Difference in Auditory Evoked Potentials in Children with Autism Spectrum Disorder Using Magnetoenecphalography

Background: Communication deficits are a core clinical feature for individuals with Autism Spectrum Disorders.  Consequently, understanding the integrity of the auditory cortex, beginning with early cortical activity in the superior temporal sulcus is essential. Evoked related potential studies of cortical processing suggest ineffective regulation of auditory sensory input and disruption in early hemispheric specialization.  While there are contradictory findings in peak latencies/amplitudes depending on technique, developmental age, and diagnostic variation, several studies have reported diminished amplitudes of the early cortical peak.

Objectives: To determine if children with autism have atypical early cortical response (latency and amplitude) to simple repeated and deviant tones relative to healthy controls.

Methods: We assessed children with ASD (autism n=5, PDD, NOS n=2, Asperger’s syndrome n=3) and 11 healthy controls. Subjects watched a silent movie while listening passively to an auditory paradigm: binaural presentation of a standard 1 kHz tone  repeated with an intermittent deviant 1.2 kHz tone as deviant.  Recordings were made using the whole head OMEGA 275-Channel Magnetoencepholography System (CTF Systems Inc.)  Stimulus locked responses to the pre-deviant, deviant and post-deviant tones were analyzed for each hemisphere. These responses were band pass filtered (1-40 Hz), averaged over all the presented trial and root mean squared over parietal and temporal sensors.  Two raters (KK and SN) choose M100 peaks. The latency and amplitude of these peaks were compared by diagnostic group using t-tests.  In addition, the conditional and group effects were assessed using a mixed linear effects model.

Results: In accordance with previous reports, we see a diminution in amplitude in the ASD group relative to controls.  This finding is limited to the right hemisphere but exists for the pre-deviant standard, the post-deviant standard and the deviant tone (all < p=0.01).  We see a conditional effect across groups with the deviant having a larger amplitude relative to the standard tones in both hemispheres (LH p=0.002, RH p=0.008). However, this difference is primarily driven by the ASD group.  The latencies also show a striking discrepancy between the ASD and HC groups with the ASD group showing earlier latencies in the left hemisphere (p<0.05).  Conditional comparison revealed shorter latencies in the deviant condition relative to the repeated tones.  This was also limited to the left hemisphere and largely dependent on the ASD group (p=0.04).

Conclusions: These data suggest atypical early cortical activity in children with ASD in response to repeated and novel tones. The early latencies may be related to heightened attention and auditory sensitivity that has been widely described in the literature.  A deeper explanation may be that the temporal cortex has impaired inhibitory input, creating a hyper-excitable or “noisy cortex.”  In this state, the cortex would be prone to early, yet poorly coordinated (low amplitude) responses. The cross hemispheric findings are intriguing and may suggest atypical specialization of tonal processing as has been suggested for language processing in ASD.