February 2025
·
4 Reads
Background: Recent models of the development of adolescent psychopathology emphasize the dynamic interplay between substance use and internalizing and externalizing symptoms. Importantly, this interplay may be moderated by known risk factors of substance use and internalizing/externalizing symptoms, particularly impaired working memory capacity and high-risk taking behavior. However, thus far studies into the temporal symptom/substance use dynamics were not designed to test moderation of this dynamic interplay. To fill this critical gap, the present study introduces a novel adaptation of the cross-lagged panel network approach (CLPN) for identifying interaction effects between outside factors and cross-lagged estimates. We use this moderated CLPN approach (mCLPN) to examine how working memory and risk-taking moderate the temporal associations between substance use and internalizing and externalizing symptoms.Methods: Using data from the IMAGEN study (N = 1,364), we tested how working memory and risk-taking at age 14 moderated temporal associations between internalizing/externalizing symptoms and substance use over two years (ages 14 → 16). Results: Alcohol use showed reciprocal associations with externalizing symptoms and predicted heavier tobacco and cannabis use at age 16. Externalizing symptoms at age 14 predicted more externalizing symptoms and substance use at age 16. Poor working memory and high risk-taking moderated the temporal associations between both symptom domains and substance use. When risk-taking was high, the link between internalizing and externalizing symptoms at age 14 and cannabis use at age 16 was stronger.Conclusions: These findings highlight cognitive risk factors in the substance use/ symptom dynamics and illustrate the value of the moderated CLPN approach in clinical-developmental science.Keywords: executive functioning; cross-lagged panel network; substance use; internalizing symptoms; externalizing symptoms