Resting State EEG and Sensory Responsivity in ASD and Schizophrenia

Background:  Many individuals with autism spectrum disorder (ASD) report and display atypical sensory experiences and responses. These manifest heterogeneously: individuals experience both hyper- and hyposensitivity to different sensory modalities and react with both sensory-avoidant and sensory-seeking behaviors. Similar experiences are frequently reported by individuals with schizophrenia spectrum disorders (SZ) and are endorsed by many typical adults, suggesting that sensory features associated with ASD vary continuously throughout the population. The mechanisms underlying the heterogeneity of sensory profiles in individuals with and without ASD are poorly understood; the study of these mechanisms will inform basic understanding of sensory responsivity and guide clinical response to sensory sensitivities and behaviors that impact quality of life in multiple clinical populations.

Objectives:  This study examines (a) relationships between EEG power spectra and individual differences in sensory responsivity and (b) whether these relationships differ among individuals with ASD, SZ, and TD adults.

Methods: Resting EEG data were acquired from 33 adult participants (enrollment ongoing; ASD: n=18,13 male; SZ: n=10 9 male; TD: n=5,3 male) using a 128-channel net. Power spectra for delta, theta, alpha, beta, and gamma frequency bands were generated for frontal, temporal, parietal, posterior, and midline regions of the cortex from 2-second segments of EEG data. Sensory sensitivity and responsivity were evaluated using the Glasgow Sensory Questionnaire (GSQ), a 42-question self-report questionnaire that quantifies hypo- and hypersensitivity in 7 sensory modalities.

Results:  Principal components analysis identified a single component accounting for the majority of variance in the GSQ’s subscales, so preliminary analyses examined participants’ total scores on the measure. Multiple regression results showed that diagnostic group (ASD/SZ) and GSQ score were significant independent predictors of higher left and right temporal delta power (left: ASD, β=140.46, p<0.001; SZ, β=165.58, p<0.001; GSQ, β=4.96, p<0.001; right: ASD, β=129.01, p=0.005; SZ, β=148.3, p=0.007; GSQ, β=4.338, p=0.0012). While higher GSQ scores were associated with higher temporal delta power in TD controls, they were associated with lower delta power in participants with ASD or SZ (left: ASD:GSQ, β=-5.13, p=<0.001; SZ:GSQ, β=-4.95, p=0.001). This pattern was also seen in the right parietal region: delta power increased with GSQ score in TD controls, and decreased with GSQ score in participants with a diagnosis of ASD or SZ (GSQ, β=3.47, p=0.017; ASD:GSQ, β=-4.05, p=0.009; SZ:GSQ, β=-3.878, p=0.017). Midline gamma (30-80 Hz) power showed significant effects of schizophrenia diagnosis and GSQ score independently, as well as interactions between GSQ and diagnosis of either ASD or SZ (SZ, β=-3.08, p=0.013; GSQ, β=-0.07, p=0.01; ASD:GSQ, β=0.07, p=0.015; SZ:GSQ, β=0.096, p=0.004), suggesting that midline gamma power decreased with higher score on the GSQ in TD controls and increased with GSQ score in participants with SZ, staying constant in participants with ASD.

Conclusions: Neural oscillations in specific regions and frequency bands varied with sensory sensitivity and responsivity; associations between the power in these spectra and sensory sensitivity were modulated by diagnosis of ASD or schizophrenia in a direction opposite of that observed in TD participants. These results suggest common neurological mechanisms underpinning atypical sensory processing and response in ASD and SZ.