Youth cannabis and alcohol use expectancies mediate associations between pre-adolescent cognitive function and subsequent use initiation.

Youth substance use increases risk for developing substance use disorders. Previous work using the longitudinal ABCD Study® found distinct neurocognitive factors contribute to youth tobacco and alcohol initiation. Using data for 7776 ABCD Study® participants, this study expands prior work to examine prospective associations between early neurocognitive factors (general ability, executive function, learning and memory, and visuospatial and mental rotation) at ages 9-10 years (sample enrolled 2016-2018) and cannabis use by ages 13-14 years. We also test whether positive and negative tobacco smoking, alcohol, and cannabis expectancies mediate associations between neurocognitive factors and substance use initiation. Higher performance in general ability was associated with increased risk [OR = 1.23, 95 % CI 1.07-1.42] for cannabis use; positive cannabis expectancies mediated 72.6 % (p-value = 0.003) of the effect [Indirect effect: OR = 1.16, 95 % CI 1.12-1.20] and negative expectancies mediated -10.2 % (p-value = 0.04) of the effect [Indirect effect: OR = 0.98, 95 % CI 0.97-0.99]. Accuracy in visuospatial reasoning was protective [OR = 0.83, 95 % CI 0.73-0.95)] for cannabis use; the effect was not mediated by expectancies. Positive alcohol use expectancy mediated 36.3 % of the association between general ability and early alcohol use [OR = 1.15, 95 % CI 1.05-1.25; Indirect effect: OR = 1.05, 95 % CI 1.03-1.07]. Associations with early tobacco use were independent of tobacco expectancies. Developmentally appropriate expectancy-based interventions lowering positive expectancies and bolstering negative expectancies may be effective for preventing youth cannabis initiation. Interventions lowering positive alcohol expectancies may help prevent youth alcohol use.