Smarter Kids Try Cannabis Earlier — But Positive Expectations Explain Why | Research

The takeaway

Higher general cognitive ability at ages 9-10 predicted cannabis use by ages 13-14, but 73% of this effect was explained by developing more positive expectations about cannabis — suggesting expectancy-based prevention could be highly effective.

School-based prevention program designers, child psychologists, parents, and adolescent health researchers.

What the researchers found

Higher general ability at ages 9-10 was associated with cannabis initiation by 13-14 (OR=1.23), with positive cannabis expectancies mediating 72.6% of the effect (p=0.003). Better visuospatial reasoning was protective (OR=0.83). Positive alcohol expectancies mediated 36.3% of the general ability-alcohol use association.

Why it matters

This counterintuitive finding — that cognitive ability predicts cannabis use — is almost entirely explained by positive expectancies, meaning prevention programs that challenge positive cannabis beliefs could be targeted and effective.

The numbers in context

N=7,776; general ability→cannabis OR=1.23; 72.6% mediated by positive expectancies (p=0.003); -10.2% mediated by negative expectancies (p=0.04); visuospatial reasoning protective OR=0.83; positive alcohol expectancy mediated 36.3%

How the study worked

Prospective longitudinal analysis of 7,776 ABCD Study participants enrolled at ages 9-10 (2016-2018) through ages 13-14, examining how pre-adolescent neurocognitive factors predict substance use initiation through expectancy mediation models.

What this study cannot tell us

ABCD Study cohort may not represent all US youth; expectancies measured concurrently with some substance use; mediation models assume temporal ordering; cannot fully separate curiosity from actual use; cannabis use by 13-14 is relatively uncommon.

How to read the evidence

Large prospective ABCD Study cohort with validated neurocognitive measures and sophisticated mediation analysis, providing strong evidence for the expectancy pathway.

When this study was published

Published 2026; uses ABCD Study data from 2016-2022.

The bigger picture

Prevention programs should not just target 'at-risk' youth with behavioral problems — cognitively capable youth who develop positive cannabis expectations are also on a path toward early use.

Questions still open

How do positive cannabis expectancies form in pre-adolescents? Would media literacy programs reduce positive expectancy development? Does the visuospatial reasoning protective effect reflect better decision-making capacity?

Common questions

Are smarter kids more likely to try cannabis?

Somewhat — higher general cognitive ability at ages 9-10 predicted cannabis use by 13-14, but this was almost entirely (73%) explained by these children developing more positive expectations about cannabis, not by intelligence itself.

How can we prevent early cannabis use?

This study suggests targeting positive cannabis expectations is key — developmentally appropriate programs that build negative expectations about cannabis and challenge positive ones could prevent up to 73% of the cognitive ability-linked risk for early use.

Read the original research

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

Addictive behaviors, 173, 108533

View on PubMed ↗Publisher’s paper ↗

Citation

Jones, Stephanie K; Tomko, Rachel; Ramer, Nolan; Wolf, Bethany J. (2026). Youth cannabis and alcohol use expectancies mediate associations between pre-adolescent cognitive function and subsequent use initiation.. Addictive behaviors, 173, 108533. https://doi.org/10.1016/j.addbeh.2025.108533