Testing the dual process model of adolescent cannabis use with prospective three-way interactions between self-regulation, negative outcome expectancies, and implicit cannabis attitudes

Objective: Dual-process models of substance use (Wiers et al., 2007) propose that whether automatic processes (i.e., implicit attitudes) influence use depends on self-regulation, such that an individual is more likely to act in accordance with automatically activated implicit attitudes when there is limited capacity for self-regulation (a two-way interaction). In this model, the relevance of self-regulation likely depends on whether an individual recognizes reasons or the need to inhibit substance use. The current study tested a three-way interaction between implicit cannabis attitudes, self-regulation, and negative expectancies to prospectively predict adolescent cannabis use. Method: A community sample of late adolescents (N = 246; M age = 19.02) were assessed across two annual time points. Negative binomial regressions predicting adolescent cannabis use were estimated to test the proposed interaction using two indictors of self-regulation (effortful control and working memory) above and beyond prior cannabis use and demographic covariates. Results: There was support for a three-way interaction with effortful control, but not working memory. As hypothesized, implicit cannabis attitudes were positively associated with cannabis use for adolescents with low levels of negative expectancies and low levels of effortful control. Conclusion: Findings suggest expecting negative outcomes of cannabis use is germane to dual process interactions between controlled and automatic processes. How to best operationalize self-regulation is an important consideration for future work.