Biological motion stimuli reveals the severity of the social deficits of ASD: an EEG study.

The integration of individuals with autism spectrum disorder (ASD) into society and their capacity to lead fulfilling lives are significantly hindered by challenges in social communication. Consequently, it is crucial to explore the underlying neural mechanisms and establish early diagnostic approaches to accurately assess the severity of social interaction impairments in ASD. To investigate the social deficits in ASD, particularly in identifying reliable biomarkers for predicting the severity of social impairments, the current study focused on the sample entropy, functional connectivity, and network properties under biological motion tasks (biological motion (BM) and scramble motion (Scr) in both ASD and typically developing (TD) children. The findings suggest that, compared to TD, those with ASD exhibit higher sample entropy in localized brain regions, specifically the frontal and occipital lobes, regardless of whether they are under BM or Scr conditions. Moreover, ASD is characterized by enhanced long-range connectivity involving the frontal, parietal, and occipital lobes. These results collectively highlight the abnormal neural mechanisms of ASD when engaging with BM tasks, which further found a significant correlation between the network properties and the ADOS social score. Notably, by utilizing network properties and sample entropy as features, the severity of social impairments in autism can be effectively predicted through multiple stepwise regression analyses. These findings illuminate the pathophysiological mechanisms of ASD's social deficits from both local and global perspectives, offering potential biomarkers for quantifying social dysfunction in autism.