Evidence for Domain Specificity of Cortical Auditory and Somatosensory Response Delays in ASD

Background: Cortical auditory response delays are a well-replicated finding in individuals with Autism Spectrum Disorder (ASD) and have been associated with language deficits. It is unclear, however, whether these delays are specific to auditory processing, are present in other sensory domains, or are associated with generalized cortical latency delays. It is also unclear whether these delays are present in other groups who experience sensory dysfunction but do not meet other symptom criteria for ASD.

Objectives: to determine whether cortical sensory response delays are specific to ASD and whether the latency delays previously reported in ASD are specific to auditory processing.

Methods: Auditory and somatosensory cortical response latencies were compared in 8-12 year-old male participants with ASD (N=18), those with sensory processing dysfunction (SPD) who do not meet criteria for ASD (N=13), and in typically developing control (TDC) participants (N=19). Auditory and somatosensory latencies were measured via MEG evoked responses to standard stimuli presented in two passive mismatch negativity tasks (i.e., response to a 1000Hz tone for the auditory evoked response and application of a tap to the left middle fingertip for the somatosensory evoked task). Measures of communication (i.e., phonemic processing, language pragmatics, and verbal intellectual abilities) were administered to examine associations between auditory processing delays and language deficits associated with ASD.

Results: ANCOVAs were performed (age as covariate) to determine if groups differed in M100 or M200 auditory response latency in each hemisphere, or contralateral somatosensory response. Significant group differences were identified for the left auditory M200 latency, F(2,45)=3.61, p=.035, and right somatosensory latency F(2,46)=3.63, p=.035. Specifically, the ASD left M200 latency was significantly delayed relative to both the TDC and SPD groups, whereas the somatosensory response in the ASD group was only delayed relative to the TDC group and did not significantly differ from the SPD group, who presented with an intermediate somatosensory latency. Likewise, response latencies for the M100 response followed a similar pattern to the M200, with the clinical groups showing longer mean latencies bilaterally, although group differences were not statistically significant. A hierarchical regression analysis performed with nonverbal intelligence entered in Step 1, language variables entered in Step 2, and left M200 latency as the dependent variable demonstrated that language abilities accounted for 60.3% of the variance in left M200 latency after controlling for the effects of nonverbal intelligence (which accounted for 5.7% of the variance) in the ASD group, with F(3,10)=5.91, p=.014. These associations were not significant in the SPD or TDC groups. Finally, the cortical auditory M200 response latency delays detected in the ASD group were not significantly associated with delayed somatosensory latencies (r=.172, p=.494).

Conclusions: These results provide evidence of auditory and somatosensory response delays in ASD, and support prior findings demonstrating associations between delayed auditory latencies and language abilities. Further, while these cortical sensory responses were delayed in ASD compared to SPD and TDC groups, the weak association between these delays across sensory domains suggests that these deficits are domain specific rather than reflective of generalized cortical processing delays in ASD.