Social Gaze Elicits Social Learning

Social gaze is a fundamental cue that regulates approach and avoidance in social interactions, yet it remains unclear how individuals learn from others’ eye gaze behavior. Using reinforcement learning modeling, we examined how micro-level gaze behaviors shape social preferences and closeness. Participants repeatedly chose between two interaction partners who predominantly (70%) responded with either direct or averted gaze and rated their closeness toward each partner. Results showed increased choice frequency and higher closeness ratings for the direct-gaze partner relative to the averted-gaze partner. Computational modeling revealed that others’ gaze direction generated a learning signal updating the latent value of interaction partners. This signal was calibrated by individual differences in how rewarding direct gaze was perceived. Initial attractiveness and valence of the interaction partner predicted this reward calibration, indicating that pre-existing impressions shaped how gaze feedback was valued. In turn, calibration was reflected in participants’ social evaluations of the partners, demonstrating how learning from gaze experiences contributed to impression formation. Together, these findings show that even minimal gaze interactions can shape social preference and closeness through reinforcement learning mechanisms, highlighting eye contact as a powerful but individually calibrated social signal with implications for social engagement in everyday and clinical contexts.