An Event-Related Potential Study of Familiar Face Processing In Autism Spectrum Conditions

Background:  

Clinical reports and behavioural studies of face perception in autism spectrum conditions suggest that the recognition of familiar faces is impaired. However, there have been inconsistent findings from functional brain imaging studies of familiar face perception in autism spectrum conditions. The N250 event-related potential component is a negative going wave maximal over occipito-parietal areas which peaks between 230 and 320 milliseconds after the presentation of a stimulus. Previous studies in the typical population have found that the N250 is larger to familiar faces than unfamiliar faces. Hence, this component is a means for investigating the processing of face familiarity in autism spectrum conditions.

Objectives:  

The current study aimed to investigate the acquisition of new face representations in autism spectrum conditions. Specifically, it was hypothesised that people with autism spectrum conditions would show a decreased N250 to a previously unfamiliar target face that was repeated and so became familiar during the course of the experiment. Thus, this paradigm differed from previous ERP and fMRI studies of face familiarity in autism spectrum conditions by directing the attention of participants toward the relevant face.

Methods:  

A sample of 15 adults with autism spectrum conditions and 15 age and intelligence quotient matched typical controls were shown a single, previously unfamiliar face and asked to remember it. This face was then inserted as the target into a randomised odd-ball sequence with seven non-target faces. During the odd-ball sequence, each face was repeated 100 times and participants were asked to indicate with a button press whether each face was the target or one of the non-target faces.

Results:  

The autism spectrum conditions and typical control groups were comparable in their behavioural performance for identifying the target and non-target faces. Consistent with the N250 literature in the typical population, the N250 was larger for the target face than the non-target faces across both participant groups. And, consistent with the hypothesis for this experiment, there was an interaction between participant group and face type for the N250 in which the autism spectrum conditions group showed a smaller N250 component to the target face than the typical control group but a similar N250 to the non-target faces.

Conclusions:  

The reduced N250 for the target face in the autism spectrum conditions group suggests that the development of face familiarity, indexed by the N250, is impaired in autism spectrum conditions. This is consistent with clinical accounts and behavioural studies that have found impaired recognition of familiar faces in autism spectrum conditions. That a decreased N250 was found in this study but not in previous studies of face familiarity in autism spectrum conditions may be because this paradigm deliberately directed the attention of participants toward the relevant face. This suggests that the impairments in familiar face processing found in autism spectrum conditions may be due to decreased attentional modulation of individual face representations.