et al. 2000). In addition, as hypothesized, externalizing polygenic score predicted greater peer substance use, which in turn was associated with adolescents’ subsequent externalizing behavior, while controlling for prior level of externalizing behavior. Our results indicated that peer substance use serves as a mediating pathway of genetic influences on adolescent externalizing behavior, presumably reflecting active and evocative gene-environment correlation processes as adolescents select and shape their environments (Scarr and McCartney 1983). These findings are consistent with previous research, with evidence from mostly twin and family studies, showing that children’s genetic predispositions are associated with their peer group selections/ affiliation (Elam et al. 2017; Fowler et al. 2007; Harden et al. 2008; Mann et al. 2016; TenEyck and Barnes 2015), as well as with the robust relationship between peer substance use and adolescents externalizing behavior (Hawkins et al. 1992; Pratt et al. 2010). It is also possible that adolescents with a genetic predisposition toward externalizing may be considered “more fun” for other like-mind peers, making other children with similar externalizing tendencies more likely to want to spend time with them. Taken together, our results demonstrate that genetic risk conferred not only direct effects on adolescent externalizing behavior but also impacted subsequent
