It turns out that there is a subset of social psychology known as ‘coalitional psychology’. It studies the formation and consequences of coalitions and group behavior. Coalitions form fairly easily: something as simple as the identification of a shared interest can create one. Interestingly, there is a large difference in how the two sexes (genders?) interact with coalitions. Geary, in his own research and in reviews of the relevant literature (e.g., [Geary, 1998], [Geary, 1999] and [Geary et al., 2003], has provided evidence showing that sex differences in group activity are widely shared across culture, influenced by prenatal hormonal differences, emerge in early childhood and continue across the life course.’ So take that for what it’s worth.
The first study (Yamagishi and Mifune, 2009) divides participants between those who prefer Klimt and those who prefer Klee. They are then asked to perform a variant of the prisoner’s dilemma: each is given 100 Yen and independently (and simultaneously) are given the option to transfer some amount of that to a partner, who gets twice that amount. They only get a given partner once, so the results aren’t biased by the expectation of infamy. Each round, each partner is secretly either given the coalition-status of their opponent or not.
Men and women both expected an equal level of cooperation from their in-group when both partners were aware of the coalition. However, men expected punishment from the out-group while women did not. The experiments showed that both men and women had a roughly equal in-group bias in the common-knowledge condition. However, in the private-knowledge condition, men showed a significant in-group bias (with no negative bias to out-group) whereas women acted as if they had no knowledge of group membership. The conclusion is that the in-group bias in men comes from in-group love rather than out-group hate.
In a similar experiment, (Yuki and Yakota, 2009) first had participants do a “language study”. They were asked to circle all nouns in three essays. In the “outgroup threat” condition, the second study was written by a member of another culture who was offensively criticizing the participants culture. Participants then were asked to perform a second, unrelated experiment. They were shown Klee/Klimt paintings and told they were separated according to preference (in reality, they were all assigned to Klee). They then each privately asked to divide 500 yen between an in-group member and an “out-group” member. Males showed an in-group bias in money only when they were primed in the “outgroup threat” condition in the first experiment; women never showed a bias.
The conclusion of these first two experiments is that men show a gender-specific in-group bias that is a result of in-group preferences rather than out-group hate. It is when the in-group is threatened that out-group attacks occur.
(Oxford et al. 2009) offer hormonal evidence for in-group/out-group differences. He set participants to war in Unreal Tournament, and measured testosterone levels. After between-group play in Unreal Tournament, men on the winning team had an increased testosterone immediately after the competition, with those who contributed the most to the win showed the largest increase. High-scoring men on the losing teams didn’t show this effect, though they did show a delayed slow increase in testosterone after the match. During within-group tournaments, high-ranking men had lower testosterone but higher cortisol. Further, if the within-group play occurred before the between-group play, there was no effect on testosterone after the between-group play. This probably signals that the ‘group’ idea was lost due to an initial disruption of coalition. Strangely, this effect was seen on McCain voters after the 2008 elections…
Another study found a similar, but muted, testosterone effect on soccer-playing women. Does testosterone contribute to sex- or gender-specific coalitional differences? Who knows. One extremely large caveat to the first two studies is that it’s entirely possible that women show similar coalitional behavior in other paradigms or situations. Presumably women have a group-building hormone, too. After all, women who were injected with testosterone make significantly fairer offers in the Ultimatum game than do those injected with a placebo. The fact that it is so effective in both sexes indicates that there’s a need for it in both sexes.
Yamagisihi and Mifune, 2009. “Social exchange and solidarity: in-group love or out-group hate?” DOI.
Yuki and Yokota, 2009. “The primal warrior: Outgroup threat priming enhances intergroup discrimination in men but not women.” DOI.
Jonathan Oxforda, Davidé Ponzia and David C. Geary, 2009. “Hormonal responses differ when playing violent video games against an ingroup and outgroup.” DOI.
Daphne Blunt Bugentala and David A. Beaulieub, 2009. “Sex differences in response to coalitional threat.” DOI.