Somatosensory Gamma-Band Phase-Locking Deficits in Autism

Background: Autism spectrum disorders (ASD) are highly heritable. Although a range of behavioral changes has been observed in first-degree relatives, collectively termed the broad autism phenotype (BAP), fewer findings have been reported for biological markers. Previously, we have reported a deficit in auditory gamma-band phase-locking and evoked magnetic field power in persons with autism and their first-degree relatives. 

Objectives: In the current investigation, we were motivated to extend the previous auditory findings to the somatosensory system to examine whether the findings would be modality specific.

Methods:  Fifteen individuals with ASD and 16 healthy adult controls (HC) participated in the study. Magnetoencephalographic (MEG) recordings were made using a 248-channel MEG system while participants received passive vibrotactile stimulation to their index finger tips at 25 Hz. Source analyses were performed to derive left and right somatosensory cortical locations for the steady-state somatosensory evoked response between 20-30 Hz. Time-frequency analysis was conducted in source space to examine phase-locking and phase-locked, or evoked, power. Statistical analysis of the time-frequency results was conducted at all time-frequency bins, correcting for multiple comparisons using the false discovery rate at q < .05.

Results: In addition to steady-state somatosensory evoked power in the 20-30 Hz range, we observed significant evoked power in the 45-55 Hz range at the first harmonic of the stimulus frequency. In this higher frequency range, we observed a significant reduction in evoked power the left and right hemisphere responses for the pASD subjects relative to HC subjects (Figure 1). This was also observed for the phase-locking factor, a more direct measure of trial-to-trial phase-consistency. In the left hemisphere, evoked power reductions were restricted to the early portion of the steady-state response, but in the right hemisphere significant reductions were observed across the time-window of the stimulus.

Conclusions: Phase-locking errors are not restricted to the auditory domain. Reduced evoked, or phase-locked power, accompanied by reduced phase-locking factor, in the ASD group relative to the HC group, confirmed our hypothesis generalizing the deficits to the somatosensory system. Phase-locking problems may be a cortex-wide endophenotype in ASD. Mechanisms underlying higher frequency phase-locking should be explored in more detail in ASD.