Auditory Processing Differences in Autism Spectrum Individuals with and without Language Delay: An fMRI Study

Background: Autism spectrum (AS) individuals exhibit widely differing language developmental trajectories.  While some individuals show a delay in language acquisition, echolalia, stereotyped language and pronoun reversal, others do not. Delays in language acquisition might be associated with atypical development of the connections among the cortical regions involved in auditory and language processing.

Objectives: The present objective was to test if developmental language acquisition differences in AS individuals may be related to differences in auditory cortical processing of sounds of varying spectral and temporal complexity.

Methods: Using a 3T MRI system, we studied 27 AS and 13 typically developing (TYP) individuals while they listened to eight auditory stimuli constructed by crossing two carrier signals (pure and harmonic tones) with four levels of modulation depth (0, 25, 50 and 100%). Participants were asked to detect the presence or absence of the modulations. In order to investigate the association between the timing of language acquisition and auditory system development, language delay (first words after 24 months/first sentences after 33 months) was used as a grouping variable. The three groups were matched on IQ (FSIQ, VIQ and PIQ), age (mean: 22yr 9mo), sex and manual preference.  The Language Delay (LD) and Non Language Delay (NLD) groups were also matched on their ADI scores (Social, Communication and Repetitive behaviours). Activation patterns within auditory and language specific brain regions, including the transverse temporal gyrus (TTG), superior temporal gyrus (STG) and inferior frontal gyrus (IFG) were examined for the TYP, NLD and LD groups.

Results: There were no between-group task performance differences in response time and accuracy. For both the pure and harmonic carrier signal tones, the linear LD>NLD>TYP contrast revealed more primary auditory cortex and IFG (BA 45/47) activity. The TYP>NLD>LD contrast showed greater activity in non-primary auditory areas (STG and middle temporal gyrus, BA22) and in the IFG (BA 44).  There were no between-group differences in spectral complexity processing as examined in the harmonic vs. non-harmonic condition contrast. Non-primary auditory areas were more sensitive to temporal complexity increases (modulation depth) in the TYP group while the reverse contrast (LD>NLD>TYP) revealed no significant activity.

Conclusions: These results indicate that auditory processing of complex stimuli more strongly engages primary auditory areas in the LD group relative to the NLD and TYP groups while non-primary auditory areas and areas outside the auditory cortex are engaged less strongly. The level of task-related activity in the NLD group was intermediate between the LD and TYP groups. This suggests that the differential atypical low-level auditory processing in the LD and NLD groups may follow the clinical distinction made between autism and Asperger syndrome. At this stage however, it is not possible to determine if the observed differences in auditory processing are a cause or an effect of variable language delay.