Developmental Differences in N170 Morphology in Children with Autism Spectrum Disorder: Results from the ABC-CT Interim Analysis

Background: Individuals with autism spectrum disorder (ASD) exhibit impairments in face recognition that are associated with reduced neural efficiency reflected in increased peak latency of the N170, a face-sensitive event-related potential (ERP). However, the shape and morphology of these ERPs are highly variable, and N170 to social stimuli are bifid, i.e., showing two peaks, in 65% of typically developing (TD) children aged 4 to 12 years (Taylor, Batty, & Itier, 2014). Though researchers have acknowledged the variability in N170 waveform morphology, few have quantified it across development. Given the centrality of the N170 as a putative biomarker in ASD, investigating atypical waveform morphology is critical for understanding the neural mechanisms associated with social perception in ASD.

Objectives: To quantify relationships among N170 morphology, age, diagnosis, and clinical characteristics in children with ASD and TD controls.

Methods: ERPs to upright faces were collected from 172 participants (124 male, 115 ASD) 6 to 11 years of age (M = 8.72, SD = 1.64). Visual inspection was used to determine presence of bifid N170 morphology in the left hemisphere. A bifid was defined as two negative peaks between 100 and 350ms after stimulus presentation. Clinical measures are presented in Table 1.

Results: 29.13% (n = 67) of children with ASD exhibited bifid morphology; 33.33% (n = 37) of TD children. A chi-square test revealed that diagnosis did not predict presence of bifid morphology [χ²(2, N = 171) = 2.11, p = .348]. Independent-samples t-tests compared means of age and clinical measures between individuals with bifid morphology versus those without (Table 1). Older children were more likely to exhibit bifid morphology (M = 9.11, SD = 1.68) (M = 8.50, SD = 1.59) [t(168) = -2.33, p = .021]. Among children with ASD, children with bifid morphology had lower FSIQ (M = 92.33, SD = 16.66) than those without (M = 101.42, SD = 18.25) [t(113) = 2.60, p = .011] and higher ADOS severity scores (M = 8.21, SD = 1.77) (M = 7.36, SD = 1.90) [t(113) = -2.33, p = .022]. Bivariate correlations were performed to assess the relationship among variables with the time between bifid peaks (Table 2). Time between peaks was reduced in older children [r(127) = -.25, p = .004] and in children with higher NEPSY scores [r(127) = -.19, p = .029]. Furthermore, time between peaks was greater in children with later N170 latencies [r(127) = .26, p = .003] and reduced in children with more negative N170 amplitudes [r(127) = -.31, p < .001].

Conclusions: Results show that bifid morphology of N170 waveform associates with demographic and clinical characteristics. Reduced time between bifid peaks correlating with age and face processing supports the idea that the maturational time course of the N170 may reflect increasing efficiency of a specific face-processing system. These data hold significant promise for understanding the neural mechanisms guiding the development of social performance and face processing, and increasing the utility of the N170 as an effective biomarker for diagnosis and stratification of ASD.