Recognition of Familiar Faces In Infants at Low and High Risk for Autism

Background:  Ability to remember and recognize faces is important for successful social functioning (Schultz et al., 2005). Children and adults with ASD are often reported to have difficulty processing faces, including poor recognition of familiar faces (Boucher et al., 1998; Dawson et al., 2002; Klin et al., 1999). 

Objectives:  The purpose of this study was to investigate whether infant siblings of children with autism process familiar and novel faces differently from typical infants and whether sensitivity to face familiarity is associated with infants’ social and communicative behaviors.

Methods:  Visual event-related potentials (ERPs) were recorded in 36 infants, age 9 months +/- 15 days (20 infants with no family history of autism and 16 infant siblings of children with autism, sibs-ASD). Infants repeatedly viewed photographs of their mother’s smiling face and on 30% of the trials, a smiling face of an unfamiliar female (varied across participants) was presented. In addition, eye tracking data were recorded in a subset of infants (10 sibs-ASD, 14 TD infants) in response to a different stranger face (same for all infants). Mothers of infant participants completed Receptive and Expressive Communication, and Interpersonal Relationships subscales of VABS-II.

Results:  The two infant groups did not differ in receptive communication or interpersonal relationships scores, but sibs-ASD had significantly lower expressive communication scores (albeit within the typical range).

ERPs revealed that infants in both groups differentiated between their mothers and strangers as reflected in amplitude modulations of face-specific posterior N290/P400 as well as general novelty-sensitive frontocentral Nc and temporal PSW responses. Group differences were present only in the latency of the P400 response, where shorter latency for the mother’s than stranger’s face was observed in typical infants but not in sibs-ASD. When entered into a logistic regression predicting risk group membership, latency of the P400 to the stranger’s face improved classification accuracy for sibs-ASD (81.3% correct) compared to the classification model based on expressive communication scores alone (68.8% correct). There was no change in classification accuracy for the TD infants (85% correct).

Eye tracking data indicated no group differences in the number or duration of fixations on the stranger’s face or any of its features. However, for sibs-ASD and TD infants combined, increased number of fixations on the mouth area of the stranger’s face was associated with smaller amplitudes of the P400, Nc, and PSW responses to the stranger’s face.

Similarly, in the combined sample, shorter Nc latency to mother’s face was associated with better VABS-II interpersonal relationships V-scores (r=-.382, p=.021).

Conclusions:  Nine-month-old infants at low and high risk for autism utilize similar face scanning strategies and comparable brain mechanisms when viewing familiar faces. However, typical infants appear to be faster at detecting familiar faces than are sibs-ASD. Individual differences in the speed of detecting familiar faces are associated with stronger interpersonal skills and may therefore be an informative marker of elevated risk for autism.