Content uploaded by Jolanda Jetten
Author content
All content in this area was uploaded by Jolanda Jetten on Sep 09, 2014
Content may be subject to copyright.
http://pss.sagepub.com/
Psychological Science
http://pss.sagepub.com/content/early/2014/09/04/0956797614545886
The online version of this article can be found at:
DOI: 10.1177/0956797614545886
published online 5 September 2014Psychological Science
Brock Bastian, Jolanda Jetten and Laura J. Ferris
Pain as Social Glue: Shared Pain Increases Cooperation
Published by:
http://www.sagepublications.com
On behalf of:
Association for Psychological Science
can be found at:Psychological ScienceAdditional services and information for
http://pss.sagepub.com/cgi/alertsEmail Alerts:
http://pss.sagepub.com/subscriptionsSubscriptions:
http://www.sagepub.com/journalsReprints.navReprints:
http://www.sagepub.com/journalsPermissions.navPermissions:
What is This?
- Sep 5, 2014OnlineFirst Version of Record >>
by guest on September 7, 2014pss.sagepub.comDownloaded from by guest on September 7, 2014pss.sagepub.comDownloaded from
Psychological Science
1 –7
© The Author(s) 2014
Reprints and permissions:
sagepub.com/journalsPermissions.nav
DOI: 10.1177/0956797614545886
pss.sagepub.com
Research Report
Painful experiences are an important component of
social rituals in many cultures across the world. From
secular initiation rites to religious practices, these rituals
may include burning, scarifying, and other forms of muti-
lation (Whitehouse, 1996). Durkheim (1912/1995) argued
that painful experiences function to promote cohesion
and solidarity within groups. This accords with accounts
of soldiers bonded together by the trauma of war (Elder
& Clipp, 1989; Whitehouse, 2012) and of camaraderie
promoted by shared pain within sporting contexts
(Turner & Wainwright, 2003). No empirical evidence has
been reported, however, for the proposition that bonding
and cooperation are enhanced among people who share
painful experiences. We tested the possibility that sharing
the experience of pain with other people promotes inter-
personal bonding and cooperation, aiming to provide
evidence for a causal link.
In all three experiments reported here, each experi-
mental session was conducted with a small group of
participants. On the basis of past experience, we aimed
to collect data from more than 30 participants in each
condition. Data collection was stopped when this
requirement was met. Some variation in sample size was
caused by availability of participants, foreseeable exclu-
sions, and group sizes.
Experiment 1
We first examined whether sharing a painful experience
with other people in a small group might promote bond-
ing more than sharing a similar but nonpainful social
experience. We also measured affect and pain appraisal
to examine whether any effects could be explained by
these responses to pain.
Method
Fifty-four university students1 (39 female, 15 male; mean
age = 22.24 years) were paid $10 (Australian) to participate
545886PSSXXX10.1177/0956797614545886Bastian et al.Pain and Cooperation
research-article2014
Corresponding Author:
Brock Bastian, University of New South Wales, School of Psychology,
UNSW Sydney, New South Wales, 2052, Australia
E-mail: b.bastian@unsw.edu.au
Pain as Social Glue: Shared Pain
Increases Cooperation
Brock Bastian1, Jolanda Jetten2, and Laura J. Ferris2
1University of New South Wales and 2University of Queensland
Abstract
Even though painful experiences are employed within social rituals across the world, little is known about the social
effects of pain. We examined the possibility that painful experiences can promote cooperation within social groups. In
Experiments 1 and 2, we induced pain by asking some participants to insert their hands in ice water and to perform leg
squats. In Experiment 3, we induced pain by asking some participants to eat a hot chili pepper. Participants performed
these tasks in small groups. We found evidence for a causal link: Sharing painful experiences with other people,
compared with a no-pain control treatment, promoted trusting interpersonal relationships by increasing perceived
bonding among strangers (Experiment 1) and increased cooperation in an economic game (Experiments 2 and 3). Our
findings shed light on the social effects of pain, demonstrating that shared pain may be an important trigger for group
formation.
Keywords
pain, group cohesion, bonding, cooperation, decision making, open data, open materials
Received 4/30/14; Revision accepted 6/28/14
Psychological Science OnlineFirst, published on September 5, 2014 as doi:10.1177/0956797614545886
by guest on September 7, 2014pss.sagepub.comDownloaded from
2 Bastian et al.
and were randomly allocated to either a pain condition
(n = 27) or a no-pain condition (n = 27). Group sizes
ranged between 2 and 5, with a median of 4 (M = 3.65).
Pain was elicited through two separate performance
tasks. The first involved an adapted cold pressor task
(Walsh, Schoenfeld, Ramamurthy, & Hoffman, 1989). In
the pain condition, participants submerged their hands in
ice water (< 3 °C) for as long as possible. Participants in
the no-pain condition completed the same task with
room-temperature water (≥ 24 °C) for a fixed duration (90
s). In both conditions, participants were required to locate
metal balls in the bottom of the water vessel and to place
as many of them as possible into a small container affixed
underwater. The sorting requirement ensured that in both
conditions, participants felt there was a purpose to the
task. In the second task, participants in the pain condition
were asked to maintain an upright wall squat, with back
straight and knees bent at 90°, for as long as possible.
Participants in the no-pain condition were invited to bal-
ance on one leg for a fixed duration of 60 s and instructed
to switch legs and use balance aids to avoid any tiredness.
All groups of participants in both conditions were able to
have a similar amount of interaction.
Next, participants completed the Positive and Negative
Affect Schedule (PANAS; Watson, Clark, & Tellegen, 1988;
positive affect: α = .90; negative affect: α = .73) and the
Appraisal of Life Events Scale (ALES; Ferguson, Matthews,
& Cox, 1999). Items on both measures were rated on a
scale from 1 (very slightly or not at all) to 5 (extremely).
The ALES is a primary appraisal that includes subscales
for threat and challenge. Participants rated the degree to
which their perception of the physical tasks was described
by 12 adjectives (threat: “fearful,” “worrying,” “hostile,”
“threatening,” “frightening,” and “terrifying”; M = 1.25,
SD= 0.47, α = .87; challenge: “enjoyable,” “stimulating,”
“exciting,” “exhilarating,” “informative,” and “challeng-
ing”; M = 2.52, SD = 0.89, α = .89).
Next, participants were asked to rate seven statements
designed to measure their feeling of bonding to the other
participants: “I feel a sense of solidarity with the other
participants,” “I feel connected to the other participants,”
“I feel part of this group of participants,” “I feel a sense of
loyalty to the other participants,” “I feel I can trust the
other participants,” “I feel that the participants in this
study have a lot in common,” and “I feel like there is
unity between the participants in this study.” Ratings
were made on a 5-point scale (1 = strongly disagree, 5 =
strongly agree; M = 3.43, SD = 1.08, α = .91). Finally, par-
ticipants rated the physical pain of the tasks by respond-
ing to an item gauging intensity (“How intense was the
pain you experienced?”; 0 = not at all painful, 10 =
intensely painful) and an item gauging unpleasantness
(“How unpleasant was the pain you experienced?”;
0=not at all, 10 = the most intense bad feeling imagin-
able; Price, McGrath, Rafii, & Buckingham, 1983).
Results
Manipulation checks revealed that reported pain inten-
sity was higher in the pain condition (M = 6.07, SD =
1.99) than in the no-pain condition (M = 1.67, SD = 0.92),
t(52) = 10.41, p < .001. Reported unpleasantness was also
greater in the pain condition (M = 6.00, SD = 1.96) than
in the no-pain condition (M = 1.74, SD = 1.19), t(52) =
9.63, p = .001. There were no significant differences
between conditions in positive affect (pain condition:
M= 3.05, SD = 0.82; no-pain condition: M = 2.80, SD =
0.83), t(52) = 1.09, p = .283, or negative affect (pain con-
dition: M = 1.34, SD = 0.45; no-pain condition: M = 1.27,
SD = 0.37), t(52) = 0.60, p = .554. Compared with the
control tasks, the pain tasks were viewed as marginally
more threatening (pain tasks: M = 1.36, SD = 0.58; control
tasks: M = 1.11, SD = 0.30), t(52) = 1.97, p = .054, but not
more challenging (pain tasks: M = 2.67, SD = 0.87; con-
trol tasks: M = 2.37, SD = 0.91), t(52) = 1.22, p = .227.
We predicted that participants who shared a painful
experience, compared with those who shared a similar
but nonpainful social experience, would feel more
bonded together. A one-way analysis of variance
(ANOVA) revealed that pain had a medium-sized effect
on bonding, F(1, 52) = 4.09, p = .048, d = 0.54 (see Fig.1);
participants in the pain condition reported higher bond-
ing (M = 3.71, SD = 1.01, 95% confidence interval, or CI=
[3.33, 4.09]) than did those in the no-pain condition (M =
3.14, SD = 1.09, 95% CI = [2.73, 3.55]).
This effect of pain remained when controlling for age
(p = .048), gender (p = .052), and group size (p = .050).
None of these variables were significantly correlated
with experimental condition (ps > .136) or perceived
bonding (ps > .925). To determine whether the marginal
tendency for the pain tasks to be viewed as more threat-
ening than the control tasks mediated the effect of pain
on perceived bonding, we conducted a bootstrap analy-
sis (Preacher & Hayes, 2008) using 5,000 resamples. The
results of this analysis revealed that threat was not a
significant mediator, indirect effect = −0.11, SE = 0.09,
95% CI = [−0.34, 0.03].
Experiment 2
Our first experiment found that sharing painful experi-
ences (compared with sharing nonpainful experiences)
increased perceived bonding among strangers, providing
an important insight into how shared pain may promote
trusting interpersonal relationships. In our next experi-
ment, we examined whether these effects would extend
by guest on September 7, 2014pss.sagepub.comDownloaded from
Pain and Cooperation 3
to cooperation. We predicted that sharing a painful expe-
rience (compared with sharing a nonpainful experience)
would enhance cooperative behavior.
Method
Sixty-two university students2 (47 female, 15 male; mean
age = 21.87 years) were paid $10 (plus game winnings)
to participate and were randomly allocated to a pain con-
dition (n = 34) or a no-pain condition (n = 28). Group
sizes ranged from 2 to 6, with a median of 4 (M = 3.54).
As in Experiment 1, participants completed the pain or
control tasks, the PANAS (positive affect: α = .90; nega-
tive affect: α = .73), and the ALES (perceived challenge:
α = .87; perceived threat: α = .89). Next, the cooperation
game was introduced as an ostensibly separate compo-
nent of the experiment. Finally, as in Experiment 1, par-
ticipants rated how much pain they had experienced
during the physical tasks.
We used an economic-game paradigm to measure
cooperation (Hirshleifer, 1983). This game was played in
a group setting in which each participant could choose a
number between 1 and 7. Choosing “7” could bring the
highest payoff, but only if all other group members chose
“7” also. When group members’ choices differed, partici-
pants who chose lower numbers received higher payoffs
than those who chose higher numbers (for the full payoff
schedule, see Table 1). In essence, choosing “1” was the
least cooperative option, because it ensured that the par-
ticipant would receive a moderate payoff but minimized
economic outcomes for the group. Choosing “7” was the
most cooperative option because it maximized potential
group outcomes, but the participant’s own outcome was
at risk if another group member defected. Participants
played six trials of this game. To minimize iterative strat-
egizing, we advised participants that their final payoff
would be determined from a random trial. On each trial,
participants chose numbers simultaneously. At the end of
each trial, they were told the lowest number chosen and
their earnings for that trial. Cooperation was indexed by
averaging responses across all six trials; higher scores
indicate more cooperative behavior.
Results
The pain manipulation was successful. Participants in the
pain condition reported higher pain intensity (M = 6.09,
SD = 2.12) than did those in the no-pain condition (M =
1.36, SD = 0.78), t(58) = 11.19, p < .001. Likewise, partici-
pants in the pain condition reported greater unpleasant-
ness (M = 6.16, SD = 2.01) than did those in the no-pain
condition (M = 1.46, SD = 1.14), t(58) = 10.95, p < .001.
There were no significant differences between conditions
in positive affect (pain condition: M = 2.76, SD = 0.99;
no-pain condition: M = 2.65, SD = 0.91), t(60) = 0.45,
p = .710, or negative affect (pain condition: M = 1.21,
Experiment 1 Experiment 2 Experiment 3
2.0
2.5
3.0
3.5
4.0
4.5
5.0
2.0
2.5
3.0
3.5
4.0
4.5
5.0
Mean Bonding Rating
Mean Number Choice
Pain Condition
No-Pain Condition
Fig. 1. Results for Experiments 1, 2, and 3. Mean bonding ratings in Experiment 1 (left y-axis) and mean
number choices in Experiments 2 and 3 (right y-axis) as a function of condition. Error bars indicate 95%
confidence intervals.
by guest on September 7, 2014pss.sagepub.comDownloaded from
4 Bastian et al.
SD= 0.26; no-pain condition: M = 1.27, SD = 0.37), t(60)=
0.33, p = .741. Compared with the control tasks, the pain
tasks were not viewed as more threatening (pain tasks:
M= 1.11, SD = 0.41; control tasks: M = 1.17, SD = 0.41),
t(57) = 0.59, p = .561, but they were viewed as marginally
more challenging (pain tasks: M = 2.62, SD = 0.90; control
tasks: M = 2.15, SD = 0.93), t(52) = 1.94, p = .057.
We predicted that participants in the pain condition,
compared with those in the no-pain condition, would
engage in more cooperative behavior by selecting higher
numbers in the game. An ANOVA revealed a medium- to
large-sized effect of condition on cooperation, F(1, 60) =
7.81, p = .007, d = 0.72; participants in the pain condition
opted for higher numbers (M = 4.35, SD = 1.21, 95% CI =
[3.95, 4.75]) than did those in the no-pain condition (M =
3.58, SD = 0.91, 95% CI = [3.25, 3.92]; see Figs. 1 and 2).
Age and gender were not significantly correlated with
experimental condition or responses in the cooperation
game (age: ps < .106; gender: ps < .101), and the effect of
experimental condition on cooperation remained when
we controlled for age (p = .005) and for gender (p =
.012). Group size correlated significantly with experi-
mental condition (r = −.31, p = .016) and with responses
in the cooperation game (r = −.35, p = .005); nonetheless,
the effect of experimental condition on cooperation
remained when we controlled for group size (p = .042).
To determine whether the marginal tendency for the
pain tasks to be viewed as more challenging than the
control tasks mediated the effect of pain on cooperation,
we conducted a bootstrap analysis using 5,000 resam-
ples. The results of this analysis revealed that challenge
was not a significant mediator, indirect effect = −0.01,
SE= 0.08, 95% CI = [−0.02, 0.32].
Experiment 3
Our second experiment provided a behavioral demon-
stration that sharing painful experiences enhanced coop-
eration (compared with sharing nonpainful experiences).
One potential criticism of our pain induction is that the
physical tasks involved non-pain-related factors such as
whether participants felt they performed well on the
tasks. To better isolate the effects of pain on cooperation,
we used a different type of pain induction—consumption
of a hot chili pepper.
Method
Fifty-seven university students3 (36 female, 21 male;
mean age = 24.14 years) were paid $10 (plus game win-
nings) to participate and were randomly allocated to a
pain (n = 28) or no-pain (n = 29) condition. Group sizes
ranged from 2 to 5, with a median of 2 (M = 2.84).
Participants were recruited if they were prepared to
potentially consume a hot chili pepper.
Participants first completed a pain task or a control
task, which they were told was a consumer-preferences
task. Participants in the pain condition were given one
raw Bird’s Eye chili (which is very hot) and instructed to
eat as much as possible. Participants in the no-pain
Table 1. Payoff Schedule for Experiments 2 and 3
Number chosen
by participant
Lowest number chosen in the group
1 2 3 4 5 6 7
1 $4.20 — — — — — —
2 $3.60 $4.80 — — — — —
3 $3.00 $4.20 $5.40 — — — —
4 $2.40 $3.60 $4.80 $6.00 — — —
5 $1.80 $3.00 $4.20 $5.40 $6.60 — —
6 $1.20 $2.40 $3.60 $4.80 $6.00 $7.20 —
7 $0.60 $1.80 $3.00 $4.20 $5.40 $6.60 $7.80
0
123456
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
Trial
Mean Number Choice
Pain Condition
No-Pain Condition
Fig. 2. Results for the cooperation game in Experiment 2. Mean num-
ber choice is graphed as a function of trial number and condition.
Higher numbers indicate more cooperation. Error bars indicate 95%
confidence intervals.
by guest on September 7, 2014pss.sagepub.comDownloaded from
Pain and Cooperation 5
condition were given a hard candy (also known as a
hard-boiled sweet). All participants were given 2 min to
complete the task. Participants in the no-pain condition
were instructed to hold the candy in their mouths rather
than to chew it for the entire 2 min. Yogurt and water
were provided as necessary in the pain condition.
After the consumer preferences task, participants com-
pleted the PANAS (positive affect: α = .90; negative affect:
α = .73) and the ALES (perceived challenge: α = .87;
perceived threat: α = .89). They next played the coopera-
tion game, as in Experiment 2, and then rated the pain-
fulness of the consumer preferences task. All groups of
participants in both conditions were able to have a simi-
lar amount of interaction.
Results
Participants in the pain condition reported higher pain
intensity (M = 6.29, SD = 1.78) than did participants in the
no-pain condition (M = 1.41, SD = 1.41), t(55) = 11.50,
p< .001. Participants in the pain condition also reported
greater unpleasantness (M = 5.96, SD = 1.81) than did
those in the no-pain condition (M = 1.52, SD = 1.60),
t(55) = 9.84, p < .001. Independent-samples t tests
revealed significant differences between the conditions in
positive affect (pain condition: M = 2.94, SD = 0.90; no-
pain condition: M = 2.41, SD = 0.85), t(55) = 2.28, p=.027,
and negative affect (pain condition: M = 1.75, SD = 0.51;
no-pain condition: M = 1.25, SD = 0.37), t(55) = 4.29, p <
.001. Compared with the control task, the pain task was
viewed as more threatening (pain task: M = 2.20, SD =
0.93; control task: M = 1.26, SD = 0.55), t(53) = 4.62, p <
.001, and more challenging (pain task: M = 2.99, SD = 0.52;
control task: M = 1.97, SD = 0.82), t(53) = 5.44, p < .001.
We predicted that participants in the pain condition,
compared with those in the no-pain condition, would
engage in more cooperative behavior by selecting higher
numbers in the economic game. An ANOVA revealed a
medium-sized effect of pain on cooperation, F(1, 55) =
4.09, p = .048, d = 0.53; participants in the pain condition
chose higher numbers (M = 4.33, SD = 1.62, 95% CI =
[3.81, 4.85]) than did those in the no-pain condition (M =
3.52, SD = 1.42, 95% CI = [2.92, 4.12]; see Figs. 1 and 3).
This effect of pain remained when we controlled for
age (p = .045), gender (p = .054), and group size (p =
.050). None of these variables was significantly correlated
with experimental condition (ps > .694) or cooperation
(age and group size: ps > .414; gender: p = .068).
To determine whether significant differences in affect
and task perceptions mediated the effect of pain on
cooperation, we conducted bootstrap analyses using
5,000 resamples. The results of these analyses revealed
that neither positive affect (indirect effect = −0.01, SE =
0.14, 95% CI = [−0.21, 0.39]) nor negative affect (indirect
effect = −0.01, SE = 0.25, 95% CI = [−0.58, 0.45]) signifi-
cantly mediated the effect of pain on cooperation.
Likewise, perceptions of the task as challenging (indirect
effect = −0.01, SE = 0.31, 95% CI = [−0.22, 1.02]) or threat-
ening (indirect effect = −0.01, SE = 0.24, 95% CI = [−0.36,
0.63]) were not significant mediators of the effect of pain
on cooperation.
Because Experiments 2 and 3 focused on the same
dependent variable, we collapsed the data across the
experiments for a more powerful test of our key research
question. An ANOVA revealed a medium-sized effect of
pain on cooperation, F(1, 117) = 11.10, p = .001, d = 0.61;
participants in the pain condition selected higher num-
bers (M = 4.34, SD = 1.39), 95% CI = [4.18, 4.51] (i.e.,
cooperated more) than did those in the no-pain condi-
tion (M = 3.55, SD = 1.19), 95% CI = [3.38, 3.72].
Discussion
Across three experiments, we found support for our
hypothesis that shared pain promotes cooperation.
Experiment 1 demonstrated that sharing pain promotes
bonding among strangers. We then found evidence that
shared pain enhances cooperative behavior, using the
cold pressor task and leg squats or consumption of a hot
chili pepper to induce pain (Experiments 2 and 3). None
of these effects were explained by affective responses to
pain or by appraisal of the painful tasks as challenging or
threatening.
Our findings provide novel experimental evidence for
the role of pain in promoting cooperation. This possibil-
ity has long been suggested by social theorists (e.g.,
0123456
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
Trial
Mean Number Choice
Pain Condition
No-Pain Condition
Fig. 3. Results for the cooperation game in Experiment 3. Mean num-
ber choice is graphed as a function of trial number and condition.
Higher numbers indicate more cooperation. Error bars indicate 95%
confidence intervals.
by guest on September 7, 2014pss.sagepub.comDownloaded from
6 Bastian et al.
Durkheim, 1912/1995). We argue that pain promotes
cooperation because of its well-demonstrated capacity to
capture attention and focus awareness on the immediate
painful event (Craig, 2003, 2009; Eccleston & Crombez,
1999). Painful experiences are selected for attention over
other competing demands, which makes painful events
especially salient. Our interpretation aligns with the
accounts of Whitehouse and his colleagues (Richert,
Whitehouse, & Stewart, 2005; Whitehouse & Lanman, in
press), who argued that dysphoric rituals prompt consid-
erable reflection, which in turn generates richer repre-
sentations of the episodes and their significance. When
these experiences are shared, they not only make the
events more salient but also enhance the salience of the
other people who shared in those events. Sharing pain
therefore is an especially powerful form of shared experi-
ence (cf. Campbell, 1958; Pinel, Long, Landau, Alexander,
& Pyszczynski, 2006; Wiltermuth & Heath, 2009) that
enhances the salience of the group and promotes bond-
ing, solidarity, and, ultimately, cooperation.
Our findings make several novel contributions to the
literature. First, our studies focused on personal perfor-
mance or consumer preference in a context in which no
prior group memberships or identities were salient.
Therefore, the enhanced bonding and cooperation that
we observed emerged from the experience of pain rather
than from the experience of pain for the group (i.e.,
which would increase the symbolic value of group mem-
bership; Aronson & Mills, 1959; Olivola & Shafir, 2013).
Our research thus goes beyond work focusing on costly
behaviors and group commitment (Henrich, 2009;
Xygalatas et al., 2013) or the influence of preestablished
social identities on cooperation (Kramer & Brewer, 1984;
van Vugt & Hart, 2004). Second, in our studies, partici-
pants were exposed to functionally similar tasks (com-
mon fates) in the pain and no-pain conditions, and the
tasks varied only in how painful they were; thus our find-
ings extend beyond explanations based on common fate
(e.g., Brewer & Kramer, 1986; Campbell, 1958) and place
the burden of explanation on qualities related to pain.
Third, we focused on what happens after pain, rather
than before pain, thus going beyond work focusing on
the role of fear or anxiety related to future pain (e.g.,
Schachter, 1959). Finally, although we did not empirically
demonstrate a mechanism for the effects we observed,
our design did allow us to rule out alternative explana-
tions, showing that merely sharing painful experiences
with other people promotes cooperation.
Our findings afford new insight into the ways in
which pain interacts with human sociality. Evolved
responses to pain serve to generate social support (e.g.,
Hadjistavropoulos et al., 2011); when shared, however,
pain may promote higher-order effects such as bonding
and group formation. Pain, it seems, has the capacity to
act as social glue, building cooperation within novel
social collectives.
Author Contributions
B. Bastian and J. Jetten developed the study concept and
designed the studies. L. J. Ferris assisted with data collection. B.
Bastian and L. J. Ferris analyzed the data and prepared the draft
manuscript. J. Jetten and L. J. Ferris provided critical revisions.
All authors approved the final version of the manuscript for
submission.
Acknowledgments
We thank Glen Russell for assistance with data collection and
Harvey Whitehouse, William B. Swann, Kathleen Vohs, Tom
Denson, and Ryan McKay for helpful feedback.
Declaration of Conflicting Interests
The authors declared that they had no conflicts of interest with
respect to their authorship or the publication of this article.
Funding
This research was supported by Australian Research Council
Discovery Grant DP140103716 (awarded to B. Bastian and
J. Jetten).
Open Practices
All data and materials have been made publicly available via
Open Science Framework and can be accessed at http://osf
.io/9k3sw. The complete Open Practices Disclosure for this
article can be found at http://pss.sagepub.com/content/by/
supplemental-data. This article has received badges for Open
Data and Open Materials. More information about the Open
Practices badges can be found at https://osf.io/tvyxz/wiki/
view/ and http://pss.sagepub.com/content/25/1/3.full.
Notes
1. Of the 58 participants in the original sample, 4 were excluded:
One had insight into the study’s aims, 1 reported high levels of
pain in the control condition (> 3 SD above the mean), and 2
presumably did not experience group rapport because of their
overt annoyance at a group member’s loud sneezing.
2. Of the 69 participants in the original sample, 7 were excluded:
Five had preexisting friendships with other members of their
group, and 1 misunderstood the instructions for the game.
3. Of the 71 participants in the original sample, 14 were excluded:
Eleven had preexisting friendships with members of their group
(these participants were identified early in the experimental
session and were marked for exclusion, but they were allowed
to complete the procedure in the control condition so that they
could receive payment), 2 verbally communicated a strategy at
the beginning of the game, and 1 misunderstood the instruc-
tions for the game.
by guest on September 7, 2014pss.sagepub.comDownloaded from
Pain and Cooperation 7
References
Aronson, E., & Mills, J. (1959). The effect of severity of initia-
tion on liking for a group. Journal of Abnormal and Social
Psychology, 59, 177–181.
Brewer, M. B., & Kramer, R. M. (1986). Choice behavior in social
dilemmas: Effects of social identity, group size, and deci-
sion framing. Journal of Personality and Social Psychology,
50, 543–549.
Campbell, D. T. (1958). Common fate, similarity, and other indi-
ces of the status of aggregates of persons as social entities.
Behavioral Science, 3, 14–25.
Craig, A. D. (2003). A new view of pain as a homeostatic emo-
tion. Trends in Neurosciences, 26, 303–307.
Craig, A. D. (2009). How do you feel—now? The anterior insula
and human awareness. Nature Reviews Neuroscience, 10,
59–70.
Durkheim, E. (1995). Les formes élémentaires de la vie religieuse
[The elementary forms of religious life]. New York, NY:
Free Press. (Original work published 1912)
Eccleston, C., & Crombez, G. (1999). Pain demands attention:
A cognitive–affective model of the interruptive function of
pain. Psychological Bulletin, 125, 356–366.
Elder, G. H., & Clipp, E. C. (1989). Combat experience and
emotional health: Impairment and resilience in later life.
Journal of Personality, 57, 311–341.
Ferguson, E., Matthews, G., & Cox, T. (1999). The Appraisal
of Life Events (ALE) Scale: Reliability and validity. British
Journal of Health Psychology, 4, 97–116.
Hadjistavropoulos, T., Craig, K. D., Duck, S., Cano, A., Goubert,
L., Jackson, P. L., . . . Fitzgerald, T. D. (2011). A biopsycho-
social formulation of pain communication. Psychological
Bulletin, 137, 910–939.
Henrich, J. (2009). The evolution of costly displays, coopera-
tion and religion: Credibility enhancing displays and their
implications for cultural evolution. Evolution & Human
Behavior, 30, 244–260.
Hirshleifer, J. (1983). From weakest-link to best-shot: The volun-
tary provision of public goods. Public Choice, 41, 371–386.
Kramer, R. M., & Brewer, M. B. (1984). Effects of group identity
on resource use in a simulated commons dilemma. Journal
of Personality and Social Psychology, 46, 1044–1057.
Olivola, C. Y., & Shafir, E. (2013). The martyrdom effect: When
pain and effort increase prosocial contributions. Journal of
Behavioral Decision Making, 26, 91–105.
Pinel, E. C., Long, A. E., Landau, M. J., Alexander, K., &
Pyszczynski, T. (2006). Seeing I to I: A pathway to inter-
personal connectedness. Journal of Personality and Social
Psychology, 90, 243–257.
Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resam-
pling strategies for assessing and comparing indirect effects
in multiple mediator models. Behavior Research Methods,
40, 879–891.
Price, D. D., McGrath, P. A., Rafii, A., & Buckingham, B. (1983).
The validation of visual analogue scales as ratio scale mea-
sures for chronic and experimental pain. Pain, 17, 45–56.
Richert, R. A., Whitehouse, H., & Stewart, E. A. (2005).
Memory and analogical thinking in high-arousal rituals.
In H. Whitehouse & R. McCauley (Eds.), Mind and reli-
gion: Psychological and cognitive foundations of religiosity
(pp.127–145). Walnut Creek, CA: AltaMira Press.
Schachter, S. (1959). The psychology of affiliation: Experimental
studies of the sources of gregariousness. Stanford, CA:
Stanford University Press.
Turner, B. S., & Wainwright, S. P. (2003). Corps de ballet: The
case of the injured ballet dancer. Sociology of Health &
Illness, 25, 269–288.
van Vugt, M., & Hart, C. M. (2004). Social identity as social glue:
The origins of group loyalty. Journal of Personality and
Social Psychology, 86, 585–598.
Walsh, N. E., Schoenfeld, L., Ramamurthy, S., & Hoffman, J.
(1989). Normative model for cold pressor test. American
Journal of Physical Medicine & Rehabilitation, 68, 6–11.
Watson, D., Clark, L. A., & Tellegen, A. (1988). Development
and validation of brief measures of positive and negative
affect: The PANAS scales. Journal of Personality and Social
Psychology, 54, 1063–1070.
Whitehouse, H. (1996). Rites of terror: Emotion, metaphor and
memory in Melanesian initiation cults. Journal of the Royal
Anthropological Institute, 2, 703–715.
Whitehouse, H. (2012). Ritual, cognition, and evolution. In R.
Sun (Ed.), Grounding social sciences in cognitive sciences
(pp. 265–284). Cambridge, MA: MIT Press.
Whitehouse, H., & Lanman, J. A. (in press). The ties that bind
us: Ritual, fusion, and identification. Current Anthropology.
Wiltermuth, S. S., & Heath, C. (2009). Synchrony and coopera-
tion. Psychological Science, 20, 1–5.
Xygalatas, D., Mitkidis, P., Fischer, R., Reddish, P., Skewes, J.,
Geertz, A. W., . . . Bulbulia, J. (2013). Extreme rituals pro-
mote prosociality. Psychological Science, 24, 1602–1605.
by guest on September 7, 2014pss.sagepub.comDownloaded from