ScholarWorks@성균관대학교

초록

Prior research on social decision-making has largely focused on monetary trade-offs between self and others, often in static contexts with known payoffs. However, it remains unclear how interpersonal closeness and individual differences in prosocial orientation influence information processing when individuals must concurrently make decisions for themselves and others in dynamic environments. In such contexts, decision-makers must allocate cognitive resources across multiple payoff streams—learning not only for themselves but also on behalf of others. To address this gap, we employ a modified Social Gambling Task (SGT) in which participants learn optimal decisions through trial-and-error for both self and a partner—either a close friend or a stranger. This setup removes direct payoff conflict, enabling us to isolate how social distance and stable prosocial traits shape concurrent learning processes. Behavioral analyses, learning indices, and computational reinforcement learning models reveal that both social closeness and prosocial orientation modulate the weighting of others' outcomes during learning, with closeness exerting a stronger influence than dispositional traits. These findings advance our understanding of the layered cognitive and motivational mechanisms underlying value-based learning in social contexts.

키워드