Pyszczynski, T., & Solomon, S. (1986). The causes and
consequences of a need for self-esteem: A terror management
theory. In R. F. Baumeister (Ed.), Public self and private self
(pp. 189-212). New York: Springer-Verlag.
Joseph, S., Williams, R., & Yule, W. (1993). Changes in outlook fol-
lowing disaster: The preliminary development of a measure to
assess positive and negative responses. Journal of Traumatic
Stress, 6, 271-279.
Kasser, T. (2002). The high price of materialism. Cambridge, MA:
MIT Press.
Kasser, T., & Ryan, R. M. (1996). Further examining the American
dream: Differential correlates of intrinsic and extrinsic goals.
Personality and Social Psychology Bulletin, 2, 280-287.
Kasser, T., & Sheldon, K. M. (2000). Of wealth and death:
Materialism, mortality salience, and consumption behavior.
Psychological Science, 11, 348-351.
Kemeny, M. E. (1995). The Life Goals Inventory. Unpublished man-
uscript.
Kissell, R. (2002). Auds show 9/11 apathy. Viewers again gravitate
towards CBS documentary. Retrieved November 25, 2005, from
http://www.variety.com/index.asp?layout=story&articleid=VR11
17872705&categoryid=14&cs=1
Kloner, R. A., Leor, J., Poole, W. K., & Perritt, R. (1997). Population-
based analysis of the effect of the Northridge earthquake on car-
diac death in Los Angeles County, California. Journal of the
American College of Cardiology, 30, 1174-1180.
McGregor, B. A., Antoni, M. H., Boyers, A., Alferi, S. M., Blomberg,
B. B., & Carver, C. S. (2004). Cognitive-behavioral stress man-
agement increases benefit finding and immune function among
women with early-stage breast cancer. Journal of Psychosomatic
Research, 56, 1-8.
Peterson, C., & Seligman, M. E. P. (2003). Character strengths before
and after September 11. Psychological Science, 14, 381-384.
Pyszczynski, T., Greenberg, J., & Solomon, S. (1999). A dual-process
model of defense against conscious and unconscious death-related
thoughts: An extension of terror management theory.
Psychological Review, 106, 835-845.
Routledge, C., Arndt, J., & Goldenberg, J. L. (2004). A time to tan:
Proximal and distal effects of mortality salience on sun exposure
 12 PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN
intentions. Personality and Social Psychology Bulletin, 30,
1347-1358.
Schmuck, P., Kasser, T., & Ryan, R. M. (2000). Intrinsic and extrin-
sic goals: Their structure and relationship to well-being in German
and U.S. college students. Social Indicators Research, 50, 225-241.
Schuster, M. A., Stein, B. D., Jaycox, L. H., Collins, R. L., Marshall,
G. N., Elliott, M. N., et al. (2001). A national survey of stress
reactions after the September 11, 2001, terrorist attacks. The New
England Journal of Medicine, 345, 1507-1512.
Sheldon, K. M., Ryan, R. M., Deci, E. L., & Kasser, T. (2004). The
independent effects of goal contents and motives on well-being:
It’s both what you pursue and why you pursue it. Personality and
Social Psychology Bulletin, 30, 475-486.
Solomon, G. F., Segerstrom, S. C., Grohr, P., Kemeny, M., & Fahey, J.
(1997). Shaking up immunity: Psychological and immunologic changes
after a natural disaster. Psychosomatic Medicine, 59, 114-127.
Solomon, S., Greenberg, J., & Pyszczynski, T. (1991). A terror man-
agement theory of social behavior: The psychological functions of
self-esteem and cultural worldviews. In M. P. Zanna (Ed.),
View publication stats
Advances in experimental social psychology (Vol. 24, 91-159).
San Diego, CA: Academic Press.
Tedeschi, R. G., & Calhoun, L. G. (1996). The Posttraumatic Growth
Inventory: Measuring the positive legacy of trauma. Journal of
Traumatic Stress, 9, 455-471.
Tedeschi, R. G., & Calhoun, L. G. (2004). Posttraumatic growth:
Conceptual foundations and empirical evidence. Psychological
Inquiry, 15, 1-18.
Thompson, S. C. (1985). Finding positive meaning in a stressful event
and coping. Basic and Applied Social Psychology, 6, 279-295.
Wallerstein, J. S. (1986). Women after divorce: Preliminary report
from a 10-year follow-up. American Journal of Orthopsychiatry,
56, 65-77.
Yum, Y.-O., & Schnuck-Hamlin, W. (2005). Reactions to 9/11 as a
function of terror management and perspective taking. The Journal
of Social Psychology, 145, 265-268.
Received February 6, 2006
Revision accepted January 21, 2007