Executive Functioning as a Mediator of Relations Between Neural Structure and General Psychopathology in Pre-Adolescents

Abstract

Dimensional models of psychopathology conceptualize mental illness as a spectrum of manifest symptoms representing latent, higher-order domains. Evidence suggests that a general ‘p’-factor, which captures shared variation among all psychopathology symptoms has relations both to neural structure and to executive functioning (EF). However, a large portion of this work has been cross-sectional and cannot establish causation. Therefore, the goal of this study was to test longitudinal mediation of relations between neural structure and general and specific psychopathology factors via EF. Data from 9807 preadolescents (aged 9-10 at baseline) from the Adolescent Brain Cognitive Development study were utilized. Neural structure was captured via structural MRI scans at baseline, and EF was divided into inhibitory control (IC) and working memory (WM) tasks measured at two-year-follow-up. Psychopathology at three-year-follow-up was modeled using a bi-factor confirmatory factor model previously validated within the ABCD dataset. Longitudinal mediation of 87 cortical and subcortical gray matter volume (GMV) regions was conducted utilizing structural equation modeling (SEM) controlling for age, sex, total intracranial volume, and image quality. Results indicated that WM was significantly associated with all psychopathology factors, whereas IC was only associated with the internalizing, neurodevelopmental, and general factors. Additionally, compared to IC, WM was a stronger mediator of relations between a wider range of GMV regions and the psychopathology factors. Indirect effects were negative for the externalizing, neurodevelopmental, and general factors, but were positive for the internalizing, somatization, and detachment factors. This study provides evidence that EF may represent a mechanistic link between altered neural structure and later psychopathology among pre-adolescents. Future work should investigate additional EF processes and include alternative models of psychopathology.