Social Cognition in Autism: Predicting Task-Based Neural Activity from Resting State Connectivity

Understanding Social Cognition (SC) in Autism Spectrum Disorder (ASD) is challenging due to difficulties in obtaining task-evoked fMRI data. This study developed a predictive model using resting-state fMRI (rs-fMRI) to estimate task-related brain activity in ASD. We trained an Activity Flow Mapping (AFM) model on Typically Developing Controls (TDC) individuals from the Human Connectome Project (HCP) and applied it to ASD rs-fMRI data from the Autism Brain Imaging Data Exchange (ABIDE). While predicted ASD brain activity was not significantly different from social cognition task activations, key social brain regions showed deviations linked to ASD symptoms. Despite limitations, this method demonstrates the potential of using rs-fMRI to predict ASD social task-evoked neural activity, offering a flexible, non-invasive method for future research, diagnosis, and interventions.