This study used a genetically informed design to assess the effects of friends' and nonfriends' reticent and dominant behaviors on children's observed social reticence in a competitive situation. Potential gene-environment correlations (rGE) and gene-environment interactions (GxE) in the link between (a) friends' and nonfriends' behaviors and (b) children's social reticence were examined. The sample comprised 466 twin children (i.e., the target children), each of whom was assessed in kindergarten together with a same-sex friend and two nonfriend classmates of either sex. Multilevel regression analyses revealed that children with a genetic disposition for social reticence showed more reticent behavior in the competitive situation and were more likely to affiliate with reticent friends (i.e., rGE). Moreover, a higher level of children's reticent behavior was predicted by their friends' higher social reticence (particularly for girls) and their friends' higher social dominance, independently of children's genetic disposition. Children's social reticence was also predicted by their nonfriends' behaviors. Specifically, children were less reticent when male nonfriends showed high levels of social reticence in the competitive situation, and this was particularly true for children with a genetic disposition for social reticence (i.e., GxE). Moreover, children genetically vulnerable for social reticence seemed to foster dominant behavior in their female nonfriend peers (i.e., rGE). In turn, male nonfriends seemed to be more dominant as soon as the target children were reticent, even if the target children did not have a stable genetic disposition for this behavior.