Revisiting the bounded generalized reciprocity model: Ingroup favoritism and concerns about negative evaluation

The bounded generalized reciprocity (BGR) model, grounded in reputation management, predicts that the motivation underlying ingroup favoritism (favoring one’s own group over other groups) is driven by avoiding a negative reputation within one’s own group. This research conducted two economic games with minimal groups in which reputational concerns (partners’ knowledge of participants’ group membership) were manipulated. We aimed to verify the replicability of the experimental results in support of the BGR model. A study (N = 394) using a dictator game (in which participants unilaterally determined their partners’ payoffs) indicated the following: (1) participants were more likely to behave cooperatively with ingroup partners than with outgroup partners, regardless of whether their partners knew the participants’ group membership; and (2) individual differences in fear of negative evaluation by others were not associated with cooperation toward the ingroup. Similar results were found in another study (N = 429) using a prisoner’s dilemma game (in which participants’ payoffs were determined by their partners). However, while sharing knowledge about group membership facilitated cooperation with the outgroup in the dictator game, no such tendency was observed in the prisoner’s dilemma game. These findings suggest that concerns about a bad reputation may not play a relatively important role in ingroup favoritism, and that generosity toward outgroup members is influenced by the presence or absence of interdependence (i.e., whether the partner’s behavior affects the participant’s payoffs). This research proposes a reconsideration of the motivations behind cooperation within or between groups.