Maturation of Face-Sensitive MEG Responses in 6-24 Months Old Infants

Background: Failure to detect or respond to faces is a predictor of autism spectrum disorder (ASD), with most research showing altered brain networks and abnormal fusiform gyrus (FFG) activity to face stimuli in adult or school-age children with ASD. Although orienting to face stimuli is an ability that emerges in the first year of life, little is known about how the brain processes social face stimuli in young populations.

Objectives: The present study employed magnetoencephalography (MEG) to study the development of face processing in infants and toddlers. The primary study goals were: (1) identify brain areas showing differences when processing Face versus Non-Face stimuli in typically-developing (TD) infants 6-12 months old (Time 1); (2) evaluate the developmental trajectory of face processing by obtaining MEG measures 12 months after the participants’ initial visit (Time 2; 18-24 months old). Associations between FFG peak latencies and age were also examined to evaluate whether the latency of the face FFG response decreases as a function of age.

Methods: MEG responses to Face and Non-Face stimuli were obtained from 35 infants aged 6-12 months. Follow-up MEG data (Time 2) were obtained from 12 TD infants 12 months after their initial visit. MEG data were obtained using an infant MEG system (Artemis 123™). MEG data were co-registered to a one-year-old MRI template. Artifact-free epochs 200 ms pre- to 500 ms post-stimulus were averaged according to stimulus type. Distributed source modeling via dSPM estimated activity throughout the brain at each digitized time point.

Results: A within-subject paired t-test showed stronger Face than Non-Face activity in multiple brain regions across time, with greater activity to Face than Non-Face stimuli most evident in right FFG at ~300 ms. Right FFG latencies in response to Face stimuli decreased as a function of age (R² = 0.43, p < 0.001). For the 12 infants with a follow-up MEG exam, the average right FFG latency difference between Time 1 and Time 2 was 110 ms (range 0 – 260 ms).

Conclusions: Face-sensitive brain activity was identified in TD infants in left and right FFG at ~300 ms post-stimulus, with right FFG activity occurring on average 110 ms earlier after a year interval. The above suggests that the face-sensitive FFG response (right hemisphere) evolves to an adult-like FFG face response (~170 ms post-stimulus) by 3 years of age. With a better understanding of face processing in infants, longitudinal studies are now needed to characterize neural responses to social stimuli during the first year of life in infants at-risk for ASD. It is expected that via such studies early brain markers that predict future (and possibly atypical) brain function will be identified, with these functional markers eventually used to identify children at risk for conversion to ASD.