Spatial heterogeneity of the excitation-inhibition balance: correlation with sleep disturbances in children with Autism Spectrum Disorder

Autism Spectrum Disorder (ASD) is a neurodevelopmental condition characterized by clinical and neurophysiological heterogeneity. Among the proposed neurobiological models, both the excitation–inhibition (E/I) imbalance and the altered connectivity hypotheses have been extensively researched. However, when considered independently, neither theory provided a fully coherent explanation of the neurobiology of ASD. This retrospective study aimed to investigate whether spatial heterogeneity of the cortical E/I balance could represent a potential biomarker for ASD and explore its possible correlation with sleep disturbances. Clinical and electroencephalographic (EEG) data were retrospectively collected from 248 children with ASD and 135 typically developing (TD) controls admitted to the Regina Margherita Children’s Hospital in Turin between 2018 and 2024. A subgroup of 55 participants with ASD underwent a follow-up assessment including the Children’s Sleep Habits Questionnaire. EEGs were analysed using the FOOOF algorithm to extract the aperiodic exponent as a proxy for cortical E/I balance, and Moran’s I was computed to quantify spatial autocorrelation across electrodes during wakefulness and sleep. Results showed a significant reduction in spatial autocorrelation of the E/I ratio in ASD compared to TD individuals, supporting the concept that ASD is characterized not by an overall alteration in average brain E/I balance but rather by increased heterogeneity of the E/I ratio across different regions. Notably, no associations were found between E/I heterogeneity and symptom severity or sleep disturbances. These findings highlight the potential of spatial E/I heterogeneity as a core, measurable neurophysiological feature of ASD. Building on these results, a prospective longitudinal study is currently underway to validate spatial E/I heterogeneity as a reliable biomarker and to thoroughly investigate its relationship with sensory processing and multisensory integration.