Developmental Trajectory of the N170 Marker of Face Processing In Children with ASD

Background:  Face processing and its role in the social phenotype of autism spectrum disorders (ASD) has received much attention in the past decade. Behavioral studies of individuals with ASD consistently note impaired performance on basic tests of face memory and face processing during the early and mid-childhood years. Electrophysiological studies of face processing have noted developmental delays in the event-related N170 response to faces in young toddlers and children with autism (Dawson et al., 2002; Webb et al., 2006; 2010), but less obvious impairments in adults (Webb et al., 2009). The developmental trajectory of face processing in ASD has not been studied.

Objectives:  In a longitudinal study of children with autism spectrum disorders (ASD) at 3, 6 and 9 years of age, face processing and face memory were assessed using event-related potentials. This analysis will focus on the specialized face-processing component (the N170) at these three time points, examining individual and group change over time.

Methods:  Data were collected using a high density EEG array in children who met gold standard diagnostic criteria for ASD; comparison data from groups of age-matched children with developmental delay (DD) and neurotypical development (NT) were also collected. Participants watched pictures of a familiar and unfamiliar face (at 3 and 6 years) or a series of unfamiliar faces (at 9 years). Data were available for 24 3-yr old, 30 6-yr old, and 39 9-yr old children with ASD; of this sample, 15 children provided data at all 3 time points. Analyses examined the latency and amplitude of the N170. 

Results:  Preliminary analyses suggest that across the three time-points, the N170 response to faces in the ASD group was consistently slower than in the NT group; the DD group showed intermediate responses. However, the N170 became more negative and faster with development in all three groups; for example, from 6 to 9 years of age processing speed increased by 50 ms in both the ASD and NT groups. These findings suggest that children with ASD continue to show temporal face processing delays through early and middle childhood relative to children with neurotypical development. However, between age 3 and age 9 years developmental change in the speed of face processing proceeds at a relatively typical rate in children with ASD.

Conclusions:  In a passive face processing task, children with ASD demonstrate a temporal delay in the neural correlates of face processing that is evident by 3 years of age and is maintained across childhood. The rate of increase of face processing speed from 3 to 9 years was similar in the NT and ASD groups, suggesting that later maturation of basic-level face processing systems is relatively typical in children with ASD. Additional analyses will examine developmental trajectories of ERP responses to object and facial emotion from the same group of children.