Global cortical morphometry as a mediator of longitudinal associations between music participation and language outcomes in a population-based cohort

Associations between music participation and language outcomes have been widely reported, yet the extent to which brain structure statistically accounts for these relationships remains unclear, particularly at the population level. Using longitudinal data from the population-based Adolescent Brain Cognitive Development (ABCD) Study, we examined prospective associations between music participation at ages 9–10 and language outcomes two years later and tested whether global cortical morphometry accounted for these associations. The analytic sample included 5,993 children (baseline mean age 10.0 years; 47.1% female), including a twin subsample of 936 participants. Music participation was assessed via parent report of sustained participation, frequency, and intensity. Language outcomes were measured using NIH Toolbox assessments of picture vocabulary, oral reading recognition, and crystallized cognition. Cortical morphometry was quantified across multiple metrics, including global surface area, volume, thickness, and sulcal depth. Longitudinal associations were estimated using linear mixed-effects models, with mediation analyses conducted to quantify the indirect effect of cortical morphometry after adjusting for baseline language performance, sociodemographic factors, and relevant covariates. Sustained music participation was associated with higher crystallized cognition, picture vocabulary, and oral reading recognition scores two years later. Mediation analyses indicated that global cortical surface area and volume, but not cortical thickness or sulcal depth, statistically accounted for a modest proportion of these longitudinal associations (approximately 5–9%). Analyses of practice intensity showed weaker total associations with language outcomes but proportionally greater mediation by surface area and volume, whereas practice frequency exhibited minimal associations and no evidence of mediation. Sensitivity analyses in the twin subsample yielded qualitatively similar patterns, with larger mediated proportions observed for crystallized cognition and picture vocabulary. These findings suggest that global cortical morphometry explains a limited but reproducible component of the longitudinal association between music participation and language outcomes in population-based developmental samples.