Dynamic Inter-Modality Source Coupling Reveals Sex Differences in Children based on Brain Structural-Functional Network Connectivity: A Multimodal MRI Study of the ABCD Dataset

Abstract

Background: Sex differences in brain development are well-documented, yet the dynamic coupling between structure and function remains underexplored. We introduce dynamic inter-modality source coupling (dIMSC), extending our previous work to link structural MRI source-based morphometry (SBM) with dynamic functional network connectivity (dFNC). Methods: We used data from 500 participants in the Adolescent Brain Cognitive Development (ABCD) study (ages 9-11) and combined SBM-derived gray matter sources with sliding-window dFNC. dIMSC was computed as the time-resolved cross-correlation between these modalities to quantify structure-function coupling strength. We evaluated sex differences in these coupling profiles and used a cognitive moderation analysis to validate the sensitivity of the identified coupling states. Results: Significant sex-specific patterns emerged: males exhibited stronger positive coupling in sensorimotor regions (postcentral gyrus), while females showed stronger coupling in higher-order associative regions (inferior parietal lobule). These configurations were functionally distinct: higher positive coupling predicted better crystallized cognition (vocabulary) in females, whereas it predicted better fluid cognition (working memory) in males. Conclusion: These findings validate the functional sensitivity of dIMSC, demonstrating that the time-resolved coupling states it identifies are neurobiologically meaningful and differentially predictive of cognitive performance as a function of biological sex, consistent with known sex differences in neurodevelopmental during late childhood.