Partner Choice does not Promote Altruism in a Joint Taking Game

The theory of competitive altruism suggests that (1) individuals displaying great altruism are likely to be chosen as partners by others, and (2) when there is a chance of being chosen as partners, people tend to behave more generously. Although many studies have empirically confirmed this theory, none have examined it in situations in which observers can benefit from choosing a selfish player as a partner. Thus, we developed and conducted a joint taking game in which observers could expect to benefit from choosing a selfish player as their partner to test the theory of competitive altruism, compared between generous and self-interested partner choice. Our preregistered online experiment with 221 participants found no evidence that players tended to make altruistic or selfish choices, nor were observers more likely to choose one of these players as partners. These findings suggest that establishing competitive altruism requires a structure in which observers gain from choosing altruistic partners.