• Diffusion Tensor Imaging;
  • Adolescence;
  • Mean Diffusivity;
  • Marijuana;
  • Alcohol


The influence of repeated substance use during adolescent neurodevelopment remains unclear as there have been few prospective investigations. The aims of this study were to identify longitudinal changes in fiber tract integrity associated with alcohol- and marijuana-use severity over the course of 1.5 years.


Adolescents with extensive marijuana- and alcohol-use histories by mid-adolescence (= 41) and youth with consistently minimal if any substance use (= 51) were followed over 18 months. Teens received diffusion tensor imaging and detailed substance-use assessments with toxicology screening at baseline and 18-month follow-ups (i.e., 182 scans in all), as well as interim substance-use interviews each 6 months.


At an 18-month follow-up, substance users showed poorer white matter integrity in 7 tracts: (i) right superior longitudinal fasciculus, (ii) left superior longitudinal fasciculus, (iii) right posterior thalamic radiations, (iv) right prefrontal thalamic fibers, (v) right superior temporal gyrus white matter, (vi) right inferior longitudinal fasciculus, and (vii) left posterior corona radiata (ps < 0.01). More alcohol use during the interscan interval predicted higher mean diffusivity (i.e., worsened integrity) in right (< 0.05) and left (= 0.06) superior longitudinal fasciculi, above and beyond baseline values in these bundles. Marijuana use during the interscan interval did not predict change over time. More externalizing behaviors at Time 1 predicted lower fractional anisotropy and higher radial diffusivity (i.e., poorer integrity) of the right prefrontal thalamic fibers (< 0.025).


Findings add to previous cross-sectional studies reporting white matter disadvantages in youth with substance-use histories. In particular, alcohol use during adolescent neurodevelopment may be linked to reductions in white matter quality in association fiber tracts with frontal connections. In contrast, youth who engage in a variety of risk-taking behaviors may have unique neurodevelopmental trajectories characterized by truncated development in fronto-thalamic tracts, which could have functional and clinical consequences in young adulthood.