How agency shapes social preferences: Using mouse-tracking to reveal changes in cognitive conflict

Inequity aversion is a fundamental social preference. Concern for others’ welfare is yet another fundamental social preference. What do people prefer when maximizing others’ welfare creates an inequality that disadvantages them? Prior research has shown that people’s preferences depend on their level of agency, that is, their control over the allocation. Behaviorally, agentic decision makers have been shown to be more likely to prefer allocations that give another person more than them, compared to non-agentic ones. However, the cognitive processes underlying this shift of preference remain unclear. Here, we used mouse-tracking and reaction times to investigate how agency shapes social preferences in such cases. We examined decision conflict, the level of conflict experienced in arriving at a given choice, as a way to uncover whether different underlying social preferences are at play when one has versus does not have agency. We predicted an interaction effect of agency and choice on conflict, where choosing the prosocial option would be associated with less conflict in the Agency condition compared to the No-Agency condition, and that the opposite would be true for the equitable option. In our task, on each of trial, participants were presented with two allocation options: an equitable option (e.g., 50 cents for self, 50 cents for other) and an inequitable and prosocial option, favoring the other participant over oneself (e.g., 50 cents for self, 70 cents for other). Agentic participants chose one option to be implemented, whereas non-agentic participants indicated their preference, knowing that the lab manager had already made the decision. In two studies (N = 586), including a preregistered one, an interaction effect emerged in both mouse trajectory and reaction time analyses. These results suggest that agency alters social preferences during the decision-making process, and challenge traditional inequity aversion models claiming that people inherently dislike inequitable outcomes. Granting agency may be a powerful tool for maximizing social welfare in situations where equity and prosocial behavior are at odds.