Longitudinal Neurocognitive Trajectories in a Large Cohort of Youth Who Use Cannabis: Combining Self-Report and Toxicology.
Natasha E Wade, Ryan M Sullivan, Alexander L. Wallace, Rachel Visontay, Veronica Szpak, Krista M. Lisdahl, Marilyn A. Huestis, Priscila Dib Goncalves, Hollie Byrne, Louise Mewton, Joanna Jacobus, Susan F Tapert
bioRxiv : the preprint server for biology March 6, 2026 preprint DOI: 10.64898/2025.12.20.695698 (opens in new tab)
Study at a glance
AI-extracted from the abstract| Characteristics | Longitudinal cohort study |
|---|---|
| Sample size | 11,036 |
| Population | Adolescents aged 9-17 from the ABCD Study |
| Topics | Cannabis |
| Key findings | Cannabis use onset interacts with age to alter neurocognitive trajectories across multiple domains, with reduced improvement or flattened trajectories over time. THC exposure specifically predicted worse episodic memory compared to controls, while CBD exposure showed no difference. |
Abstract
Adolescents experience extensive neurocognitive development, with cannabis use potentially impacting developmental trajectories. Here, we comprehensively assess the influence of adolescent cannabis use onset on neurocognitive trajectories and consider how recent delta-9-tetrahydrocannabinol (THC) and cannabidiol (CBD) may influence neurocognition. We use the large, diverse longitudinal Adolescent Brain Cognitive Development (ABCD) Study dataset, combining self-reported substance use with objective toxicological tests (hair, urine, breath, oral fluid). Longitudinal mixed methods of the full cohort (n=11,036, ages 9-17; 47% Female/53% Male) investigate time-varying cannabis onset on neurocognitive performance. Primary model covariates include sociodemographics, family history of substance use disorder, prenatal substance exposure, early psychopathology, other substance use, and nesting for participant ID, study site, and family ID. Secondarily, in participants with repeat toxicological hair testing (n=645; 38% Female/62% Male) at ages 12-16, we consider the influence of THC v. CBD v. Controls. Primary models included false discovery rate corrections (FDR-p<.05) while secondary models were interpreted at p<.01. Cannabis group interacted with age to show altered neurocognitive trajectories across domains (immediate recall and delayed memory, processing speed, inhibitory control, visuospatial processing, language, and working memory; βs=-0.11- -0.52). Secondary models indicated hair-identified THC exposure*age predicted worse episodic memory than in Controls (β=-0.60, p=.007), with no difference between CBD exposed and Controls. Data suggest those who use cannabis show likely pre-existing better cognitive performance during late childhood, with reduced improvement or flattened trajectories over time. These neurocognitive trajectories in youth (ages 9-17) who initiate cannabis use were demonstrated after accounting for within-person change and numerous known confounds and improving accuracy in identifying cannabis use through incorporating toxicological measures. Continued monitoring of this cohort will clarify cannabinoid-cognition relationships into young adulthood, including the impact of timing of cannabis use initiation.