In this paper, I propose and test a new model of peer influence to analyze how cultural forms can mediate social influence. Peer influence is a fundamental process in social life. Network scientists generally conceptualize influence as an additive process, with increasing exposure to others’ behavior or attitudes associated with a greater likelihood of adopting them. Based on cognitive theories of symbolic boundaries, I propose an alternative model that regards peer influence as conditional on membership in the same symbolic group. I test this model with data from a large, nationally-representative dataset. I use a counterfactual logic and instrumental variable methods to estimate peer influence on adolescent drinking. I find that, for religious adolescents, having a same-religion friend who drinks is more strongly predictive of drinking behavior than exposure to drinking and same-religion friends alone. For some behaviors, peer influence should not be understood simply as a function of exposure, but also of the violation of symbolic group boundaries.