EEG Correlates of Social-Communication Difficulties in Typically Developing Individuals: Functional Connectivity and Frequency-Domain Insights

Autism Spectrum Disorder (ASD) is characterized by persistent social communication challenges and repetitive behaviors. However, similar traits measured using self-report questionnaires such as the Autism-Spectrum Quotient (AQ) are also present at subclinical levels in typically developing (TD) individuals, a phenomenon known as the broader autism phenotype (BAP). To investigate the neurophysiological correlates of such traits, this study examined resting-state electroencephalography (EEG) data from 49 TD adults categorized into high AQ [subjects = 19, (AQ \(\:\ge\:\:\)26), (mean age: 25.53 \(\:\pm\:\) 6.18; 09 females, 10 males)] and low AQ [subjects = 30, (AQ\(\:\:\le\:\) 16), mean age: 25.3 \(\:\pm\:\) 6.70; 21 females, 09 males)] groups.Frequency-domain (FD) features and functional-connectivity (FC) metrics including phase locking value (PLV), weighted phase lag index (WPLI), and phase lag index (PLI) were computed across five frequency bands. Global network measures (GNMs), [global efficiency (GE), transitivity (T), modularity (MM), and assortativity (AC)] were also derived to assess large-scale brain dynamics. Group differences in FD and FC/GNM features were assessed with Welch’s t-tests; associations with AQ scores (continuous) were tested using multiple linear regression models adjusted for age and gender. Moreover, correlations between FD and FC/GNM EEG features and AQ subscales were evaluated using Pearson’s correlation.Group comparisons revealed significant group differences across most FD EEG features (except gamma). However, when AQ was treated as a continuous variable in multiple linear regression models, none of the FD features showed significant associations. Additionally, theta-band in GNM features derived from FC connectivity showed significant differences between high- and low-AQ groups and remained significantly associated with total AQ scores in multiple linear regression models. Furthermore, both FD and FC/GNM EEG features exhibited significant correlations with AQ subscales particularly social skills, attention to detail, and communication in the theta, delta, and gamma bands, core challenges faced by individual with ASD.Our results indicate that theta-band EEG activity reflects subclinical variations in social-communication abilities within the general population. The results support the continuum model of autism and highlight the potential of EEG-derived metrics as biomarkers for early identification of social-communication difficulties in TD individuals