Building Functional Connectivity Neuromarkers of Behavioral Self-Regulation in Children with and without Autism Spectrum Disorder

Abstract

Children with Autism Spectrum Disorder (ASD) are known to struggle with behavioral self-regulation, which associates with greater daily-life challenges and an increased risk for psychiatric comorbidities. Despite these negative outcomes, little is known about the neural expression of behavioral regulation in children with and without ASD. Here, we examined whole-brain linear associations between brain functional correlations (FC) and behavioral regulation through connectome predictive modelling (CPM), a data-driven protocol for developing predictive models of brain-behavior relationships from data, assessing neuromarkers using cross-validation. Using data from two sites of the ABIDE II dataset comprising 276 children with and without ASD (8-13 years), we identified functional brain networks whose FC predicted individual differences in two, of three, behavioral regulation subdomains. These distributed network models predicted novel individuals inhibition and shifting from resting-state FC data both in a leave-one-out, as well as split halves, cross-validation. We observed commonalities and differences in the functional networks associating with these subdomains, with inhibition relying on more posterior networks, shifting relying on more anterior networks, and both involving regions of the DMN. Our findings present a substantial addition to our knowledge on the neural expressions of inhibition and shifting across the spectrum of children with and without ASD, demonstrating the utility of this trans-diagnostic modelling approach. Given the numerous cognitive and behavioral issues that can be quantified dimensionally in neurodevelopmental disorders, further refinement of whole-brain neuromarker techniques may thus pave a way for functional neuroimaging to meaningfully contribute to individualized medicine.