American Military Children Engage in Competitive Altruism Article Review -Use an article related to social psychology. -Prepare a 1-2 page summary of the article in your own words including specifics regarding the overall purpose of the research in question, a clear statement of the researcher’s hypothesis, details regarding the study methodology, and pertinent results of the manipulation. -Also include your personal opinion of the work. Should it be repeated/ how can it be improved? What was your overall impression of the work? What are the implications of the study? Be sure to cite the article appropriately at the end of your review. Papers should be typed, double-spaced, and formatted according to APA style. Journal of Experimental Child Psychology 179 (2019) 176–189
Contents lists available at ScienceDirect
Journal of Experimental Child
Psychology
journal homepage: www.elsevier.com/locate/jecp
Children engage in competitive altruism
Esther Herrmann a,⇑, Jan M. Engelmann b,c, Michael Tomasello d
a
Minerva Research Group of Origin of Human Self-Regulation, Max Planck Institute for Evolutionary Anthropology, 04103
Leipzig, Germany
b
Department of Psychology, Yale University, New Haven, CT 06520, USA
c
Department of Developmental Psychology, Georg-Elias Müller Institute of Psychology, University of Göttingen, 37073
Göttingen, Germany
d
Department of Psychology and Neuroscience, Duke University, Durham, NC 27708, USA
a r t i c l e
i n f o
Article history:
Received 26 July 2018
Revised 13 November 2018
Available online 8 December 2018
Keywords:
Competitive altruism
Reputation
Cooperation
Partner choice
Dyadic sharing game
5- and 8-year-old children
a b s t r a c t
Humans cultivate their reputations as good cooperators,
sometimes even competing with group mates, to appear most
cooperative to individuals during the process of selecting partners.
To investigate the ontogenetic origins of such ‘‘competitive
altruism,” we presented 5- and 8-year-old children with a dyadic
sharing game in which both children simultaneously decided
how many rewards to share with each other. The children were
either observed by a third-person peer or not. In addition, the children either knew that one of them would be picked for a subsequent collaborative game or had no such knowledge. We found
that by 8 years of age, children were more generous in the sharing
game not only when their behavior was observed by a third party
but also when it could affect their chances of being chosen for a
subsequent game. This is the first demonstration of competitive
altruism in young children, and as such it underscores the important role of partner choice (and individual awareness of the
process) in encouraging human cooperation from an early age.
Ó 2018 Elsevier Inc. All rights reserved.
Introduction
Humans, in comparison with other primates, are remarkable for their degree of cooperation and
altruistic acts toward unrelated individuals (Axelrod & Hamilton, 1981; Tomasello, 2009). Researchers
⇑ Corresponding author.
E-mail address: eherrman@eva.mpg.de (E. Herrmann).
https://doi.org/10.1016/j.jecp.2018.11.008
0022-0965/Ó 2018 Elsevier Inc. All rights reserved.
E. Herrmann et al. / Journal of Experimental Child Psychology 179 (2019) 176–189
177
have consistently been interested in the mechanisms supporting such hyper-cooperation in humans
and in how altruism—which on the surface often seems counter to the actor’s best interests—is maintained. There are a limited number of ways in which altruism may be stabilized within a population.
Most commonly referred to is direct reciprocity in which individuals in essence trade favors over time
(Trivers, 1971). Also important is so-called indirect reciprocity in which individuals cultivate their reputation for cooperation with others (because cooperators naturally want to interact with one another
and not cheaters; Alexander, 1987). In situations where an observer chooses among potential partners
for a mutually beneficial collaborative enterprise, the potential partners might even compete with one
another to appear to be the best cooperator. Roberts (1998) labeled this process competitive altruism. It
has been argued that competitive altruism has played an especially important role in the evolution of
humans‘ ultra-cooperative (even moral) tendencies (Baumard, André, & Sperber, 2013; Tomasello,
Melis, Tennie, Wyman, & Herrmann, 2012). Unlike their great ape relatives, early humans became obligate collaborative foragers, which meant that choosing a good partner and being chosen as a partner in
turn were essential to an individual’s success.
A number of studies have provided evidence that adult humans are motivated to develop a reputation
for being a good partner and, better yet, for being the most cooperative in their group (Barclay & Willer,
2007; Debove, André, & Baumard, 2015; Sylwester & Roberts, 2010; see also Hardy & Van Vugt, 2006, for
reputation-based generosity). For example, Barclay and Willer (2007) conducted a study consisting of
three conditions in which two players took part in a continuous prisoner’s dilemma while a third player
either had knowledge of the contributions of both players or not. In a second step, the third player either
was randomly paired with one of the two initial players (either having knowledge about their contributions or not) or was allowed to choose one of them after being informed about their contributions. Only
the two conditions in which the observer was informed about the donations allowed the two players to
develop their reputations for being collaborative partners, whereas competition for being chosen as a
future partner was present only in the last condition. As predicted, people donated more money, and
thus actively competed to be more generous, than others in the condition where partner choice played
a role than when donation could only result in self-presentation as a good partner (see also Experiment 1
in Sylwester & Roberts, 2010, for similar findings). Furthermore, competing for being the most altruistic
partner paid off given that the most generous individuals were chosen more often as cooperative partners than those who donated less. Compellingly, after being chosen, players donated near zero in the second game, indicating that being generous in the first round was merely a strategic investment that was
unnecessary in the second game where generosity had no effect on future interactions. In another study
(Sylwester & Roberts, 2010), participants played a public goods game in which they could contribute any
amount of the money they received to a common pool, after which they had the opportunity to select a
partner for a second cooperative game. Thanks to the possibility of partner choice, individuals who had
cooperated in the first game were more likely to be recruited for the second game than non-cooperators
and received higher payoffs. Finally, in a recent study, Debove et al. (2015) not only showed that partner
choice can result in competitive altruism when the biological market is uneven—in this case, when there
are fewer individuals selecting partners than individuals seeking to be chosen—but also showed that it
can lead to fairness in contexts where choosers and those being chosen are equally present. The results of
these studies demonstrate that in the right situation adult humans will engage in competitive altruism.
Whereas these studies document the presence of competitive altruism in human adults, little is
known about the ontogeny of human children’s skills and motivations for such forms of altruism. Early
in ontogeny, human infants seemingly perform altruistic acts indiscriminately (Warneken &
Tomasello, 2009). After this stage, and relevant to the current investigation, young children begin to
care about their reputation. As early as 5 years of age, children not only behave more generously when
observed by others (Engelmann & Rapp, 2018; Engelmann, Herrmann, & Tomasello, 2012; Leimgruber,
Shaw, Santos, & Olson, 2012) but also invest in their reputation in order to benefit at a later point by
indirect reciprocity (e.g., Engelmann, Over, Herrmann, & Tomasello, 2013). Recently, it has been shown
that by 4 years of age children invest in their reputation in a competitive group setting and that by
5 years children flexibly apply different reputational strategies depending on context (Rapp,
Engelmann, Herrmann, & Tomasello, in press). Preschoolers’ concern for their reputation is even more
nuanced in that they take into account not only the fact of being observed but also the nature of the
audience (e.g., authority, team, in-group member). (Banerjee, 2002a; Engelmann et al., 2013; Goffman,
178
E. Herrmann et al. / Journal of Experimental Child Psychology 179 (2019) 176–189
1959; Piazza, Bering, & Ingram, 2011). Also relevant for the current investigation is a recent finding
demonstrating that children compare their own generosity with the generosity of a peer. Specifically,
Tasimi, Dominguez, and Wynn (2015) showed that children’s preference for a generous child was
reduced when a comparison with their own (lesser) generosity was possible and, hence, their reputation was at stake. Finally, and most pertinent to the current study, Aloise-Young (1993) asked 6-, 8-,
and 10-year-old children to explain to a potential team member why they should be picked as a partner for a game. From 8 years of age, participants exhibited appropriate self-promoting behavior in
order to convince another child to pick them as a partner for a skill-based game (Aloise-Young,
1993). Although this documents the ability of children to present themselves in ways that maximize
their chances of being picked as a partner, it does not provide evidence for competitive altruism. First,
children in this study were not in direct competition with peers. Second, the relevant dimension for
partner choice in the study was not altruism but rather children’s skill at playing a game. Thus, none
of these previous studies investigated children’s concern for their cooperative reputation in a partner
choice scenario where they needed to compete with another peer to be chosen.
We wished to investigate the ontogeny of competitive altruism, specifically the act of investing in
one’s reputation at a personal cost in a competitive context. In the current study, children competed to
be chosen by a peer to play a highly rewarding game. In Study 1 (8-year-olds) and Study 2 (5-yearolds), child dyads participated in a dyadic sharing game where both children simultaneously decided
how many balls to share with each other. In Study 1, each dyad was assigned to one of four conditions
that varied in two ways. First, children either were observed by a peer (observer conditions) or played
the game anonymously (anonymous conditions). Second, either the children were told that after finishing the initial game one of them would be picked by a third peer to play a special and more rewarding second game (choice conditions) or they had no knowledge of this second game (no choice
conditions). We predicted that the lowest sharing rates would occur in the two anonymous conditions
where there was no possibility of reputational benefit for being generous. We further predicted that
sharing rates would increase in the audience condition (observer/no choice) where an observer was
aware of children’s generosity and, thus, reputation building was possible. However, the highest sharing rates should occur in the competitive altruism condition (observer/choice). Participants should
invest in their reputation for being generous persons in order to outcompete another peer for being
chosen by the observer as a future interaction partner for a mutualistic game. Finally, we predicted
that in the competitive altruism condition children’s sharing would decrease from the first game, in
which their choices may be strategic, to the second game, in which their distributive choices remain
private and do not influence future interactions. We did not expect a similar decrease in sharing in the
anonymous/choice condition given that reward allocations remained anonymous in both games.
Based on previous studies showing that children’s concern about their reputation and their use of
self-presentational strategies increase between 5 and 8 years of age (Aloise-Young, 1993; Banerjee,
2002a; Engelmann et al., 2012, 2013; Leimgruber et al., 2012; Piazza et al., 2011; Shaw et al.,
2014), we predicted that 5-year-olds would care about their reputation and donate more in observer
conditions compared with anonymous conditions. However, based on the findings by Aloise-Young
(1993), we expected only our older age group (8-year-olds) to attempt to strategically outcompete
potential rivals for the best collaborative reputation through generous donations, that is, to engage
in competitive altruism.
Study 1
Method
Participants
Participants were 128 8-year-old children (64 girls; age range = 7.9–8.3 years) from a mediumsized German city (population 500,000). The children were tested in same-sex pairs (64 dyads in
total; 16 dyads per condition as set a priori). In addition, 6 female confederates (ages 7–9 years)
and 8 male confederates (ages 7–9 years) participated throughout the study. Confederates and participants were recruited from different day-care centers to ensure that children did not know each other
E. Herrmann et al. / Journal of Experimental Child Psychology 179 (2019) 176–189
179
before coming to the child laboratory for this study. All participants (predominantly Caucasian) were
native German speakers and came from mixed economic backgrounds. Informed consent was
obtained for all children from their parents. An additional 12 8-year-old dyads were tested but
excluded. The reasons for exclusion were as follows: one child or both children told their partner
how many balls they shared or looked at the outcome of the other child (one child: n = 5; both children: n = 4; resulting in a total of n = 14 children), confederate errors (n = 2, resulting in a total of n = 4
children), stopped playing (n = 1, resulting in a total of n = 2 children), and technical problems (n = 2,
resulting in a total of n = 4 children).
Apparatus
A wooden rectangular box (66 100 12 cm) topped by a plastic occluder (50 100 cm) was
attached to a small wooden table. Two opaque plastic ducts, 5 cm in diameter, ran through the box
at a diagonal such that each duct started on one side and ended on the opposite side. The bottom hole
of each duct ended in a small wooden box (81 40 30 cm) akin to a cupboard with a hinged door at
the front. Each door had a different symbol (car or house) attached to it with Velcro. A second apparatus, identical to the first except for its color (gold), was placed opposite the first one (see Fig. 1). In
addition, cardboard boxes with a lid and marked by either a car or house symbol served as containers
in which balls were provided. Two little toy suitcases depicting either a car or a house served as
reward collecting boxes (see Fig. 2). Plexiglas tubes (length = 25 cm, diameter = 4 cm) with a rubber
cork could be baited with green or golden Styrofoam balls (diameter = 3 cm) and sent through the opaque ducts into the small wooden boxes. Stickers served as rewards, which were exchanged for the collected balls at the end of the session.
Procedure
In the study, two children played a sharing game. The goal of the game was to collect as many balls
as possible. At the end of the game, the balls were exchanged for prizes. The children sat on opposite
sides of the apparatus with a blocked view of each other (see Figs. 1 and 2). Before every trial (four in
total), each child received a box containing seven balls and a Plexiglas tube. The children were told by
the first experimenter (E1) that they could put as many balls in this tube as they wished to send to the
other child through an opaque duct. In addition, children were informed that they should leave the
balls they wanted to keep in their box and close the lid so that no one could see how many balls were
inside. The small tube containing the balls that were sent to the other child fell into an opaque wooden
box hidden to both participants. Each dyad was assigned to one of four conditions that varied in two
aspects. In two conditions (observer/choice and observer/no choice), another child (confederate) was
present. The confederate sat in front of a second apparatus identical to the one with which the dyad
was playing but golden in color (see Fig. 1). In the other two conditions (anonymous/choice and
Fig. 1. Experimental setup in the observer conditions (observer/choice and observer/no choice) (A) and anonymous conditions
(anonymous/choice and anonymous/no choice) (B). Participants sat on opposite sides of the apparatus with no visual access to
each other. In the observer conditions, a peer confederate sat in front of the second golden apparatus. In the beginning of each
trial, participants received a tube and seven green balls. The shared balls were sent through a duct and landed in a wooden box.
The number of shared balls was visible to the observer only after opening the doors to the boxes.
180
E. Herrmann et al. / Journal of Experimental Child Psychology 179 (2019) 176–189
Fig. 2. Basic experimental setup of the first sharing game for all four conditions from a bird’s-eye view.
anonymous/no choice), no confederate was present. Depending on condition, either the children were
told that after finishing the initial game one of them would be picked by another child (the confederate) to play the golden game (observer/choice and anonymous/choice) or they were not told about it
(observer/no choice and anonymous/no choice). The assignment to conditions was randomized, and
all four conditions were run in parallel.
Before each dyad played the game together, each child was individually introduced to the apparatus in the familiarization phase as follows. E1 entered the room with the child, who was given a bracelet with either a car or house symbol on it. E1 told the child that the goal of the game was to collect
as many balls as possible. Then E1 pointed out that there were two games present (apparatuses) that
are played using the same rules. She also emphasized that the golden apparatus was very special and
that playing with it even included playing with golden balls. Then she told the child that they would
first play with the ordinary game and that the child might be able to play with the fancy golden one at
a later time. The child was told to sit down on the side of the apparatus that depicted the same symbol
as his or her bracelet and was told that another child with the other symbol would later sit on the
other side. E1 continued explaining how the apparatus works: ‘‘If you put something in this duct
on your side, it will slide to the other child’s side and will end up in their opaque wooden box. And
if something is placed in the duct on the other side, it will end up in your box.” Then E1 gave the child
a box with two green balls and emphasized that they belonged to the child. Then she handed over a
transparent tube depicting the other child’s symbol and explained how the game is played while
demonstrating by putting one ball in the tube and sending it to the other side. E1 continued to explain
that the remaining balls, which were still in the box, belonged to the child and would be collected in a
little suitcase depicting the same symbol next to the apparatus (see Fig. 2). Then E1 showed the child
where the tube containing the ball landed by opening both opaque boxes. E1 explained that the ball
arrived in the other child’s box and that it would be collected in his or her little suitcase, whereas
nothing had yet arrived on the child’s side. After this demonstration, the child was allowed to play
alone (i.e., without instructions by E1). E1 handed the child a box containing seven balls and a tube
while explaining that all the balls belonged to the child. Then E1 repeated that the child could decide
to put as many balls in the tube as he or she wished to send to the other child. The child should leave
the balls that he or she wanted to keep in the child’s box and should close the box after making the
decision. After that, E1 went to the other side, where she also baited a tube and sent the tube through
E. Herrmann et al. / Journal of Experimental Child Psychology 179 (2019) 176–189
181
the apparatus. After they both sent their tubes, E1 explained that from then on they were not allowed
to look in any of the box…
Purchase answer to see full
attachment
Uncategorized