Social norms guide our behaviors in many decision contexts and influence our decisions such as whether to trust or cooperate with others. To do so, individuals observe and interpret the social signals that their peers provide through their behavior. In this paper, we investigate how individuals navigate strategic environments that are characterized by different distributions of behavior and we contribute to a burgeoning literature probing how different social environments inform one’s own decisions. We focus on the difference between tight (i.e., characterized by low behavioral variance), loose (i.e., characterized by high behavioral variance), and polarized (i.e., characterized by U-shaped behavior) environments. Our results show that individuals strongly adapt their actions to the variance and distribution (polarized/single-peaked) of one’s peers. In particular, higher variance environments generate greater variance of replies, and polarized environments generate polarized responses. This implies that tight, loose, and polarized (empirical) norms are self-sustaining. Moreover, we find that personal values have a stronger importance for actual behavior in polarized and loose than in tight environments.