Electrophysiological Response to Words In Infants at Risk for ASD

Background: In typical development, the developing brain responds to linguistic stimuli in an increasingly focused and lateralized manner as infants gain experience with using language (Mills et al., 2005).  However, converging research suggests that individuals with ASD commonly show atypical lateralization of language networks in the brain such as reduced or even reversed asymmetry of frontal language areas.  The high prevalence of language delay or impairment in ASD can make it difficult to determine the aspects of this atypical response that are symptomatic of ASD specifically and which more closely related to language ability. Here, we compare infants at risk for ASD (HRA) against low-risk controls (LRC) who are in the process of acquiring language in order to tease apart the effects of language and ASD symptomology on brain response.

Objectives: To investigate whether infants at risk for ASD show atypical neural response to words and, if so, to examine whether this response is related more closely to language ability or to more general traits indicative of an ASD endophenotype.

Methods: As part of a larger longitudinal study, we recorded event related potential (ERP) in response to words at 18 months in HRA infants (n=18) and low-risk control infants (LRC; n=8).  Infants listened to a stream of words that are either understood or not understood by the majority of 18-month-olds.  To characterize language and ASD status, behavioral and parent report measures of language ability and ASD symptomology are collected at 12 and 18 months.

Results: Preliminary results revealed the negative response between 600 and 1000ms over frontal and temporal/central electrode sites (N600) that was faster to known words that to unknown words over the left hemisphere (condition by hemisphere interaction: F(1,24)=4.91, p=.036). This was qualified by a trend suggesting a three-way interaction (p=.09) such that the differential N600 response to known versus unknown words may have been restricted to the left hemisphere of the low-risk group.  Additional analyses are in progress and are focused on examining whether this pattern of lateralization is more closely related to language ability or to ASD symptomology/group status.

Conclusions: Our preliminary results suggest a possibility of different patterns of N600 response to comprehended words in HRA and LRC infants.  Additional investigation of language ability will help us to more closely determine whether the difference in response is more closely related to risk status for ASD or to experience with language.  Future work will ultimately look longitudinally to examine the predictive value of language ability in earlier infancy on the pattern of ERP response to words at 18 months.