The effect of intergroup competition outcome on ingroup cooperation: insights from the male warrior hypothesis

Abstract

Introduction
The Male Warrior Hypothesis (MWH) proposes that sex-specific selective pressures have promoted male cooperation with the ingroup members to outcompete rival groups. However, intergroup conflicts do not occur in isolation and the outcomes of previous competitions may influence group cooperativeness. Since this phenomenon is not well understood, we aimed to shed light on the effect of previous competition outcome on later cooperative behavior under intergroup conflicts. Based on the MWH, we hypothesized that repeated contests between groups could enhance ingroup cooperation, regardless of the outcome of the previous contest because status is at risk, but when competition is not present, participants would move to the symmetric equilibria.

Methods
To test this hypothesis, we recruited 246 individuals organized in groups of 6 and measured cooperation using a threshold public good game over two rounds, manipulating the outcome in the first round to create groups of winners and losers.

Results
Our results show that intergroup conflict scenarios promoted cooperation in both victory and defeat conditions, whereas, in the control scenario only losers increased their cooperation.

Discussion
We argue that winners under the presence of an external threat may enhance in-group cooperation in order to assure their status; whereas, losers may be attempting to regain it.

Full text

Frontiers in Psychology 01 frontiersin.org
The eect of intergroup
competition outcome on ingroup
cooperation: insights from the
male warrior hypothesis
MontserratBelinchón
1, PabloPolo
1
*, CarlosRodriguez-Sickert
1,
OrianaFigueroa
1, NoheliaValenzuela
1, PaulaPavez
2 and
JoséAntonioMuñoz-Reyes
1
1 Laboratorio de Comportamiento Animal y Humano, Centro de Investigación en Complejidad Social,
Facultad de Gobierno, Universidad del Desarrollo, Santiago, Chile, 2 Facultad de Educación,
Universidad San Sebastián, Santiago, Chile
Introduction: The Male Warrior Hypothesis (MWH) proposes that sex-specific
selective pressures have promoted male cooperation with the ingroup members
to outcompete rival groups. However, intergroup conflicts do not occur in isolation
and the outcomes of previous competitions may influence group cooperativeness.
Since this phenomenon is not well understood, weaimed to shed light on the
eect of previous competition outcome on later cooperative behavior under
intergroup conflicts. Based on the MWH, wehypothesized that repeated contests
between groups could enhance ingroup cooperation, regardless of the outcome
of the previous contest because status is at risk, but when competition is not
present, participants would move to the symmetric equilibria.
Methods: To test this hypothesis, we recruited 246 individuals organized in
groups of 6 and measured cooperation using a threshold public good game
over two rounds, manipulating the outcome in the first round to create groups
of winners and losers.
Results: Our results show that intergroup conflict scenarios promoted
cooperation in both victory and defeat conditions, whereas, in the control
scenario only losers increased their cooperation.
Discussion: Weargue that winners under the presence of an external threat may
enhance in-group cooperation in order to assure their status; whereas, losers
may beattempting to regain it.
KEYWORDS
male warrior hypothesis, intergroup conflict, ingroup cooperation, competitive
outcome, public good game
1 Introduction
Humans are adapted to live and cooperate in social groups due to the vast benets that
group living involves, such as division of labor, acquiring and maintaining reproductive
resources, or avoiding predators (Tattersall, 2011; McDonald etal., 2012). In parallel, group
living came with competition for resources, giving rise to very dierent patterns of intra and
intergroup aggression. Focusing on the latter, competition between social groups has been
present since early hominids (Keeley, 2014; Lahr etal., 2016) to modern societies and hunter-
gatherer tribes (Chagnon, 1988) as well as in non-human primates (Wrangham and Peterson,
1996). is intergroup competition is thought to have played an important role in human
evolution eliciting an intergroup psychology that enables individuals to cooperate with the
OPEN ACCESS
EDITED BY
Anastassia Zabrodskaja,
Tallinn University, Estonia
REVIEWED BY
Sefora Nemțeanu,
Babeș-Bolyai University, Romania
Sam Whitt,
High Point University, UnitedStates
*CORRESPONDENCE
Pablo Polo
p.polo@udd.cl
RECEIVED 29 September 2023
ACCEPTED 13 May 2024
PUBLISHED 24 May 2024
CITATION
Belinchón M, Polo P, Rodriguez-Sickert C,
Figueroa O, Valenzuela N, Pavez P and
Muñoz-Reyes JA (2024) The eect of
intergroup competition outcome on ingroup
cooperation: insights from the male warrior
hypothesis.
Front. Psychol. 15:1303372.
doi: 10.3389/fpsyg.2024.1303372
COPYRIGHT
© 2024 Belinchón, Polo, Rodriguez-Sickert,
Figueroa, Valenzuela, Pavez and
Muñoz-Reyes. This is an open-access article
distributed under the terms of the Creative
Commons Attribution License (CC BY). The
use, distribution or reproduction in other
forums is permitted, provided the original
author(s) and the copyright owner(s) are
credited and that the original publication in
this journal is cited, in accordance with
accepted academic practice. No use,
distribution or reproduction is permitted
which does not comply with these terms.
TYPE Original Research
PUBLISHED 24 May 2024
DOI 10.3389/fpsyg.2024.1303372

Belinchón et al. 10.3389/fpsyg.2024.1303372
Frontiers in Psychology 02 frontiersin.org
ingroup members while, at the same time, increasing hostility towards
outsiders (Choi and Bowles, 2007; Halevy etal., 2008; Bowles, 2009;
Weisel and Böhm, 2015).
Cooperation is a fundamental aspect of human behavior, but its
manifestation can vary depending on various social factors (Bowles
and Gintis, 2011). Of particular signicance is the presence of a
competitive environment since, as wealready mentioned, there are
studies that have found that competition plays a crucial role in shaping
cooperative behavior (McDonald etal., 2012). For instance, men
exhibit more altruistic and cooperative behaviors toward members of
their own group during intergroup conicts (Stirrat and Perrett, 2012;
Muñoz-Reyes et al., 2020), while female cooperation remains
unaected by such scenarios (Van Vugt etal., 2007; Yuki and Yokota,
2009). In an attempt to explain these ndings, researchers have
proposed the “male warrior hypothesis” (Van Vugt et al., 2007).
According to this functional proposal, men have evolved psychological
mechanisms that enhance intragroup cooperation during intergroup
conicts because of the substantial benets derived from aggressive
competitions throughout human evolution (Van Vugt and Hardy,
2010; McDonald et al., 2012). is disparity between sexes can
beattributed to men’s lower minimum obligatory parental investment,
which provides them with higher potential reproductive success and
greater advantages resulting from direct competitions (Trivers, 1972;
Clutton-Brock and Parker, 1992; Betzig, 2012). In this regard,
intergroup contests represent an intrasexual competition scenario to
obtain and protect resources that are turned into reproductive
resources, such as sexual mates, territory, or social status (Cosmides
and Tooby, 1992; Van Vugt and Hardy, 2010; McDonald etal., 2012).
However, under some circumstances of social instability, these
contests may occur as a series of successive competitions instead of
isolated ones because social hierarchies are not dened (Boehm,
2012). In these cases, the outcome of previous competition may aect
the motivation and willingness to compete again (Geniole etal., 2017).
en, competition outcome is another potential factor that can impact
cooperative behavior in future competitive interactions.
Previous research regarding competition outcomes has been
based on the biosocial model of status (Mazur, 1985), which indicates
that individuals that win a conict should bemore predisposed to get
involved in future competitive interactions to defend social status but,
on the contrary, losers should adopt a submissive role in order to
prevent future status decline and physical damage (Mazur and Booth,
1998). is theory is consistent with the “winner-loser” eect, which
shows that individuals tend to increase their levels of testosterone aer
winning a contest but decrease them aer losing, which in turn shapes
status-seeking behaviors increasing or decreasing competitiveness,
respectively (Mazur et al., 1992; Archer, 2006; Aguilar etal., 2013;
Geniole etal., 2017). Nonetheless, other studies have failed to nd the
“winner-loser” eect (Gonzalez-Bono etal., 1999; Schultheiss etal.,
2005; Mehta and Josephs, 2006), suggesting that changes in
testosterone depend on a number of psychological variables that
moderate the eect of winning and losing a competition. In this sense,
Mehta and Josephs (2006) found that increases in testosterone aer
competition in losers were related to their willingness to compete
again, then, some individuals may attempt to reclaim status aer
losing a competition. Moreover, losers of close competitions and the
unpredictability of social hierarchies increase levels of testosterone
and the motivation to compete again (Zilioli etal., 2014; Zilioli and
Watson, 2014). Beyond the eect on testosterone levels, winning or
losing a contest aects dierentially the mood and satisfaction and
that may also inuence behavior in future competitions through
reappraisal of the situation as challenging or threatening (Salvador
and Costa, 2009; Leis and Lautenbach, 2020). ese theoretical
models, as well as most of the empirical data in this eld, are based on
how the outcome of an individual competition aects testosterone and
further aggressive behavior (e.g., Mazur and Lamb, 1980; Elias, 1981;
Archer, 2006; Carré etal., 2009; Zilioli and Watson, 2014). However,
there is little research on how previous outcomes of intergroup
competitions aect directly ingroup cooperation in immediately
subsequent conicts.
Social dilemmas, as the public good game, have been used
extensively to study cooperation under laboratory conditions. In turn,
a specic type of public good game, the threshold public good game,
has been used in some of the previous studies investigating ingroup
cooperation within intergroup conict (e.g., Van Vugt etal., 2007;
Stirrat and Perrett, 2012). In this type of public good game, the
common pool disappears if the sum of the contributions in the group
fails to reach a given threshold. is may represent an appropriate
context to study intergroup conicts towards monopolizable resources
(as territories or mates) since failing to reach a certain degree of
cooperation and coordination may lead to losing all the potential
gains (i.e., losing the contest). In the absence of intergroup
competition, Cadsby and Maynes (1999) proposed that Nash
equilibrium theory will predict participants’ cooperation in a repeated
public good game bound to reaching or not the threshold. Groups of
individuals without the possibility of communicating are expected to
move towards two symmetric strategies over time in this game. While
symmetric pure strategy equilibrium occurs when all participants
contribute zero, the symmetric threshold equilibrium occurs when
each participant donates just enough to reach collectively the
threshold. Hence, when the threshold is not achieved, participants
may stop cooperating and may adopt the symmetric pure strategy,
whereas when the threshold is achieved, participants are expected to
maintain the symmetric threshold equilibrium.
Under intergroup conict scenarios, the Nash equilibrium may
not beappropriate to explain cooperative behavior over time since
status is added to the payo and it acts as a strategic incentive to win.
en, winning (losing) the contest increase (decrease) social status
and the utility derived from cooperation. Two previous studies found
that, in a repeated public good game with strategic incentives to win,
deserved losers tended to increase their donations regardless of the
previous individual contribution (Tan and Bolle, 2007; Kiss et al.,
2020). is increment in cooperation is somehow contrary to the
expectations from the biosocial model of status that postulates that
losers should decrease their predisposition to compete again (Mazur
and Booth, 1998), and then, to reduce cooperation in the next round,
but instead, the increment of cooperation might represent an
opportunity for losers to regain status (Daly and Wilson, 1988; Mehta
and Josephs, 2006). Results regarding the victory condition were
ambiguous: while in one study winners decreased donations (Tan and
Bolle, 2007), in the other acted as conditional cooperators (Kiss etal.,
2020), which are not expected responses in conventional approaches
in which winners increase their competitive motivations (Mazur,
1985; Geniole etal., 2017). Consequently, due to the lack of consistency
between empirical data and within theoretical predictions, our goal is
to contribute to clarify the eect of competition outcome on male
ingroup cooperation under intergroup conicts.

Belinchón et al. 10.3389/fpsyg.2024.1303372
Frontiers in Psychology 03 frontiersin.org
Considering the postulates of the male warrior hypothesis (Van
Vugt etal., 2007), weproposed that a repeated interaction between
groups may enhance ingroup cooperation, independently of the
previous group’s history of victory or defeat, given the signicant
benets for men related to status acquisition during intergroup
conicts. Losers may be attempting to regain their status/limited
resources lost in the last interaction by increasing their ingroup
cooperation as suggested by Mehta and Josephs (2006) whereas
winners, under the presence of this external threat, may enhance
cooperation in order to ensure their status/limited resources. is
would beespecially true under circumstances of unstable or undened
group hierarchies (Zilioli and Watson, 2014) like competition between
groups that do not meet each other before. In order to test this general
hypothesis, wemeasured cooperation via the threshold public goods
game in an intergroup conict scenario and a control context to
investigate how the competitive outcome (winning or losing the
contest and reaching or not the threshold, respectively) inuenced
cooperation in the next round. We set out the following specic
predictions. First, in a intergroup conict context, wepredicted that
the high sensitivity of men to this conict would result in that both
groups of winners and losers in the rst round would increase their
contributions to maintain or reclaim status/limited resources in the
second round. On the contrary, in the control context (i.e., without the
intergroup conict scenario), wepredicted that participants in groups
that did not reach the threshold would decrease their contributions
(i.e., move to the pure symmetric strategy), but winners would tend to
maintain their contributions (i.e., symmetric threshold equilibrium)
according to the Nash equilibria of the game.
2 Methods
2.1 Participants
A total of 246 young men (age: M = 22.21 years, SD = 3.20 years)
were recruited from universities and general population in the Region
of Valparaiso (Chile) through public announcements on social
networks and in the laboratory website. Participants were organized
into groups of 6 members. Weintended to form groups in which
individuals knew each other in order to gain ecological validity.
Weexcluded four individuals because they did not complete the entire
procedure. e main reason for choosing young male adults between
18 and 39 years old is because intrasexual competition is more intense
in that period of life (Wilson and Daly, 1985). At the end of the
experiment, participants received $15,000 Chilean pesos (CLP) per
individual (around $23 USD) for showing up. In addition, they could
receive an additional payment of up to $15,000 CPL according to their
individual (and group) performance in the game. Most of the
participants (90%) received the total amount of $30,000 CLP.
2.2 Ethics committee
is experiment, including protocols and data treatment, was
approved by the Institutional Bioethics Committee of the Universidad
de Playa Ancha (Chile). Participants must have read and signed the
informed consent form prior to the experiment. In this document, all
the procedure and anonymity protocols were presented. In order to
ensure anonymity, we linked all the individuals’ responses to
alphanumeric codes.
2.3 Procedure
We had two manipulated sets of conditions. First, groups were
assigned to an intergroup competition scenario or a control scenario.
In the intergroup competition condition, participants were informed
that they were playing synchronously against a group of men from
another university in the country that was participating in the same
project, so the group that reached the threshold faster would win.
Only the winners would get the bonus. is group was ctitious to
simulate an intergroup competitive scenario with one of three
potential universities in the country that dier in their nature as public
or private institutions (two public and one private) and in the access
score. Participants were not informed which institution were
competing against. Weincluded in the potential pool of competitors
public and private institutions because regardless of the origin and
socioeconomic level of the participants, there would always bea
potential competing group with dierent characteristics. ese
institutions corresponded to the aliations of some of the project’s
co-investigators, but for logistical reasons, data was only collected at
the institution of the principal investigator (a fourth institution). In
the control condition, participants were informed that they were
playing synchronously among them. ey were informed that they
could gain a bonus if they reached the threshold as a group. No
mention was made of other groups. e second condition was whether
the group was assigned to the winner or the loser condition in the rst
round of the game. is rst round of the game was manipulated in
terms of the group performance. In the winner condition, groups were
informed that they reached a total amount of $20,000 Chilean pesos
in their common good and accordingly, they won the bonus regardless
of the actual group performance. In the context of intergroup
competition, it was added that the group exceeded the threshold
(18,000 Chilean pesos) before the rival group. In the loser condition,
groups were informed that they reached a total amount of $17,000
Chilean pesos in their common good and accordingly, they lost the
bonus regardless of the actual group performance. Individuals only
had information about their own contribution and the manipulated
contribution of the group. Accordingly, they were unable to know
their real performance and therefore doubting the credibility of the
group performance. In both conditions, in the second round, the
participants were informed about their real performance. at is,
groups that exceeded the threshold were informed that they won the
bonus. Otherwise, they were informed that did not gain the bonus.
Groups were randomly assigned to one of these conditions.
Participants rst completed a sociodemographic questionnaire in
which they responded to questions about their age, sex, sexual
orientation, socioeconomic status, and place of residence. Aer that,
they were informed about the procedure of the threshold public good
game in a meeting room. ese instructions were provided by a
researcher and the protocol of the speech is provided in
Supplementary material. e instructions were also provided inside
the game (see Supplementary material) before they played a practice
game. en, they played the rst round of the game. Aer the outcome
manipulation, they were asked to play a second round in which the
outcome was not manipulated. e experiment was conducted in the

Belinchón et al. 10.3389/fpsyg.2024.1303372
Frontiers in Psychology 04 frontiersin.org
Laboratorio de Comportamiento Animal y Humano of the University
of Playa Ancha (Chile) in six isolated experimental cabins with
computers so participants could not communicate with each other. In
addition, participants only had information about their individual
contributions to the public good. In other words, participants were not
able to know their real performance as a group in the rst round of the
game (the round in which the outcome was manipulated).
2.4 Behavioral measure
Cooperation was measured by the threshold public good game.
We employed the contribution of the individuals to assess their
cooperative behavior (Zelmer, 2003). We followed the protocol
established by Van Vugt etal. (2007) and replicated by Stirrat and
Perrett (2012) to measure contributions under the presence and
absence of intergroup conict and following winning and losing the
rst round. en, participants played two consecutive rounds of the
game. e threshold public good game was played on computers using
z-Tree soware (Fischbacher, 2007). e starting endowment was
$5,000 Chilean pesos which they could destinate any amount of that
for the group endowment. ey were told that if they exceeded a
threshold (total invested) of $18,000 Chilean pesos (which would
involve a mean cooperation of $3,001, i.e., 60% of their initial
endowment), they would receive a bonus of $11,000 Chilean pesos,
regardless of the aggregate contributions. In the case of the intergroup
conict scenario, they were told that, in the case that the rival group
exceeded the threshold, they needed to reach that objective before the
rival group to gain the bonus. On the contrary, if the group did not
exceed that threshold, participants would receive just the amount they
decided not to contribute. As wehave mentioned, the outcome of the
rst round was manipulated regardless of group performance but
participants were not aware of this manipulation. e outcome of the
second round depended on whether the group exceeded the threshold
in both conditions (intergroup competition and control) since the
rival group was ctitious.
2.5 Data analysis
To test our predictions, weemployed hierarchical linear models
(HLM) in order to account for the repeated nature of our data (Wes t
etal., 2007). Level-1 variables were those regarding within individual
measures and level-2 variables were those regarding between
individual measures. To test our rst prediction, weselected only data
for groups in the intergroup conict scenario and wetted a HLM
with contributions in the public good game (level-1 variable) as the
dependent variable. e round of the game (level-1 variable) and the
condition (losing or winning the rst game, level-1 variable) were our
independent variables. Weincluded the interaction between round
and condition. We controlled for age (level-2 variable) and
socioeconomic status (level-2 variable). Individual ID was included as
a random eect. To test our second prediction, weselected data for
groups in the control condition and tted the same model indicated
above. Wedecided to t two independent models since comparisons
between control and intergroup conict conditions were published
elsewhere (Muñoz-Reyes etal., 2020). For both models, wespecied
full maximum likelihood estimation and Type III variance. Post-hoc
tests (using Bonferroni correction) followed whenever a signicant
interaction eect was detected. Since hierarchical linear models entail
residuals at dierent levels, wecalculated for each signicant result the
eect size following the following expression:
ƒ22
2
1
2
2
2
1
=−
−
RR
R
where
R2
2
represents the variance explained for a full model and
R
1
2
the variance explained for a model in which a given eect was
removed (Lorah, 2018). In order to calculate
R2
of the models,
weemployed the following expression:
RFF
EE
2
22
22
1=− +
+
στ
στ
where
σ
F
2
is level-1 error variance of the full model,
τ
F
2
is level-2
error variance of the full model,
σ
E
2
is level-1 error variance of the
empty or null model and
τ
E
2
is level-2 error variance of the empty or
null model (Lorah, 2018). e global signicance level was set at
α = 0.05. Models were performed with IBM SPSS 25 soware.
3 Results
Table1 shows mean contributions (and standard deviations) in
the threshold public good game in each round according to the
competition outcome condition for both intergroup conict and
control conditions.
Regarding our rst prediction in which weexpected an increase
in cooperation in both winners and loser when facing a intergroup
conict condition, the results (Table2) show that there was no main
eect of competition outcome on cooperation, (F(1, 126) = 0.84;
p = 0.360) but there was a main eect of round (F(1, 126) = 7.19;
p = 0.008; ƒ
2
= 0.008). Overall, contributions were higher in the second
round (estimated mean = 3892.54, SE = 93.72) than in the previous one
(estimated mean = 3638.90, SE = 93.72). e interaction eect between
round and competition outcome was not signicant, (F(1, 126) = 0.58,
p = 0.447), that is, both winners and losers increased their
contributions and no dierence was found between them in any
round (Figure1A). Finally neither age (F(1, 126) = 2.36, p = 0.127) nor
socioeconomic status (F(1, 126) = 3.42, p = 0.067) were signicant in
the model. Since only round was a signicant predictor, alternatively
wecan test its eect with a simpler t-test of repeated measures analysis.
In this case, the dierences remained signicant (t = −2.631, df = 125,
p = 0.010, Cohen’s dz = 0.234).
Regarding our second prediction in which weexpected the winners
to keep cooperation unchanged, while the losers to decrease it in a
control condition, the results (Table3) show that there was no main
eect of competition outcome on cooperation, (F(1, 120) = 0.06,
p = 0.939), but there was a main eect of round, (F(1, 120) = 10.91,
p = 0.001; ƒ
2
= 0.031). Overall, cooperation is higher in the second round
(mean = 2561.63, SE = 93.48) than in the rst one (mean = 3313.23,
SE = 93.47). However, interaction eect between round and competition
outcome was signicant, (F(1, 120) = 14.412, p < 0.010; ƒ
2
= 0.019).
Pairwise comparisons showed that the mean contribution was

Belinchón et al. 10.3389/fpsyg.2024.1303372
Frontiers in Psychology 05 frontiersin.org
signicantly dierent in losers between the rst and second rounds
(mean dierences = −0.498, df = 120, p < 0.001) but winners’
contributions did not signicantly dier between rounds (mean
dierences = 0.035, df = 120, p = 0.740) (Figure 1B). In addition,
wefound that losers and winners did not dier in their contribution
either in the rst round (mean dierences = −0.254, df = 164.23,
p = 0.149) or in the second round (mean dierences = 0.279, df = 164.23,
p = 0.114). Finally, neither age (F(1, 120) = 1.67, p = 0.199) nor status
socioeconomic (F(1, 120) = 0.09, p = 0.923) were signicant in the model.
4 Discussion
In this study, we aimed to investigate the role of intergroup
competition outcomes on intragroup cooperation. Concretely,
wetested for dierences in contributions between winners and losers
in two consecutive rounds in a threshold public good game in two
contexts: competing against a rival group and in the absence of this
competition. Wefound support for our rst prediction as, under a
competitive scenario, cooperation is heightened in the second round
TABLE1 Mean contributions and standard deviations (SD) in the threshold public good game according to the round (1st Round vs. 2nd Round), the
competition outcome (Defeat vs. Victory) and condition (Intergroup conflict vs. Control).
Control condition Intergroup conflict condition
Defeat (N= 66) Victory (N= 54) Defeat (N= 60) Victory (N= 66)
Mean SD Mean SD Mean SD Mean SD
1st Round 3,185 1,169 3,440 951 3,515 1,236 3,761 948
2nd Round 3,719 1,032 3,403 932 3,760 1,220 3,898 881
TABLE2 Estimated parameters for contributions in the threshold public good game in the intergroup competitive scenario (N = 126).
Fixed eect B t p-value
Intercept 5517.53 6.797 < 0.001
Condition = 0 −106.20 −0.563 0.574
Round = 0 −136.38 −1.390 0.167
Age −41.59 −1.536 0.127
SES −232.32 −1.850 0.067
Condition = 0 * Round = 0 108.50 −0.763 0.447
Covariance parameter Estimate SE ICC
Residual 317862.02 40046.85 0.712
Intercept (ID) 786293.31 120758.72
Condition [0 = Loser]; Round [0 = First]; SES, Socioeconomic Status.
FIGURE1
Estimated mean contribution ± standard error of the mean (SEM) in the threshold public good game in the first and the second round for the victory
and defeat condition in the intergroup conflict (A) and the control context (B). Orange line depicts victory condition and blue line depicts defeat
condition.

Belinchón et al. 10.3389/fpsyg.2024.1303372
Frontiers in Psychology 06 frontiersin.org
of the game independently of the previous competition outcome. at
is, both losers and winners increased their cooperation in the second
round when competition with another group was mentioned.
However, weonly found partial support for our second prediction. As
expected, winners did not change their contributions in the second
round in the absence of intergroup competition, but against our
prediction, losers also contributed signicantly more in the second
round. ese results provide some hints about the strategic use of
cooperation under intergroup competitive scenario as predicts the
male warrior hypothesis.
e male warrior hypothesis (Van Vugt and Hardy, 2010;
McDonald etal., 2012) argues that intergroup conicts represent an
opportunity for men to acquire or defend status. In line with this
hypothesis, we proposed in our rst prediction that, aer a
competition, losers may beattempting to regain and acquire their
status lost in the last interaction, and winners to defend it. erefore,
wepredicted that the competition outcome would not aect ingroup
cooperation in a subsequent contest because increasing cooperation
has potential benets for both winners and losers regarding to social
status. Our results indicate that male groups increased cooperation in
a subsequent competition independently of the outcome of the
previous contest supporting our prediction. On the one hand, these
results are partially in accordance with studies under similar
methodological conditions, that is, using the threshold public good
game under an intergroup contest. ese studies found that losers
increased contributions when their group contributed less than the
rival group (deserving losers) (Tan and Bolle, 2007; Kiss etal., 2020).
On the other hand, for the victory condition, results dier between
studies: while in one study winners acted as conditional cooperators
(Kiss etal., 2020) in the other winners decreased contributions with
and without monetary incentives (Tan and Bolle, 2007). Both results
contrast with the increase in cooperation that wefound. A potential
reason that may explain the dierences between our results and the
two mentioned studies is that, in the previous ones, groups were
formed by men and women and according to the male warrior
hypothesis (Van Vugt and Hardy, 2010; McDonald etal., 2012), group
performance could bebiased because females are not aected by
intergroup conicts as males are. In addition, in the study of Kiss etal.
(2020) the probability of being chosen as the winner was proportional
to the performance of the group relative to the other group so the
winning group was not always the one that contributed the most. is
adds an element of chance that our design did not contemplate and
that could beaecting the logic of the competition for a monopolizable
resource. And nally, in our study participants played only two rounds
whereas in the mentioned studies played 10 and 20 rounds. e
behavioral response to win or lose may bedierent in the rst rounds
compared to the last ones in a sequence of 10 or 20. In fact, in the
study of Tan and Bolle (2007) participants playing with partners and
in the intergroup competition condition with incentives seem to
increase their contributions regardless of the previous results in the
second and third rounds. is resembles our results, unfortunately,
this study only reports results from all rounds averaged.
In addition, our results are in contrast to predictions derived from
the biosocial model of status (Mazur, 1985), which argues that winners
may beinvolved in further competitions, but losers would tend to
withdraw in order to avoid physical aggression and status decline. is
theory is in line with the challenge hypothesis (Wingeld etal., 1990;
Archer, 2006) which proposes that, under situations of status threat,
there are physiological responses associated with testosterone levels
that drive dominance-related behaviors producing the “winner-loser”
eect: winners increase testosterone levels to reinforce dominant
behaviors and losers decrease them (Mazur, 1985). en, under this
model, wemight nd that winners increase cooperation but losers
decrease it because of testosterone levels decline. However, in our
study participants played the second round following the rst round
without any delay, therefore, changes in the cooperation levels cannot
beexplained by changes in circulating testosterone if weconsider that
the hormonal eects of winning or losing a competition are delayed
15–20 min (Casto and Edwards, 2016). e increased cooperation
among losers can beexplained by other psychological factors, such as
individual attributions or mood, that also modulate the “winner-loser”
eect (Gonzalez-Bono etal., 1999). For example, Kiss etal. (2020)
observed that chance losers acted as conditional cooperators but
deserved losers increased cooperation, which reects that attributions
related to the outcome may modulate ingroup cooperation. Moreover,
circumstances of social instability may lead to a reverse “winner-loser”
eect (Geniole et al., 2017), which may promote status-seeking
behavior in losers who would beinvolved in future competitions to
reclaim status (Daly and Wilson, 1988; Mehta and Josephs, 2006).
en, when social hierarchies are not dened, as when groups that
have not interacted before, like in our study design, wecan expect a
dierent tendency for losers (Zilioli and Watson, 2014).
TABLE3 Estimated parameters for contributions in the threshold public good game in the control scenario (N = 120).
Fixed eect B t p-value
Intercept 4186.52 5.665 < 0.001
Condition = 0 298.62 1.590 0.114
Round = 0 37.06 0.332 0.740
Age −36.74 −1.292 0.199
SES 11.46 0.096 0.923
Condition = 0 * Round = 0 −570.90 −3.796 < 0.001
Covariance parameter Estimate SE ICC
Residual 335841.78 43356.99 0.676
Intercept (ID) 702075.80 114389.08
Condition [0 = Loser]; Round [0 = First]; SES = Socioeconomic Status.

Belinchón et al. 10.3389/fpsyg.2024.1303372
Frontiers in Psychology 07 frontiersin.org
Regarding our second prediction, we proposed that the Nash
equilibrium would explain participant’s behavior in consecutive games
in a control condition without intergroup competition (Cadsby and
Maynes, 1999). In this sense, participants that do not overcome the
threshold will move to the symmetric pure equilibrium, but winners will
tend toward the symmetric threshold equilibrium. Our results show that
participants in the victory condition maintained contributions to
overcome the threshold using the previous successful strategy,
suggesting that they moved to the symmetric threshold equilibrium.
However, participants in the defeat condition tended to behave
oppositely as expected by the symmetric pure strategy: they increased
contributions. is nding could beexplained partly considering the
utility and the payos from the game because when participants
overcome the threshold, they receive a monetary incentive that
promotes cooperation. Furthermore, social identity is known to
decrease “free-riding” in social dilemmas as participants try to maximize
ingroup outcomes (Simpson, 2006). en, a collective goal—as the
threshold in this case—may enhance intrinsic motivation to cooperate
and succeed in the group objective even without monetary incentives.
e eect of intergroup conict on cooperation has taken
attention recently because there is robust evidence that cooperation is
exacerbated in groups of men when competing against same-sex rival
groups (McDonald etal., 2012; Stirrat and Perrett, 2012; Muñoz-Reyes
etal., 2020). Nonetheless, intergroup conicts are more complex than
have been represented in experimental settings. Firstly, in some cases,
competitions do not occur isolated as most of the time humans
become revenge-seekers (Boehm, 2012). Moreover, competition
outcome may inuence and regulate social hierarchies aecting status
and, therefore, further status-seeking behaviors and social interactions
(Sidanius and Pratto, 2001; Geniole et al., 2017). In this sense,
considering previous competition outcomes will help us to understand
more precisely the role of intergroup conict scenarios on cooperation.
However, our results can only be extended to the mentioned
competitive settings. For instance, face-to-face contests could lead to
dierent predictions as physical threat could be present and
participants have the opportunity to evaluate their rivals’ features
dening social hierarchies, which is known to play a key role in
modulating competitive behavior (Flinn etal., 2012). en, under this
scenario, there are some possibilities that losers would decrease their
competitive behavior (Mazur, 1985). In addition, in our study, the
contest was compounded by two games. However, there is evidence
indicating that cooperation declines when the game is played in
repeated interactions (e.g., Burton-Chellew and West, 2021). It would
berelevant to investigate whether the eect found in this study is
sustained across rounds and, therefore, if the intergroup conict
scenario is a key element in sustaining cooperation. It might bealso
possible that further rounds would dene group hierarchies so that
losers would admit their defeat and winners would not perceive the
contest as a challenge, therefore, adopting dierent competitive
behaviors. In addition, previous results showed that cooperation in the
public goods game also depends on males’ sexually selected traits
(Muñoz-Reyes etal., 2020). en, it would bevaluable also to test in
future studies if individual dierences may modulate the inuence of
competition outcome on cooperation. In this study the interaction
with the other group was simulated and accordingly, participants
lacked a relative measure of their performance compared to the rival’s
group performance. is information may berelevant since individuals
may change their competitive strategy from more aggressive to more
avoidance-oriented according to the formidability (the ability to inict
costs) of the rival group (McDonald etal., 2012). In this sense, if the
performance of the rival group is a proxy of their formidability, this
information may aect the degree of in-group cooperation from the
participants. Future studies in which this information is provided
either because there is a real interaction between the groups, or
because although it is simulated also simulates dierent degrees of
performance of the other group, would beinteresting to understand
this problem more deeply. Another limitation is that our study only
considered groups of men thus limiting the interpretation of our
results to the specic context of intergroup competition between them.
It would berelevant in future studies to include groups of women and
mixed-sex groups and consider dierent combinations in the
composition of competing groups to gain a deeper understanding of
the dynamics of cooperation in intergroup competition contexts. is
is because it has been found that both the sex composition of the group
and that of the rival group are relevant to the emergence of intergroup
discrimination in cooperation and outgroup bias in both men and
women (Navarrete etal., 2010; Balliet etal., 2014). Furthermore, recent
evidence suggests that group composition is more relevant than sex in
revealing dierences in cooperation when comparing an intergroup
conict and a control scenario (Muñoz-Reyes et al., 2023). Finally,
individuals were not randomly assigned to groups since weaimed to
form groups of individuals who knew each other beforehand to gain
ecological validity. However, this may introduce potential confounding
biases that limit the scope of our results in explaining the underlying
mechanisms associated with the male warrior hypothesis.
In conclusion, this is the rst attempt to understand how
competition outcome aects male ingroup cooperation to outcompete
a rival group within the male warrior hypothesis framework. Wefound
that both winners and losers increased cooperation in the second round
of a threshold public good game compared with the rst round when
competition with another group is present. ese results suggest that
under an intergroup conict scenario, both winners and losers have
incentives to increase cooperation supporting the main postulate of the
male warrior hypothesis. Wepropose that the incentives to cooperate
might bedriven by a mechanism related to resource monopolization
and status-seeking behavior: winners would defend status and losers
would try to regain it. is study supports that intergroup conicts
inuence male behavior similarly in winners and losers and suggest
that an intergroup conict scenario may enhance ingroup cooperation.
Data availability statement
e datasets presented in this study can be found in online
repositories. e names of the repository/repositories and accession
number(s) can befound below: https://osf.io/4p69d/?view_only=811
e7d5d2e4147dd92ac1d95e7ed8c.
Ethics statement
e studies involving humans were approved by Institutional
Bioethics Committee of the Universidad de Playa Ancha (Chile). e
studies were conducted in accordance with the local legislation and
institutional requirements. e participants provided their written
informed consent to participate in this study.

Belinchón et al. 10.3389/fpsyg.2024.1303372
Frontiers in Psychology 08 frontiersin.org
Author contributions
MB: Formal analysis, Investigation, Writing – original dra,
Writing – review & editing. PPo: Conceptualization, Data curation,
Formal analysis, Investigation, Writing – original dra, Writing –
review & editing. CR-S: Writing – review & editing. OF: Investigation,
Writing – review & editing. NV: Investigation, Writing – review &
editing. PPa: Investigation, Writing – review & editing. JM-R:
Conceptualization, Funding acquisition, Investigation, Methodology,
Project administration, Writing – review & editing.
Funding
e author(s) declare nancial support was received for the
research, authorship, and/or publication of this article. Research
was funded by a FONDECYT Regular project (#1170513) from the
Agencia Nacional de Investigación y Desarrollo, Government of
Chile and by a National Doctoral Scholarship (21211574) from the
Agencia Nacional de Investigación y Desarrollo, Government
of Chile.
Conflict of interest
e authors declare that the research was conducted in the
absence of any commercial or nancial relationships that could
beconstrued as a potential conict of interest.
Publisher's note
All claims expressed in this article are solely those of the authors
and do not necessarily represent those of their aliated organizations,
or those of the publisher, the editors and the reviewers. Any product
that may beevaluated in this article, or claim that may bemade by its
manufacturer, is not guaranteed or endorsed by the publisher.
Supplementary material
e Supplementary material for this article can befound online
at: https://www.frontiersin.org/articles/10.3389/fpsyg.2024.1303372/
full#supplementary-material
References
Aguilar, R., Jiménez, M., and Alvero-Cruz, J. R. (2013). Testosterone, cortisol and
anxiety in elite eld hockey players. Physiol. Behav. 119, 38–42. doi: 10.1016/j.
physbeh.2013.05.043
Archer, J. (2006). Testosterone and human aggression: an evaluation of the
challenge hypothesis. Neurosci. Biobehav. Rev. 30, 319–345. doi: 10.1016/j.
neubiorev.2004.12.007
Balliet, D., Wu, J., and De Dreu, C. K. W. (2014). Ingroup favoritism in cooperation:
a meta-analysis. Psychol. Bull. 140, 1556–1581. doi: 10.1037/a0037737
Betzig, L. (2012). Means, variances, and ranges in reproductive success:
comparative evidence. Evol. Hum. Behav. 33, 309–317. doi: 10.1016/j.
evolhumbehav.2011.10.008
Boehm, C. (2012). Ancestral hierarchy and conict. Science 336, 844–847. doi:
10.1126/science.1219961
Bowles, S. (2009). Did warfare among ancestral hunter-gatherers aect the evolution
of human social behaviors? Science 324, 1293–1298. doi: 10.1126/science.1168112
Bowles, S., and Gintis, H. (2011). A cooperative species: Human reciprocity and its
evolution. Princeton: Princeton University Press.
Burton-Chellew, M. N., and West, S. A. (2021). Payo-based learning best explains
the rate of decline in cooperation across 237 public-goods games. Nat. Hum. Behav. 5,
1330–1338. doi: 10.1038/s41562-021-01107-7
Cadsby, C. B., and Maynes, E. (1999). Voluntary provision of threshold public goods
with continuous contributions: experimental evidence. J. Public Econ. 71, 53–73. doi:
10.1016/S0047-2727(98)00049-8
Carré, J. M., Putnam, S. K., and McCormick, C. M. (2009). Testosterone responses to
competition predict future aggressive behaviour at a cost to reward in men.
Psychoneuroendocrinology 34, 561–570. doi: 10.1016/j.psyneuen.2008.10.018
Casto, K. V., and Edwards, D. A. (2016). Testosterone, cortisol, and human
competition. Horm. Behav. 82, 21–37. doi: 10.1016/j.yhbeh.2016.04.004
Chagnon, N. A. (1988). Life histories, blood revenge, and warfare in a tribal
population. Science 239, 985–992. doi: 10.1126/science.239.4843.985
Choi, J.-K., and Bowles, S. (2007). e coevolution of parochial altruism and war.
Science 318, 636–640. doi: 10.1126/science.1144237
Clutton-Brock, T., and Parker, G. A. (1992). Potential reproductive rates and the
operation of sexual selection. Q. Rev. Biol. 67, 437–456. doi: 10.1086/417793
Cosmides, L., and Tooby, J. (1992). “Cognitive adaptations for social exchange” in e
adapted mind: Evolutionary psychology and the generation of culture. eds. J. H. Barkow, L.
Cosmides and J. Tooby (Oxford: Oxford University Press), 163–228.
Daly, M., and Wilson, M. (1988). Homicide. New Jersey: Transaction Publishers.
Elias, M. (1981). Serum cortisol, testosterone, and testosterone-binding globulin
responses to competitive ghting in human males. Aggress. Behav. 7, 215–224. doi:
10.1002/1098-2337(1981)7:3<215::AID-AB2480070305>3.0.CO;2-M
Fischbacher, U. (2007). Z-tree: Zurich toolbox for ready-made economic experiments
[journal article]. Exp. Econ. 10, 171–178. doi: 10.1007/s10683-006-9159-4
Flinn, M. V., Ponzi, D., and Muehlenbein, M. P. (2012). Hormonal mechanisms for
regulation of aggression in human coalitions. Hum. Nat. 23, 68–88. doi: 10.1007/
s12110-012-9135-y
Geniole, S. N., Bird, B. M., Ruddick, E. L., and Carré, J. M. (2017). Eects of
competition outcome on testosterone concentrations in humans: an updated meta-
analysis. Horm. Behav. 92, 37–50. doi: 10.1016/j.yhbeh.2016.10.002
Gonzalez-Bono, E., Salvador, A., Serrano, M. A., and Ricarte, J. (1999). Testosterone,
cortisol, and mood in a sports team competition. Horm. Behav. 35, 55–62. doi: 10.1006/
hbeh.1998.1496
Halevy, N., Bornstein, G., and Sagiv, L. (2008). “In-group love” and “out-group hate”
as motives for individual participation in intergroup conict: a new game paradigm.
Psychol. Sci. 19, 405–411. doi: 10.1111/j.1467-9280.2008.02100.x
Keeley, L. H. (2014). “War before civilization—15 years on” in e evolution of
violence. eds. T. K. Shackelford and R. D. Hansen (New York: Springer), 23–31.
Kiss, H. J., Rosa-Garcia, A., and Zhukova, V. (2020). Conditional cooperation in group
contests. PLoS One 15:e0244152. doi: 10.1371/journal.pone.0244152
Lahr, M. M., Rivera, F., Power, R. K., Mounier, A., Copsey, B., Crivellaro, F., et al.
(2016). Inter-group violence among early Holocene hunter-gatherers of West Turkana,
Kenya. Nature 529, 394–398. doi: 10.1038/nature16477
Leis, O., and Lautenbach, F. (2020). Psychological and physiological stress in non-
competitive and competitive esports settings: a systematic review. Psychol. Sport Exerc.
51:101738. doi: 10.1016/j.psychsport.2020.101738
Lorah, J. (2018). Eect size measures for multilevel models: denition, interpretation,
and TIMSS example. Large-scale Assess. Educ. 6:8. doi: 10.1186/s40536-018-0061-2
Mazur, A. (1985). A biosocial model of status in face-to-face primate groups. Soc.
Forces 64, 377–402. doi: 10.1093/sf/64.2.377
Mazur, A., and Booth, A. (1998). Testosterone and dominance in men. Behav. Brain
Sci. 21, 353–363. doi: 10.1017/S0140525X98001228
Mazur, A., Booth, A., and Dabbs, J. M. Jr. (1992). Testosterone and chess competition.
Soc. Psychol. Q. 55:70. doi: 10.2307/2786687
Mazur, A., and Lamb, T. A. (1980). Testosterone, status, and mood in human males.
Horm. Behav. 14, 236–246. doi: 10.1016/0018-506X(80)90032-X
McDonald, M. M., Navarrete, C. D., and Van Vugt, M. (2012). Evolution and the
psychology of intergroup conict: the male warrior hypothesis. Philos. Trans. R. Soc.
Lond. B Biol. Sci. 367, 670–679. doi: 10.1098/rstb.2011.0301
Mehta, P. H., and Josephs, R. A. (2006). Testosterone change aer losing predicts the
decision to compete again. Horm. Behav. 50, 684–692. doi: 10.1016/j.yhbeh.2006.07.001
Muñoz-Reyes, J. A., Polo, P., Valenzuela, N., Pavez, P., Ramírez-Herrera, O.,
Figueroa, O., et al. (2020). e male warrior hypothesis: testosterone-related cooperation

Belinchón et al. 10.3389/fpsyg.2024.1303372
Frontiers in Psychology 09 frontiersin.org
and aggression in the context of intergroup conict. Sci. Rep. 10:375. doi: 10.1038/
s41598-019-57259-0
Muñoz-Reyes, J. A., Torrico-Bazoberry, D., Polo, P., Figueroa, O., Guzmán-Lavín, E.,
Fajardo, G., et al. (2023). Evidence of the active participation of women in the intergroup
conict based on the use of aggression and cooperation. Sci. Rep. 13:17742. doi: 10.1038/
s41598-023-45012-7
Navarrete, C. D., McDonald, M. M., Molina, L. E., and Sidanius, J. (2010). Prejudice
at the nexus of race and gender: an outgroup male target hypothesis. J. Pers. Soc. Psychol.
98, 933–945. doi: 10.1037/a0017931
Salvador, A., and Costa, R. (2009). Coping with competition: neuroendocrine
responses and cognitive variables. Neurosci. Biobehav. Rev. 33, 160–170. doi: 10.1016/j.
neubiorev.2008.09.005
Schultheiss, O. C., Wirth, M. M., Torges, C. M., Pang, J. S., Villacorta, M. A., and
Welsh, K. M. (2005). Eects of implicit power motivation on men's and women's implicit
learning and testosterone changes aer social victory or defeat. J. Pers. Soc. Psychol. 88,
174–188. doi: 10.1037/0022-3514.88.1.174
Sidanius, J., and Pratto, F. (2001). Social dominance: An intergroup theory of social
hierarchy and oppression. Cambridge: Cambridge University Press.
Simpson, B. (2006). Social identity and cooperation in social dilemmas. Ration. Soc.
18, 443–470. doi: 10.1177/1043463106066381
Stirrat, M., and Perrett, D. I. (2012). Face structure predicts cooperation: men with
wider faces are more generous to their in-group when out-group competition is salient.
Psychol. Sci. 23, 718–722. doi: 10.1177/0956797611435133
Tan, J. H. W., and Bolle, F. (2007). Team competition and the public goods game. Econ.
Lett. 96, 133–139. doi: 10.1016/j.econlet.2006.12.031
Tattersall, I. (2011). “Cooperation, altruism, and human evolution: introduction part
I” in Origins of altruism and cooperation. eds. R. W. Sussman and C. R. Cloninger (New
York: Springer), 11–18.
Trivers, R. L. (1972). “Parental investment and sexual selection” in Sexual selection and
the descent of man 1871–1971. ed. B. Campbell (Chicago: Aldine-Atherton), 136–179.
Van Vugt, M., Cremer, D. D., and Janssen, D. P. (2007). Gender dierences in
cooperation and competition: the male-warrior hypothesis. Psychol. Sci. 18, 19–23. doi:
10.1111/j.1467-9280.2007.01842.x
Van Vugt, M., and Hardy, C. L. (2010). Cooperation for reputation: wasteful
contributions as costly signals in public goods. Group Process. Intergr. Relat. 13, 101–111.
doi: 10.1177/1368430209342258
Weisel, O., and Böhm, R. (2015). “Ingroup love” and “outgroup hate” in intergroup
conict between natural groups. J. Exp. Soc. Psychol. 60, 110–120. doi: 10.1016/j.
jesp.2015.04.008
West, B. T., Welch, K. B., and Galecki, A. T. (2007). Linear mixed models: A practical
guide using statistical soware. Boca Raton: Chapman & Hall.
Wilson, M., and Daly, M. (1985). Competitiveness, risk taking, and
violence: the young male syndrome. Ethol. Sociobiol. 6, 59–73. doi: 10.1016/0162-3095
(85)90041-X
Wingeld, J. C., Hegner, R. E., Duy, A. M. Jr., and Ball, G. F. (1990). e “challenge
hypothesis”: theoretical implications for patterns of testosterone secretion, mating
systems, and breeding strategies. Am. Nat. 136, 829–846. doi: 10.1086/285134
Wrangham, R. W., and Peterson, D. (1996). Demonic males: Apes and the origins of
human violence. New York: Houghton Miin Harcourt.
Yuki, M., and Yokota, K. (2009). e primal warrior: outgroup threat priming
enhances intergroup discrimination in men but not women. J. Exp. Soc. Psychol. 45,
271–274. doi: 10.1016/j.jesp.2008.08.018
Zelmer, J. (2003). Linear public goods experiments: a meta-analysis. Exp. Econ. 6,
299–310. doi: 10.1023/A:1026277420119
Zilioli, S., Mehta, P. H., and Watson, N. V. (2014). Losing the battle but winning the
war: uncertain outcomes reverse the usual eect of winning on testosterone. Biol.
Psychol. 103, 54–62. doi: 10.1016/j.biopsycho.2014.07.022
Zilioli, S., and Watson, N. V. (2014). Testosterone across successive competitions:
evidence for a ‘winner eect’ in humans? Psychoneuroendocrinology 47, 1–9. doi:
10.1016/j.psyneuen.2014.05.001