An Exploratory Follow-Up Study of Cannabis Use and Decision-Making Under Various Risk Conditions Within Adolescence

Objective: Studies examining the associations between decision-making (DM) and cannabis use (CU) often use cross-sectional, adult samples, and composite scores or single tasks to assess DM. The present study explored differential associations between tasks assessing DM under various risk conditions (i.e., ambiguous vs. explicit; gain vs. loss) and CU frequency, CU-related problems, and CU disorder (CUD) onset across a 2-year period within adolescence. Method: Adolescents (n = 401, 90% Hispanic) aged 14-17 at baseline participated in five biannual assessments. CU frequency, CU-related problems, and CUD were assessed using the Drug Use History Questionnaire, Marijuana Problems Scale, and Structured Clinical Interview for Diagnostic and Statistical Manual of Mental Disorders, fourth edition, respectively. DM was assessed using the Iowa gambling task (IGT), Game of Dice Task (GDT), and Cups Task. We used latent growth curve modeling to examine bidirectional associations between DM and escalation in CU frequency and CU-related problems, and discrete time survival analyses to determine whether baseline performance across DM tasks predicted CUD onset. Results: Baseline performance on the GDT predicted greater escalation in CU (beta = .200, p = .008) and CU-related problems (beta = .388, p = .035). No other significant associations were found. Conclusions: DM under explicit risk may be a more salient risk factor for escalating CU and CU-related problems than DM under ambiguous risk. Deficits in executive functioning could partially explain the results. Findings suggest that neurocognitive development should inform prevention and intervention efforts focused on reducing CU. Given the exploratory nature of the present study, replication of findings is needed.