Neural Correlates of Impaired Processing of Communicative Auditory Stimuli In Children with ASD

Background: Studies have demonstrated abnormal cortical processing of socially relevant auditory stimuli in adults with Autism Spectrum Disorder (ASD).  Specifically, it has been demonstrated that a putative voice-selective region in the superior temporal sulcus (STS) is hypoactive in response to vocal sounds in adults with ASD, suggesting a deficit in processing human vocal compared with non-vocal sounds.  However, research from our laboratory suggests that this STS region may actually be sensitive to voiced sounds emitted with communicative intent, rather than being sensitive to both communicative and non-communicative voiced sounds.  This raises the possibility that reduced STS activity observed in adults with ASD may reflect a deficit in recognizing the communicative intent underlying auditory stimuli, rather than a deficit in processing voiced versus non-voiced sounds.  However, few studies have examined STS activity in response to communicative compared with non-communicative sounds in individuals with ASD.

Objectives: Our objectives were 1) to replicate findings indicating that the STS is sensitive to communicative compared with non-communicative sounds in typically-developing children, and 2) to investigate whether children with ASD exhibit reduced sensitivity to communicative compared with non-communicative sounds.

Methods: To date, data have been collected from 8 participants (4 ASD) matched on age, gender, and verbal IQ as measured by the Differential Ability Scale.  Participants listened to the following sounds while in the fMRI scanner: human speech (affective and neutral), human communicative non-speech (e.g. laughing, sounds of agreement), human non-communicative voiced (e.g. coughing, sneezing), human non-communicative non-voiced (e.g. walking, snapping), rhesus monkey calls, and sounds of water.  Each condition was presented 5 times in a block design, each block lasting 20 seconds.

Results: Whole-brain analyses were performed to identify brain regions that are responsive to human communicative (speech and communicative non-speech) compared with non-communicative sounds (human non-communicative voiced, human non-communicative non-voiced, rhesus monkey calls, and water).  A random effects analysis with p<0.01 and k>10 revealed greater activation in the right STS to communicative compared with non-communicative sounds in typically-developing children.  Furthermore, a random effects ANCOVA with a p < 0.01 and k >10 revealed that children with ASD exhibited reduced sensitivity to communicative sounds in the right STS compared to typically-developing children.

Conclusions: We observed increased activation in the right STS of typically-developing children in response to human communicative compared to non-communicative sounds, supporting the idea that the STS is sensitive not to voiced sounds, but specifically to voiced sounds emitted with communicative intent.  This finding is consistent with research indicating that the STS is specialized for detecting and reasoning about social communicative stimuli in both visual and auditory domains.  Further, our data indicate that reduced STS activation in children with ASD may not simply reflect a deficit in differentiating voiced compared with non-voiced sounds, but rather a deficit in detecting auditory sounds produced with communicative intent.  This distinction illuminates the nature of previously reported auditory deficits in ASD, an important step towards understanding the language and social communicative deficits characteristic of ASD and developing optimal interventions.