Atypical Age-Related Patterns of Neural Signal Complexity in Autism Spectrum Disorder.

BACKGROUND: Autism spectrum disorder (ASD) is increasingly conceptualized as a neurodevelopmental condition involving atypical organization of large-scale neural dynamics. While EEG studies have largely focused on spectral power and connectivity, how neural signal complexity is organized across development in ASD remains poorly understood. METHODS: Resting-state EEG was recorded from 186 children aged 3-11 years, including children with ASD and typically developing (TD) peers. Participants were further categorized into preschool-aged and school-aged groups to examine developmental differences. Multiscale entropy (MSE) was computed across 20 temporal scales to characterize neural signal complexity, and group differences, age-related patterns, and associations with autism symptom severity were examined. RESULTS: Across the full sample, children with ASD exhibited significantly higher MSE than TD children across brain regions and temporal scales. Group differences were most pronounced in preschool-aged children, whereas in school-aged children they were more restricted to lower temporal scales and frontoparietal regions. In TD children, MSE increased robustly with age across scales, while age-related associations in ASD were more variable and region-specific. Additionally, individual differences in MSE were associated with autism symptom severity, particularly in frontal and parietal regions. CONCLUSION: These findings indicate that ASD is characterized by atypical developmental organization of neural signal complexity across temporal scales, underscoring the importance of a multiscale developmental perspective for understanding neural dynamics in autism.