Reduced Cortical Response to Adjacent Finger Stimulation: Evidence for Local Underconnectivity In the Autistic Brain?

Background:

Putative structural and physiological differences in autistic versus typical brains must eventually manifest at the level of neural circuitry. Therefore, probing for differences in connectivity is a natural starting point for determining the neural phenotype. In this regard, proposals of local over-connectivity and long-range under-connectivity in the autistic brain are attractive candidates. Autism is a developmental disorder; therefore, abnormalities in the circuitry of autistic brains are likely to be pervasive and include regions involved in sensory processing. One such region is the somatosensory cortex. Afferents from adjacent fingers project via segregated neuronal clusters in the sub-cortical pathway and reach cortex, where the first neural interaction between them occurs. The cortical region to which the thalamic afferents corresponding to a given finger project is its hot spot, and the early part of its response to the stimulation of an adjacent finger assays the strength of local excitatory intracortical connectivity. 

Objectives:

Local overconnectivity predicts that the tactile stimulation of a finger would yield a stronger response in the cortical hot spot corresponding to its adjacent finger in autism versus control. We tested this prediction with magnetoencephalography (MEG).

 Methods:

(i) We recorded neural responses to the passive tactile stimulation of the thumb (D1) and index finger (D2) of the dominant hand of young adult participants (13 high-functioning persons with autism spectrum disorder or ASDs and 17 typically developing persons or TDs, matched for gender and age) while they remained awake in an eyes-closed supine posture in a 248-sensor MEG scanner.

(ii) For each participant, we computed the evoked response in source space using a mathematical tool called singular value decomposition (SVD). This method creates a virtual sensor that approximates the response of the cortical hot spot corresponding to a particular finger. In this manner, two hot spots, one each for D1 and D2, were obtained.

(iii) Somatosensory evoked potentials (SEPs) in the hot spot to the tactile stimulations of D1 and D2 were computed. For M50 (short-latency SEP, mediated mainly by feedforward sub-cortical pathways) and M100 (mid-latency SEP, mediated by cortical feedback) SEP components separately, responses of the neighboring/primary hot spots were measured and optimal least-squares slopes computed. The slopes of the two groups were then compared.

Results:

M50: No between-group differences were found in the ratios of D2/D1 hot spot responses to D1 stimulation or D1/D2 hot spot responses to D2 stimulation.

M100: There were clear between-group differences. D1 stimulation: The slope of the D2 hot spot’s response to the D1 hot spot’s response to D1 stimulation was significantly smaller in the ASD group compared with the TD group (P=0.003). D2 hot spot: The results were similar. The slope of the D1 to the D2 hot spots’ responses to D2 stimulation was again, significantly smaller in the ASD group (P=0.015).

Conclusions:

The findings indicate a weaker degree of local cortical interaction in the somatosensory cortex of the person with autism compared to the person without. Local underconnectivity in the autistic brain is consistent with our results.