Magnetoencephalography and electroencephalography can both detect differences in cortical responses to vibrotactile stimuli in individuals on the autism spectrum.

Autism Spectrum (AS) is defined primarily by differences in social interactions, with impairments in sensory processing also characterizing the condition. In the search for neurophysiological biomarkers associated with traits relevant to the condition, focusing on sensory processing offers a path that is likely to be translatable across populations with different degrees of ability, as well as into animal models and across imaging modalities. In a prior study, a somatosensory neurophysiological signature of AS was identified using magnetoencephalography (MEG). Specifically, source estimation results showed differences between AS and neurotypically developing (NTD) subjects in the brain response to 25-Hz vibrotactile stimulation of the right fingertips, with lower inter-trial coherence (ITC) observed in the AS group. Here, we examined whether these group differences can be detected without source estimation using scalp electroencephalography (EEG), which is more commonly available in clinical settings than MEG, and therefore offers a greater potential for clinical translation. To that end, we recorded simultaneous whole-head MEG and EEG in 14 AS and 10 NTD subjects (age 15-28 years) using the same vibrotactile paradigm. Based on the scalp topographies, small sets of left hemisphere MEG and EEG sensors showing the maximum overall ITC were selected for group comparisons. Significant differences between the AS and NTD groups in ITC at 25 Hz as well as at 50 Hz were recorded in both MEG and EEG sensor data. For each measure, the mean ITC was lower in the AS than in the NTD group. EEG ITC values correlated with behaviorally assessed somatosensory sensation avoiding scores. The results show that information about ITC from MEG and EEG signals have substantial overlap, and thus EEG sensor-based ITC measures of the AS somatosensory processing biomarker previously identified using source localized MEG data have a potential to be developed into clinical use in AS, thanks to the higher accessibility to EEG in clinical settings.