ArticlePDF Available

The Selfish Hero: A Study of the Individual Benefits of Self-Sacrificial Prosocial Behavior


Abstract and Figures

Twenty-four same-sex, three-person groups (a confederate plus two naive participants) completed a "group decision-making study" in which the success of the group depended upon the willingness of one of its members (the confederate) to endure pain and inconvenience. The ordeal that the altruistic confederate endured was judged to be more difficult and costly than the experience of other group members, and the altruists were ultimately awarded more money and accorded higher status. In a second study, 334 undergraduates read a description of the procedures used in Study 1 and made judgments and monetary allocations to the hypothetical people described in the scenario. The concordance of the data in the two studies support a costly signaling, rather than a reciprocal altruism explanation for such "heroic" behavior.
Content may be subject to copyright.
Dear Author: Please Do Not Post This Article on the Web!*
To maintain the integrity of peer-reviewed and editorially approved
publications in Psychological Reports, Ammons Scientic, Ltd. retains
copyright to this article and all accompanying intellectual property rights.
Ammons Scientic, Ltd. provides this copy for the author’s educational
use and research, dened as noncommercial use by the individual author,
and specically includes research and teaching at the author’s educational
institution, as well as personal educational development and sharing of
the article with the author’s close colleagues. Any other use, including, but
not limited to, reproduction and distribution through paper or electronic
copies, posting on any websites, or selling or licensing additional copies
is prohibited. This article cannot be used for any commercial purpose
whatsoever. Terms of use are available on the Ammons Scientic website.
*A code has been embedded in this pdf to allow the publisher to nd copies and remind
posters about the terms of use.
Ammons Scientic
DOI 10.2466/ ISSN 0033-2941
Psychological Reports: Mental & Physical Health
2012, 111, 1, 27-43. © Psychological Reports 2012
Knox College
University of Texas at Austin
Summary.—Twenty-four same-sex, three-person groups (a confederate plus
two naïve participants) completed a “group decision-making study” in which the
success of the group depended upon the willingness of one of its members (the
confederate) to endure pain and inconvenience. The ordeal that the altruistic con-
federate endured was judged to be more dicult and costly than the experience of
other group members, and the altruists were ultimately awarded more money and
accorded higher status. In a second study, 334 undergraduates read a description
of the procedures used in Study 1 and made judgments and monetary allocations
to the hypothetical people described in the scenario. The concordance of the data in
the two studies support a costly signaling, rather than a reciprocal altruism expla-
nation for such “heroic” behavior.
Altruism has always been a thorny issue for evolutionary theorists;
an organism engaging in behavior that comes at a great personal cost and
seems to solely benet other individuals appears dicult for natural se-
lection to explain (McAndrew, 2003). It was not until the introduction of
the concept of inclusive tness, also known as kin selection, by Hamilton
(1964) that evolutionists had a satisfactory theoretical framework for dis-
cussing altruism. The concept of kin selection, however, cannot account
for the many altruistic acts performed for individuals who are not genetic
kin. An additional form of altruism, reciprocal altruism (Trivers, 1971), ex-
plains why these important and socially necessary behaviors occur so fre-
quently. Reciprocal altruism is when one organism provides a benet to
another organism at a cost to itself because it has received, or is likely to
receive, a similar benet in return from the other organism. Since success
at reciprocal altruism depends greatly upon the ability to quickly distin-
guish those who are cooperators from those who are freeloaders, it is not
surprising that people are quite skillful at identifying cheaters (Cosmides
& Tooby, 1992, 2008) and that people are hesitant to enter into interperson-
al relationships with other individuals who are known to be highly ma-
1Address correspondence to Francis T. McAndrew, Department of Psychology, Knox College,
Galesburg, IL 61401-4999 or e-mail (
2Carin Perilloux is now at Union College. This research was supported by research grants
from the Howard Hughes Medical Institute and from Knox College. The authors wish to
thank the brave confederates in this study, Tim Rairdon and Linda Kelahan, and to also
thank Nicole Morgan for her help in conducting the pilot studies that led to this experiment.
3Portions of this study were presented at the 21st Annual Meeting of the Human Behavior
and Evolution Society, California State University at Fullerton, HBES 2009: The selsh hero?
A study of the individual benets of self-sacricial behavior in small groups.
F. T. McAndrew & C. Perilloux
nipulative (Wilson, Near, & Miller, 1998). People also seem to be primed
to quickly recognize true altruists who will be trustworthy partners in so-
cial exchange (Brown & Moore, 2000; Fetchenhauer, Groothuis, & Pradel,
2010; Schug, Matsumoto, Horita, Yamagishi, & Bonnet, 2010).
None of the aforementioned models of altruism, however, explain
large philanthropic gifts, heroic self-sacricial behavior, or handouts to
beggars that will never be directly reciprocated. Costly signaling theory
(Zahavi, 1977; McAndrew, 2002) attempts to deal with these types of al-
truistic behavior by proposing that such behaviors are a vehicle for indi-
viduals to advertise desirable personal qualities or resources. This may
ultimately benet the altruist by increasing the likelihood that he or she
will be chosen as a mate or an ally and may also be a way of positioning
oneself for greater access to resources through direct or indirect reciproca-
tion during unforeseen times of need (Zahavi, 1977; Grafen, 1990; Boone,
1998; Roberts, 1998; McAndrew, 2002; Bliege Bird & Smith, 2005; Nowak
& Sigmund, 2005). When the altruistic act is performed primarily for the
purpose of advertising just how altruistic one is, it is referred to as com-
petitive altruism because the signaler is eectively competing with others
who are attempting to establish an altruistic reputation in the eyes of oth-
ers. The main distinction between competitive altruism and reciprocal al-
truism is that reciprocal altruism requires that the altruist is reimbursed by
individuals who directly beneted from the original altruistic act, whereas
costly signaling and competitive altruism can lead to future rewards from
individuals who may not have directly beneted from the original act of
As compelling as this explanation for competitive altruism may seem,
there have not yet been enough studies to assess its value. Some experi-
ments have demonstrated that charitable donations and other acts of gen-
erosity are indeed more likely to take place when the behavior is easily
observed by others (Haley & Fessler, 2005; Bereczkei, Birkas, & Kerekes,
2010), and in a recent study Van Vugt and Hardy (2010) have even shown
that people will make wasteful contributions in “public goods” situations,
knowing full well that the contribution will not really make a dierence, as
long as the contribution is publicly observed. They reason that this occurs
because the contribution is primarily a self-presentation strategy designed
to increase the contributor’s status and prestige, with other outcomes of
the contribution being less consequential to the donor. Some research-
ers posit that conspicuous displays of philanthropy and benevolence can
be triggered by romantic motives, possibly as a way of advertising pro-
social personality traits that might be valued by prospective mates. For
males at least, these triggers are most eective if the benevolence takes the
form of risky heroism in which the male can display courage and strength
(Griskevicius, Tybur, Sundie, Cialdini, Miller, & Kenrick, 2007). Anthro-
pological studies in traditional hunter-gatherer societies (e.g., Gurven, Al-
len-Arave, Hill, & Hurtado, 2000; Smith & Bliege Bird, 2000; Sosis, 2000;
Smith, Bliege Bird, & Bird, 2003; Bliege Bird & Smith, 2005) provide nu-
merous examples of exaggerated displays of public generosity and food
sharing, but they cannot determine if any subsequent advantages enjoyed
by the altruists are directly a function of their costly altruism.
Only a few experiments have attempted to experimentally test the
outcomes experienced by altruists (i.e., Hardy & Van Vugt, 2006; Berecz-
kei, et al., 2010; Van Vugt & Hardy, 2010), and each found that people who
engage in costly altruistic activities do in fact achieve elevated social sta-
tus and recognition as a result of public generosity or cooperativeness.
However, each of these studies examined individuals behaving in a gen-
erous and cooperative manner when sharing nancial resources in experi-
mental economic games. Since some studies (Kelly & Dunbar, 2001; Far-
thing, 2005, 2007) hinted that we may prefer heroic risk takers as mates
and friends (especially if the heroic risk takers are males), it may be im-
portant to extend these ndings to situations that focus on an individu-
al’s willingness to take physical and emotional as well as nancial risks.
Since no experiments have yet addressed this directly, we are still unable
to say with great condence exactly what benets, if any, eventually come
to the person who places him- or herself at this type of disadvantage for
others’ benet. Although it may be a stretch to describe the behaviors of
the confederates in the current studies as “courageous” or “heroic,” this
study focused on altruistic behavior in which a person displays a willing-
ness to endure physically painful and potentially embarrassing ordeals
for the benet of the group. One of the goals was to establish a procedure
for studying heroic behavior in the laboratory and, given the limitations
inherent in this setting, to create a situation that is lifelike and engaging.
Furthermore, the goal was to explore the dynamics of heroic behavior in
small, same-sex groups with an eye toward assessing dierences between
male and female groups. Finally, these studies attempted to verify wheth-
er tangible benets do in fact accrue to individuals who endure non-nan-
cial costs for the benet of others. Specically, it was an attempt to dem-
onstrate experimentally that individuals who perform physically costly
altruistic behaviors on behalf of a group will be rewarded with higher so-
cial status and a greater share of the group’s resources.
In order to examine the outcomes associated with varying amounts
of eort and self-sacrice, Study 1 utilized a laboratory methodology in
which groups of three same-sex members engaged in a series of three
tasks: one with an essentially clerical function, one which required physi-
cal skill, and one which resulted in pain, inconvenience, and embarrass-
F. T. McAndrew & C. Perilloux
ment. This combination of roles resulted in a clear dierence among the
three group members both in the amount and costliness of the work done
on behalf of the group. In an eort to assess whether reciprocal altruism or
competitive altruism provides better explanations for eects that accrue to
the altruist, a second study was conducted in which uninvolved observ-
ers read an account of the experimental situation and rated the three roles
in terms of social status and resources they would accord to each group
Forty-eight undergraduate students (24 men, 24 women) participated
in this study. Most were enrolled in undergraduate psychology classes at
a liberal arts college in the American Midwest and received course cred-
it for participation. Additional participants were recruited through yers
posted around the campus. The design was approved by the institutional
review board of Knox College.
Participants reported to a laboratory for a study on “Group Decision
Making.” In each session, there were three same-sex individuals partici-
pating as a group. Two of these individuals were naïve participants; one of
them was a confederate of the experimenter posing as a naïve participant.
The same male confederate participated in all of the male groups, and the
same female confederate participated in all of the female groups. The con-
federates were typical of students at the college in appearance and dress,
and they were trained to behave in a consistent manner across all experi-
mental sessions.
The experiment was described as an attempt to study how people in
groups organize themselves to carry out problem-solving tasks and how
their feelings about each other inuence the performance of the group.
The participants were told that they would engage in a series of three
tasks, and that if the group successfully completed the three tasks, they
would receive $45.00 to divide amongst themselves. They were also told
that if the group failed to complete any of the tasks, each group mem-
ber would receive $3.00. Before participating in the group exercises, the
group had to determine the role that each member would play. From this
point on, these roles will be known as the “Astronaut,” the “Diver,” and
the “Pitcher.” The assignment of each person to a role was made after the
group had been fully informed about the duties required for each role. The
group was given 3 min. to discuss the division of labor, and the confeder-
ate always volunteered to be the Diver. The tasks were performed in the
same order in all groups, and no communication among the group mem-
bers between tasks was permitted.
Task one.—The Astronaut was responsible for leading the group in a
12-min. decision-making task based upon the “Lost on the Moon” exer-
cise developed by the National Aeronautics & Space Administration. In
this exercise, the participants pretended that they have crash-landed a
spacecraft on the lighted side of the moon. Their survival depended upon
reaching the mother ship about 200 miles away, and their task was to rank
in order of importance 15 salvaged items for their survival. At the conclu-
sion of this exercise, the Astronaut was given 5 min. to write an explana-
tion of the arguments in favor of the top three ranked items. While the As-
tronaut was nominally the leader in this task, it was very much a group
activity and the Astronaut’s responsibilities were essentially clerical.
Task two.—The Diver engaged in a painful cold-stressor test by im-
mersing his/her forearm in a tub of ice for 30 sec., which very quickly
produces pain. No communication was permitted between the confeder-
ate and the other participants during this time, so as “not to distract the
Diver from the pain.” At the conclusion of the 30 sec., the experimenter re-
marked that “this hurts a lot more than people think it will,” and (to the
confederate) “you will probably notice that it hurts even more after you
remove your arm from the ice and the blood rushes back into the arm.”
These manipulations were employed to emphasize the unpleasantness of
the confederate’s task to the other participants.
Task three.—The Pitcher was given 3 min. to hit a target with a ball.
The throwing distance of 3 m was marked by a piece of tape on the oor.
The Pitcher was given six balls to throw. If all of the balls were used up be-
fore the target was hit, the Pitcher had to scramble around the laboratory
while the timing clock continued to run, collect the wayward balls, and re-
turn to the throwing line before resuming the task. The interesting twist to
this task was that hitting the target punctured a large water balloon, which
then drenched the Diver who was required to sit underneath it. The ap-
paratus used was a Pitchburst, manufactured by WhirlWhims LLC. The
Pitchburst functions like a dunk tank, except that it is portable and uses
much less water. The apparatus is quite safe, as the Diver was seated and
did not come into contact with anything except water. In this experiment,
the Diver had to sit in a chair under the balloon for 3 min. or until the
Pitcher successfully hit the target, which all Pitchers in this study eventu-
ally did. The Diver always got completely drenched, and the other partici-
pants were under the impression that the Diver would have to walk home
soaking wet, no matter how cold it might be outside. It was clear that both
the Diver and the Pitcher fully understood the nature of their respective
responsibilities before they volunteered for these tasks.
F. T. McAndrew & C. Perilloux
At the conclusion of the three tasks, the participants were seated in
separate areas of the laboratory where they completed a questionnaire
about their experience. Each person rated all of the participants (including
him or herself) on seven dierent items using Likert-type scales anchored
by 1: Low and 7: High endorsement of the item. These items measured the
perceived importance of each individual’s contribution to the group, the
willingness to work with each person again in a future experiment, the
perceived diculty and costliness of each person’s tasks, the perceived
status of each individual, the legitimacy of considering each individual
as the leader of the group, and how much the participant liked each per-
son. It was acknowledged by the experimenter that some of the questions
seemed a bit silly when being answered about one’s self (e.g., how like-
able, willingness to work with the person again), but the participants were
asked to respond to all of the items. This section also included four ller
questions assessing reactions to the experiment.
When all of the participants had completed the rst part of the ques-
tionnaire, they then turned to a page where they anonymously recorded
how the $45.00 should be divided among the three of them. They were
told that they could allocate the money however they wished, with the re-
strictions that each person had to receive at least $1.00 and that allocations
had to be made in whole dollar amounts. They were told that the amount
of money received by each person would be equal to the average of the al-
locations, rounded to the nearest whole dollar, made to that person by the
three group members. It was emphasized, however, that decisions about
how the money should be allocated would be kept condential; each per-
son would know how much he or she received, but would not know ex-
actly how the other participants allocated their money. The confederate’s
allocation decisions did not enter into the calculation, although the naïve
participants did not know this. The amount of money that each of the na-
ïve participants actually received was equal to the average of the alloca-
tions made by the two naïve subjects. The confederate was paid $10.00 for
each experimental session.
After the allocation decisions were completed, the confederate and
the two participants waited outside of the laboratory while the experi-
menter calculated their payments and prepared paperwork to be signed
when the money was dispersed. During this time, participants interacted
and conrmed that the Diver had no hard feelings or ill eects from the
experience. Each individual was then brought into the laboratory singly
to be paid so that compensation was condential. It was explained that
“the Diver always gets paid last, because he/she has to answer some ad-
ditional questions and there is no reason to hold everyone else up while
this takes place.” The real reason was to insure that the naïve participants
would leave the laboratory before the Diver to protect the cover story. All
participants were given the opportunity to meet with the experimenter at
a later date for complete debrieng.
The rst step was to ascertain whether there were any important dif-
ferences between male and female groups or between Astronauts and
Pitchers, as this would dictate subsequent analyses to be performed. A
2 × 2 multivariate analysis of variance (MANOVA) was conducted with
sex (male/female) and group role (Astronaut/Pitcher) as the independent
variables. There was no signicant multivariate interaction between sex
and group role (Wilks Λ = 0.25, F3,21 = 1.26, p > .05), indicating that males
and females responded similarly to playing the roles of Astronaut or Pitch-
er. There was no signicant main eect of sex (Wilks Λ = 0.50, F3,21 = 0.86,
p > .05). Therefore, the data for men and women were combined in subse-
quent analyses. There was, however, a signicant multivariate main eect
of one’s role in the group (Wilks Λ = 0.25, F3,21 = 2.64, p < .01, η2 = 0.75). An
examination of the univariate ANOVAs revealed that this eect was driv-
en by a tendency for participants to perceive the other individual’s role as
more dicult and important to the success of the group than their own.
Specically, Pitchers perceived the contributions of Astronauts to be sig-
nicantly more important than did the Astronauts themselves (M = 6.00,
SD = 0.83 vs M = 4.88, SD = 2.11; F1,47 = 5.71, p < .02, η2 = 0.13) and they also
perceived the responsibilities of the Astronaut as being more dicult
than did Astronauts (F1,47 = 10.47, p < .002; M = 4.83, SD = 1.17 vs M = 3.58,
SD = 1.56). Conversely, Astronauts perceived the contribution of the Pitch-
er to be more important than Pitchers did (F1,47 = 17.91, p < .0001; M = 6.46,
SD = 0.88 vs M = 5.08, SD = 1.38) and thought that the task of the Pitch-
er was more dicult than did Pitchers (F1,47 = 14.82, p < .0001, η2 = 0.13;
M = 5.42, SD = 1.40 vs M = 3.92, SD = 1.53). There were no other signicant
univariate eects in this analysis. Since these dierences were not perti-
nent to the hypotheses being tested, and because they occurred similarly
for both Astronauts and Pitchers, it was concluded that their responses
could be safely combined for subsequent analyses.
The amount of money allocated to each person, the perceived impor-
tance of each individual’s contribution to group success, the perceived
diculty of each individual’s responsibilities, the legitimacy of that per-
son as a group leader, the status of each person, and the perception of
the costliness of each individual’s behavior were analyzed with single-
factor repeated measures ANOVAs and Tukey HSD Tests. A Bonferroni
correction indicated that a signicance level of .008 was a more conserva-
tive and appropriate guide to signicance in these analyses. The means
and standard deviations for each of these variables are presented in Table
F. T. McAndrew & C. Perilloux
1. The analyses revealed that Divers received signicantly more money
than Astronauts and Pitchers (F2,94 = 8.82, p < .0001, ηp2 = 0.16, HSD = 1.54),
and were thought to have made a more important contribution to the suc-
cess of the group (F2,94 = 5.74, p < .004, HSD = 0.63). The Divers’ responsi-
bilities in the experiment were perceived to be signicantly more di-
cult (F2,94 = 14.76, p < .0001, ηp2 = 0.39, HSD = 0.73) and their behaviors more
costly (F2,94 = 58.91, p < .00001, ηp2 = 0.69, HSD = 0.71) as well. There were no
signicant dierences between Pitchers and Astronauts in any of the afore-
mentioned analyses. Although there was a trend for Divers to be perceived
as having marginally more status than other group members, (F2,94 = 2.90,
ηp2 = .06, p < .058), this did not translate into the Diver being perceived as
the leader of the group. In fact, Astronauts were perceived as more legiti-
mate leaders than Divers and Pitchers, who were not signicantly dier-
ent from each other (F2,94 = 9.52, p < .0001, ηp2 = 0.17, HSD = 0.42).
There were two variables for which it did not make sense to conduct
a repeated-measures ANOVA, as the self-ratings of likeability would be
highly suspect and the “willingness to work” question made no sense
when applied to oneself. Therefore, these variables were analyzed via a
paired-group t test in which the ratings of the two individuals other than
the participant were compared. These analyses revealed that Astronauts
did not nd Divers and Pitchers to dier in likeability (t23 = .44, p > .05) and
they were equally willing to work with both again in a future experiment
(t23 = 0 , p > .05). Pitchers, however, were signicantly more willing to work
with Divers than with Astronauts in future experiments (t23 = 2.15, p < .04;
M = 6.71, SD = 0.46 vs M = 6.38, SD = 0.87), and showed a marginally sig-
nicant tendency to like Divers more than Astronauts as well (t23 = 2.0,
Means and standaRd deviations foR JudgMents Made about each gRoup Role in study 1
Variable Role
Astronaut Diver Pitcher
Money allocated 14.73a2.85 16.73b3.55 13.52a2.81
Importance of contribution 5.44a1.69 6.31b1.01 5.77a1.34
Diculty of responsibilities 4.21a1.50 5.79b1.38 4.67a1.56
Legitimacy of leadership 5.71a1.01 5.00b1.35 5.10b1.19
Status 4.96a1.17 5.42b1.03 5.06a1.02
Costliness of behavior 3.15a1.58 6.04b1.24 3.46a1.90
Note.—Means with dierent superscript letters in rows were signicantly dierent at p < .008
(Bonferroni-adjusted p = .05). Ratings of Likeability and Willingness to Work with individual
in future experiments were not included in this table, since the repeated-measures ANOVA
was not an appropriate analysis for these variables. All items were rated on a scale with an-
chors 1: Low and 7: High (4 was neutral).
p < .06; M = 6.29, SD = 0.75 vs M = 5.96, SD 0.81). However, feelings of guilt
for soaking the Diver were not the only reason that the Diver fared well,
since Astronauts (t23 = 2.56, p < .02) as well as Pitchers (t23 = 2.38, p < .03)
gave more money to Divers than to the other naïve participant.
The results conrm that engaging in self-sacricial, costly behavior
that is not merely nancial in kind can result in direct benets. The ordeal
the Divers had to endure was judged to be more dicult and costly, and
these individuals were rewarded with more money and higher status. The
Divers were not, however, perceived to be the leaders of the group; this
honor was bestowed upon the Astronauts.
Although everyone had positive reactions to the Divers, this was es-
pecially true for Pitchers. The Pitchers were directly responsible for soak-
ing the Divers, and accompanying feelings of guilt may have prompted
stronger feelings of liking toward the Divers along with a desire to com-
pensate Divers for their trouble. This need to restore equity may also be
behind the interest that Pitchers had in working with Divers in future ex-
periments. Thus, Study 1 conclusively demonstrated that self-sacricial
altruistic behavior does indeed lead to benecial consequences for the al-
truist, but leaves open the question of whether reciprocal altruism or cost-
ly signaling best explains these results. It could be argued that because of
the direct cost incurred by the Divers from the Pitchers, reciprocal altru-
ism may be as viable an explanation for these results as competitive altru-
ism. Study 2 was designed to distinguish these two explanations using a
separate sample.
A scenario study was conducted in which participants read a descrip-
tion of the procedure carried out in Study 1. The procedures that were de-
scribed were exactly the same, except that all of the experimental partici-
pants in the scenarios were naïve subjects and there was no confederate.
The participants then made judgments and allocation decisions about hy-
pothetical people playing the roles in the original experiment. Since these
were uninvolved observers who did not prot from the self-sacricial be-
havior of the Diver, any benets they bestowed on the Diver would be dif-
cult to explain via reciprocal altruism.
Undergraduate students (N = 334; 160 men, 174 women) from a large
state university in the Southwestern U.S. participated in this study to ful-
ll a portion of a course research requirement.
F. T. McAndrew & C. Perilloux
Participants signed up for the experiment online and were provided
with a direct link to the survey site. The study was described as a study of
“Person Perception in Group Problem-solving Situations.” After reading
and signing an online consent form and indicating their sex, each partici-
pant read the following set of instructions:
Thank you very much for participating in this study on person perception. In this
experiment, we are interested in the judgments that you will make about people
who participated in an actual experiment conducted at another university. We
need your input to help us make sense of the data being collected in that study. In
that experiment, the researchers were interested in nding out more about how
groups organize themselves to successfully carry out problem-solving tasks. The
personality and feelings of the people in that study were also examined, but we
need to get judgments from impartial, outside observers as well, which is where
you can be very helpful to us.
In the other experiment, three people reported to a laboratory where they
were told that they would be working together on a series of three tasks. If their
group successfully completed all three of these tasks, the group would receive
$45.00 to divide among the three of them any way they wished. The only restric-
tion was that each person in the group had to receive at least one dollar, and allo-
cations must be made in whole dollar amounts. If the tasks were not successfully
completed, each individual received three dollars for participating. After reading
about the tasks, the group members had to decide which person would play which
role in the group. Thus, people had to volunteer and/or negotiate to determine
which role he or she would play in the group. The group had no prior information
about the experiment, and they came dressed as they ordinarily would for what-
ever other activities had been going on that day.
YOUR TASK: Imagine that you have just watched the experiment that you
have been reading about. Also, assume that the group has successfully complet-
ed all of the tasks described below. Your task is to decide how to divide the mon-
ey among the three group members and also to make some judgments about the
group members. The three tasks that were completed in the experiment are de-
scribed below.
The participants then saw an exact description of each of the tasks
from Study 1. After reading about the tasks, the participants then rated
each of the hypothetical group members (i.e., the Astronaut, Diver, and
Pitcher) on the same seven items as in Study 1 using the same 7-point
scale. An additional item asked them to estimate that if there were peo-
ple of both sexes participating in the experiment, how likely they thought
men and women would be to volunteer for each role on a scale with an-
chors 1: Male more likely and 7: Female more likely. Three more items re-
quired participants to indicate how interested they would be in playing
the role of the Astronaut, the Diver, and the Pitcher using a 7-point scale.
Finally they divided $45.00 among the three roles according to the follow-
ing instructions:
Because the group successfully completed each of the three tasks, pretend that
you must now decide how much money should be allocated to each member
of the group. Please remember that you have $45.00 to divide among the three
group members and that it must be divided in whole dollar values. Each person
in the group must receive at least one dollar, and you MAY NOT simply divide
the money equally among them; each person must receive a dierent amount of
money. When you have nished making your allocations, please make sure that
your numbers add up to 45.
The data were analyzed with 2 (sex) × 3 (experimental role) ANOVA
with experimental role as a repeated measures factor. All results signi-
cant at the p < .05 level are reported, but a Bonferroni correction indicated
that an alpha level of .005 would be a safer, more conservative standard. A
Greenhouse–Geisser adjustment was used in the repeated measures anal-
yses to guard against violations of the sphericity assumption.
The results for the seven variables that reected judgments made
about each target person and the amount of money allocated to them by
the participants indicated a consistent main eect for the sex of the par-
ticipants on many of the variables, in that females gave higher ratings in
general to target persons in the scenarios. This was true for judgments
of the importance of the contributions made by individuals (F1,332 = 3.83,
p < .05, ηp2 = 0.01), how challenging the tasks were (F1,334 = 4.98, p < .03,
ηp2 = 0.02), the legitimacy of calling someone a group leader (F1,329 = 4.73,
p < .03, ηp2 = 0.01), perceived status within the group (F1,332 = 5.02, p < .03,
ηp2 = 0.02), and the likeability of the target persons (F1,315 = 6.59, p < .01,
ηp2 = 0.02). There were no signicant main eects of sex (p > .05) on the
amount of money allocated to each role, willingness to work with target
persons in future experiments, or judgments of the costliness of behaviors.
This general “leniency bias” on the part of female participants (or strict-
ness bias on the part of males) was not relevant to the hypotheses of this
study. For this reason, and because there were no signicant interactions
between the sex of the participant and the repeated measures variable,
this general sex dierence will not be discussed further.
The repeated measures ANOVAs revealed signicant main eects for
the role played by the target person on all of these variables. Tukey HSD
tests were used for post hoc analyses. The means and standard deviations
for each of these variables are displayed in Table 2, and the results of the
repeated measure ANOVAs are displayed in Table 3. As can be seen in
these tables, Divers received signicantly more money, were thought to be
more likeable, and their task was judged as more dicult than Astronauts,
who in turn scored signicantly higher on each of these variables than did
Pitchers. Participants were also signicantly more willing to work with
Divers in a hypothetical future experiment than with Astronauts, who in
F. T. McAndrew & C. Perilloux
turn were preferred over Pitchers as future experimental partners. The
behavior of Divers was perceived to be signicantly more costly than the
behavior of Astronauts or Pitchers (who did not dier from each other),
and they were perceived to have made a greater contribution to the suc-
cess of the group than the Astronauts or Pitchers. When it came to being
perceived as having status and being chosen as the leader of the group,
however, it was the Astronauts who scored highest, signicantly ahead of
Divers who in turn scored signicantly higher than Pitchers on these two
In addition to these items, participants guessed whether men or
women would be more likely to volunteer for each role and expressed
their own interest in volunteering for each of the three roles. A 2 (sex) × 3
(experimental role) ANOVA was run for each of these questions with ex-
perimental role as a repeated measures variable. A Greenhouse–Geisser
correction was applied and Tukey HSD Tests were used to identify where
signicant dierences occurred.
There was no signicant main eect of rater sex when estimating
whether men or women would be more likely to volunteer for dierent
roles (F1,320 = .41, p > .05). There was, however, a signicant main eect for
the role played in the experiment (F2,568 = 63.66, p < .0001, ηp2 = 0.16). Spe-
cically, it was perceived that men would be most likely to volunteer to
be Divers and that women would be most likely to volunteer to be Astro-
nauts, with Pitchers judged to be signicantly dierent from and in be-
tween the other two roles. These main eects were qualied by a signi-
cant interaction between sex and experimental role in that men thought
Means and standaRd deviations foR JudgMents Made about
each gRoup Role in study 2 (Repeated MeasuRes)
Variable Role
Astronaut Diver Pitcher
Money allocated 14.09a5.94 20.12b6.19 10.80c4.20
Importance of contribution 4.91a1.50 5.79b1.60 4.75a1.67
Diculty of responsibilities 3.91a1.58 6.02b1.33 4.43c1.53
Legitimacy of leadership 5.30a1.61 4.42b1.58 3.43c1.45
Status 5.17a1.39 4.88b1.57 4.07c1.45
Costliness of behavior 3.47a1.82 5.77b1.30 3.66a1.72
Willingness to work with person in future 5.09a1.41 5.39b1.41 4.42c1.34
Likeability 4.55a1.37 4.91b1.36 4.07c1.35
Note.—Means with dierent superscript letters were signicantly dierent at p < .005 (Bon-
ferroni-corrected p < .05). All items were rated on a scale with anchors 1: Low and 7: High (4
was neutral).
that women would be less likely to become Astronauts and more likely
to become Pitchers and Divers than women themselves did (F2,568 = 5.23,
p < .008, ηp2 = 0.01).
When participants were asked about their own interest in volun-
teering for each of the experimental roles, a signicant main eect of
role emerged (F2,653 = 16.47, p < .0001, ηp2 = 0.05) such that participants as
a group would be most interested in being Astronauts and least interest-
ed in being Divers, with a desire to be a Pitcher falling in the middle; and
each of the three roles was rated signicantly dierently from the others. A
signicant main eect of rater sex indicated that women expressed a sig-
nicantly lower interest than men in being Pitchers or Divers (F1,334 = 37.08,
p < .0001, ηp2 = 0.10). There was also a signicant interaction between rater
sex and experimental role, due primarily to the men expressing relative-
ly little preference for any experimental role, while females expressed a
much greater preference for being an Astronaut compared to a Pitcher or
Diver (F2,653 = 9.98, p < .0001, ηp2 = 0.03).
As in Study 1, the Diver was perceived more favorably than the oth-
er two roles on most traits. The Diver received signicantly more money,
was liked the most, and was perceived to have made the greatest contri-
bution to the group by engaging in the most costly and dicult behav-
ior. The Diver was clearly being rewarded for “heroic” behavior, but in
this case by individuals who were not direct beneciaries of the Diver’s
behavior. This nding does not support the reciprocal altruism explana-
tion and instead favors a costly signaling explanation for the treatment the
Diver received. The Astronaut was once again perceived as being the lead-
er of the group, and in this case also was perceived as having the highest
status in the group. Although the means for the ratings of the Pitcher were
all above the neutral point of 4, one thing that was quite dierent in Study
2 was the comparatively negative impression of the Pitcher, who was per-
Results of Repeated MeasuRes anova (astRonauts vs diveRs vs pitcheRs)
Variable F df*p < ES
Money allocated 160.82 1.70, 556.15 .00001 .33
Importance of contribution 46.24 1.82, 607.07 .0001 .12
Diculty of responsibilities 179.39 1.96, 656.83 .00001 .35
Legitimacy of leadership 127.47 1.71, 564.20 .00001 .28
Status 51.05 1.89, 629.62 .0001 .13
Costliness of behavior 210.75 1.97, 649.98 .00001 .39
Willingness to work with person in future 49.87 1.98, 636.71 .0001 .13
Likeability 32.17 1.94, 613.25 .0001 .09
F. T. McAndrew & C. Perilloux
ceived as being signicantly lower on likeability and status than anyone
else, and they were the least preferred partners in future experiments, and
they were least likely to be thought of as leaders.
Summary and Concluding Discussion
Study 2 replicated the most important ndings of Study 1 with two
additional results of interest. In both studies, the Diver was perceived to
have engaged in the most dicult and costly behavior, received the most
money, was liked the most, and was most frequently chosen as a future
work partner. In Study 1, the Diver also had the highest perceived status,
but in Study 2 the Astronaut was awarded the highest status. The Astro-
naut was also indicated to be the group leader in both studies. This may
be an artifact of the experimental tasks. Not only was the Astronaut rst to
perform his or her task, but was also the only person nominally responsi-
ble for directing the behavior of others, so in hindsight it seems only natu-
ral that this individual would be likely perceived as the leader. It is inter-
esting to note that being a leader did not necessarily translate into greater
monetary reward and higher status.
Although there was strong concordance in the results of the two stud-
ies, the fact that the participants were actively playing a role and had
higher hedonic involvement in Study 1 presented the opportunity for in-
teresting eects to emerge that simply could not happen in Study 2. For
example, although everyone in both studies had positive reactions to the
Diver, this was especially true for people playing the role of Pitchers in
Study 1, and these tendencies were especially pronounced for female
pitchers. Only occasionally did male Pitchers apologize to the Diver after
drenching him, but female Pitchers almost universally did so, and some-
times kept up a stream of apologies even while throwing the balls at the
target. Following the drenching of the Diver, female Pitchers usually ex-
pressed great concern about her well-being, often inquired about how far
she had to walk to get home, and on two occasions attempted to hug her
as well. This marked sex dierence among Pitchers rests on more than just
a subjective perceptual bias on the part of the experimenter; it was corrob-
orated by the data. Because the MANOVA in Study 1 for the interaction
between sex and group role was not signicant, none of the univariate
analyses were discussed. However, there were two signicant univari-
ate interactions which reected female Pitchers exhibiting extremely pos-
itive responses to the Diver, such as a much stronger interest in working
with the person in a future experiment (F1,44 = 6.06, p < .02) and a tendency
for female Pitchers to allocate more money to Divers (F1,44 = 6.13, p < .02).
When the amount of money allocated to the diver was broken down by
sex and group role, male Astronauts allocated an average of $16.67 to the
Diver, male Pitchers allocated $15.58, female Astronauts allocated $15.50,
but female Pitchers allocated an average of $19.17. The dierent reactions
of male and female Pitchers may be an interesting issue to explore in fu-
ture studies. It was very clear from Study 1 that a bond formed between
the Pitcher and Diver, and future experiments could examine the underly-
ing social psychology of this eect (e.g., misattribution of arousal, resolu-
tion of cognitive dissonance, restoration of equity).
Another interesting dierence between the two studies was the
strong negative reaction to Pitchers that was found in Study 2, something
that did not occur in Study 1. In comparison to the other roles, Pitchers
in Study 2 were judged much more harshly on interpersonally relevant
traits. It may simply be that it was easier to form an impression from the
scenarios of the Pitcher as an exploitative person who delighted in seizing
the opportunity to humiliate a fellow participant, whereas the individu-
als who participated in the actual experiment saw that this was not how
it usually played out.
To the extent that there was anything surprising in the results, it was
the general lack of self-serving perceptions and selsh behavior in Study
1. The participants reliably claimed that other people had made greater
contributions to the study group and had more dicult tasks than they
did, and they were usually quite generous in their allocation of money. In
fact, only one participant (a female Astronaut) stood out as being extraor-
dinarily greedy, allocating $30.00 to herself while allocating only $10.00 to
the Diver and $5.00 to the Pitcher. In hindsight, it is also a bit surprising
that such a strong eect was found for the allocation of money, since many
participants in Study 1 took the most cognitively simple route of splitting
the money evenly among the three group members (16 of 24 men and 12
of 24 women split the money evenly). This temptation was eliminated in
Study 2, and future studies should prohibit participants from being able
to split compensation equally.
The most important conclusion from these studies is the conrmation
that costly altruistic behavior does in fact lead to direct rewards for the in-
dividuals who engage in them and that costly signaling seems to be a bet-
ter explanation for this phenomenon than reciprocal altruism (although
indirect reciprocity cannot be entirely ruled out as a viable explanation).
Only same-sex groups were used in this experiment, but evolutionary hy-
potheses relevant to showing o (Hawkes, 1991) and male intrasexual
competition could lead to interesting predictions about what might hap-
pen in mixed-sex groups. The focus could also be shifted from what hap-
pens to heroes to the issue of who becomes a hero. Personality variables
might be used in studies of groups consisting entirely of naïve individuals
to predict which types of people might be predisposed toward self-sacri-
cial behavior. One always risks sounding grandiose when discussing the
F. T. McAndrew & C. Perilloux
real-world applicability of laboratory ndings, but the possibility that this
avenue of research could ultimately enhance understanding of phenom-
ena as disparate and extreme as suicide bombings and organ donations.
For now though, it is clear that the idea of competitive altruism provides
a useful framework for thinking about the dynamics of prosocial self-sac-
ricial behavior in groups.
beReczkei, t., biRkas, b., & keRekes, z. (2010) Altruism toward strangers in need: costly
signaling in an industrial society. Evolution and Human Behavior, 31, 95-103.
bliege biRd, R. l., & sMith, e. a. (2005) Signaling theory, strategic interaction, and
symbolic capital. Current Anthropology, 46, 221-248.
boone, J. l. (1998) The evolution of magnanimity: when is it better to give than to
receive? Human Nature, 9, 1-21.
bRown, w. M., & MooRe, c. (2000) Is prospective altruist-detection an evolved solu-
tion to the adaptive problem of subtle cheating in cooperative ventures? Support-
ive evidence using the Wason Selection Task. Evolution and Human Behavior, 21,
cosMides, l., & tooby, J. (1992) Cognitive adaptations for social exchange. In J. H. Bar-
kow, L. Cosmides, & J. Tooby, (Eds.), The adapted mind: evolutionary psychology and
the generation of culture. New York: Oxford Univer. Press. Pp. 163-228.
cosMides, l., & tooby, J. (2008) Can a general deontic logic capture the facts of human
moral reasoning? How the mind interprets social exchange rules ands detects
cheaters. In W. Sinnott-Armstrong (Ed.), Moral psychology. Vol. 1. The evolution of
morality: adaptations and innateness. Cambridge, MA: MIT Press. Pp. 53-119.
faRthing, g. w. (2005) Attitudes toward heroic and nonheroic physical risk takers as
mates and as friends. Evolution and Human Behavior, 26, 171-185.
faRthing, g. w. (2007) Neither daredevils nor wimps: attitudes toward physical risk
takers as mates. Evolutionary Psychology, 5, 754-777.
fetchenhaueR, d., gRoothuis, t., & pRadel, J. (2010) Not only states but traits: humans
can identify permanent altruistic dispositions in 20 s. Evolution and Human Behav-
ior, 31, 80-86.
gRafen, a. (1990) Biological signals as handicaps. Journal of Theoretical Biology, 144,
gRiskevicius, v., tybuR, J. M., sundie, J. M., cialdini, R. b., MilleR, g. f., & kenRick, d. t.
(2007) Blatant benevolence and conspicuous consumption. Journal of Personality
and Social Psychology, 93, 85-102.
guRven, M., allen-aRave, w., hill, k., & huRtado, M. (2000) “It’s a wonderful life”:
signaling generosity among the Ache of Paraguay. Evolution and Human Behavior,
21, 263-282.
haley, k. J., & fessleR, d. M. t. (2005) Nobody’s watching? Subtle cues aect generos-
ity in an anonymous economic game. Evolution and Human Behavior, 26, 257-270.
haMilton, w. d. (1964) The genetical theory of social behavior: I and II. Journal of
Theoretical Biology, 7, 1-32.
haRdy, c., & van vugt, M. (2006) Nice guys nish rst: the competitive altruism hy-
pothesis. Personality and Social Psychology Bulletin, 32, 1402-1413.
hawkes, k. (1991) Showing o: tests of another hypothesis about about men’s forag-
ing goals. Ethology and Sociobiology, 11, 29-54.
kelly, s., & dunbaR, R. i. M. (2001) Who dares, wins: heroism versus altruism in wom-
en’s mate choice. Human Nature, 12, 89-105.
McandRew, f. t. (2002) New evolutionary perspectives on altruism: multilevel-selec-
tion and costly-signaling theories. Current Directions in Psychological Science, 11,
McandRew, f. t. (2003) Evolution and the problem of altruism: current and his-
torical perspectives. In S. Shohov (Ed.), Advances in psychology research. Vol. 27
Hauppauge, NY: Nova Science. Pp. 277-288.
nowak, M. a., & sigMund, k. (2005) Evolution of indirect reciprocity. Nature, 437,
RobeRts, g. (1998) Competitive altruism: from reciprocity to the handicap principle.
Proceedings of the Royal Society of London B, 265, 427-431.
schug, J., MatsuMoto, d., hoRita, y., yaMagishi, t., & bonnet, k. (2010) Emotional ex-
pressivity as a signal of cooperation. Evolution and Human Behavior, 31, 87-94.
sMith, e. a., & bliege biRd, R. l. (2000) Turtle hunting and tombstone opening: public
generosity as costly signaling. Evolution and Human Behavior, 21, 245-261.
sMith, e. a., bliege biRd, R. l., & biRd, d. w. (2003) The benets of costly signaling:
Meriam turtle hunters. Behavioral Ecology, 14, 116-126.
sosis, R. (2000) Costly signaling and torch shing on Ifaluk Atoll. Evolution and Human
Behavior, 21, 223-244.
tRiveRs, R. l. (1971) The evolution of reciprocal altruism. Quarterly Review of Biology,
46, 35-57.
van vugt, M., & haRdy, c. l. (2010) Cooperation for reputation: wasteful contribu-
tions as costly signals in public goods. Group Processes and Intergroup Relations, 13,
wilson, d. s., neaR, d. c., & MilleR, R. R. (1998) Individual dierences in Machiavel-
lianism as a mix of cooperative and exploitative strategies. Evolution and Human
Behavior, 19, 203-212.
zahavi, a. (1977) Reliability in communication systems and the evolution of altruism.
In B. Stonehouse & C. M. Perrins (Eds.), Evolutionary ecology. London: Macmillan.
Pp. 253-259.
Accepted June 21, 2012.
Terrorists usually choose to attack soft targets, with low self-protection abilities and resistance strengths, such as schools, campus, public squares, railway stations, etc. Under great panic, civilians attacked tend to escape aimlessly and disorderly. Hence, colliding, pushing, and trampling will take place and lead to indirect deaths and injuries. If some heroes with the spirit of altruism, coming from civilians, stand up and fight terrorists bravely, the injuries and deaths will be greatly reduced. Agent-based modeling is built to explore the role of heroes during terrorist attacks. The particle system of three categories of agents, civilians, terrorists & heroes, is built to simulate the Peshawar School Case in 2014. Multiple action rules and mechanisms are introduced, such as swarm intelligence, information communication, self-organized behaviors, and heroic behaviors. We run each simulation repeatedly for 100 times to obtain averaged (robust) outcomes. The optimal combination of parameters, which best matches real outcomes, will be solved accordingly. It indicates that: (a) the self-organizing of crowd behaviors greatly improves the survival rate of civilians. Hence, well-planned trainings of counter-terrorists emergence responses can enhance capabilities of civilians; (b) people should learn from birds’ swarm behavior in crowd evacuations. As a swarm intelligence pathway, small groups and information sharing during crowd escape simulations can greatly improve survival rates; and (c) the society should encourage more people to be altruistic heroes who are acting prosocially. Besides, it suggests that more heroes bring safer overall outcomes for civilians and even for heroes themselves.
Full-text available
A society is composed of rational selfish people who consider their own interests (homo economicus) and altruistic individuals who, despite their own interests, are socially self-sacrificing. The emergence and development of a market or any other social institution is determined by the iterated behaviors of individuals within this heterogeneity. Altruistic punishment is a vital tool in sustaining cooperation and the institutional structures of the public goods of these iterated behaviors. This thesis study deals with the impact of in-group and inter-group competitions on cooperation working within the same time. The impact has been conducted with an experiment codified with z-tree and consistent with the multi level selection theories where the winners prize is paid by the losers when in-group competition and by the loser group when inter-group competition.
Full-text available
Full-text available
Farthing (2005) tested a prediction derived from costly-signaling theory, that women would prefer physical risk takers (brave, athletic, fit) over risk-avoiders as long-term mates. Using scenarios involving high-risk acts, the prediction was confirmed for heroic (brave, altruistic) but not for non-heroic (brave, non-altruistic) acts. Apparently, women's concerns over risks to their mates overrode any positive signal value of men's risk taking, when the acts were highly risky and had no redeeming practical value. The present studies revisited the costly-signaling hypothesis using both medium- and high-risk scenarios, and it was predicted that for non-heroic acts women would prefer risk takers over risk avoiders for medium-level risks but not for highly risky acts. The prediction was supported in two studies. In Study 1, risk takers were preferred for non-heroic medium-risk acts, but risk avoiders were preferred for high-risk acts. For heroic acts, risk takers were preferred for both high- and medium-risk acts. Study 2 crossed two act risk levels with two actor skill levels, with non-heroic risks. Risk takers were preferred for the least risky combination (medium-risk act, high-skill actor) and also for the two moderately risky combinations, but risk avoiders were preferred for the riskiest combination (high-risk act, medium-skill actor). In Study 1, participants compared high-level risk takers versus risk avoiders on several person adjectives. Both heroic and non-heroic risk takers were perceived as more brave, athletic, physically fit, impulsive, attention-seeking, and foolish, and less emotionally stable and self-controlled, compared to risk avoiders. But only heroic risk takers were perceived as more altruistic, agreeable, conscientious, and sexy than risk avoiders.
Full-text available
Models indicate that opportunities for reputation formation can play an important role in sustaining cooperation and prosocial behavior. Results from experimental economic games support this conclusion, as manipulating reputational opportunities affects prosocial behavior. Noting that some prosocial behavior remains even in anonymous noniterated games, some investigators argue that humans possess a propensity for prosociality independent of reputation management. However, decision-making processes often employ both explicit propositional knowledge and intuitive or affective judgments elicited by tacit cues. Manipulating game parameters alters explicit information employed in overt strategizing but leaves intact cues that may affect intuitive judgments relevant to reputation formation. To explore how subtle cues of observability impact prosocial behavior, we conducted five dictator games, manipulating both auditory cues of the presence of others (via the use of sound-deadening earmuffs) and visual cues (via the presentation of stylized eyespots). Although earmuffs appeared to reduce generosity, this effect was not significant. However, as predicted, eyespots substantially increased generosity, despite no differences in actual anonymity; when using a computer displaying eyespots, almost twice as many participants gave money to their partners compared with the controls. Investigations of prosocial behavior must consider both overt information about game parameters and subtle cues influencing intuitive judgments.
Full-text available
Costly signaling theory (CST) offers an explanation of generosity and collective action that contrasts sharply with explanations based on conditional reciprocity. This makes it particularly relevant to situations involving widespread unconditional provisioning of collective goods. We provide a preliminary application of CST to ethnographic data on turtle hunting and public feasting among the Meriam of Torres Strait, Australia. Turtle hunting appears to meet the key conditions specified in CST: it is (1) an honest signal of underlying abilities such as strength, risk-taking, skill, and leadership; (2) costly in ways not subject to reciprocation; (3) an effective means of broadcasting signals, since the collective good (a feast) attracts a large audience; and (4) seems to provide benefits to signalers (turtle hunters) as well as recipients (audience). We conclude with some suggestions as to the broader implications of this research, and the costly signaling paradigm in general, for understanding collective action and generosity in human social groups.
Conspicuous consumption associated with status reinforcement behavior can be explained in terms of costly signaling, or strategic handicap theory, first articulated by Zahavi and later formalized by Grafen. A theory is introduced which suggests that the evolutionary raison d'être of status reinforcement behavior lies not only in its effects on lifetime reproductive success, but in its positive effects on the probability of survival through infrequent, unpredictable demographic bottlenecks. Under some circumstances, such "wasteful" displays may take the form of displays of altruistic behavior and generosity on the part of high status individuals, in that it signals the ability to bear the short-term costs of being generous or "cooperative," while at the same time reinforcing the long-term benefits of higher status.
Why do people persistently contribute to public goods and does it matter to them if their donation makes a difference? A costly signalling perspective suggests that donors might be more concerned about their reputation than the utility of their helping act. We report data on two step-level public goods experiments. We find that in public (vs. private) conditions, contributions go up even when the public good is already provided (Experiment 1) or is unattainable (Experiment 2). Furthermore, these conspicuous donations appear to enhance the status and prestige of the donor because they signal some hidden quality. This research suggests that a public good contribution can be a self-presentation strategy and that the benefits of these contributions to society are sometimes of secondary importance.
Several hypotheses about attitudes toward risk takers, derived from costly signaling theory (CST), were tested. Male and female participants evaluated the attractiveness of risk takers compared with risk avoiders as potential mates, and as potential same-sex friends, in 21 different scenarios. Both females and males preferred heroic physical risk takers as mates, with the preference being stronger for females. Contrary to predictions, for nonheroic physical risks (such as risky sports), both males and females preferred risk avoiders over risk takers as mates. However, for same-sex friends, males significantly preferred nonheroic physical risk takers, whereas females preferred risk avoiders. It was concluded that insofar as nonheroic risk taking by males is a costly signal, the signal is directed more toward fellow males than toward females. Preferences for risk takers were positively correlated with reported self risk-taking tendencies, but the correlation was significantly higher for friends than for mates for both heroic and nonheroic physical risks.In a second study, both males and females accurately predicted the opposite sex's preferences for heroic risk takers as mates. However, males failed to predict females' preferences for nonheroic physical risk avoiders. Both males and females underestimated the opposite sex's preferences for drug risk avoiders.
Subjects who score high on Machiavellianism are often charming and attractive in short-term social interactions. It is unclear from their behavior whether they are using a deceptive strategy of exploitation or whether they are merely exceptionally capable social actors. We used a story-telling method to explore aspects of Machiavellianism that are not obvious from short-term social interactions. Subjects who scored high and low on Machiavellianism wrote stories in the first person, whose main characters (referred to as “I”) were evaluated by another set of subjects. The stories reveal the cooperative nature of low-Machs and the exploitative nature of high-Machs in particularly sharp focus. Judged by their fictional creations, high-Machs were rejected as social partners for most relationships, except when their exploitative skills could be directed against members of other groups.