Atypical Social Attention Patterns in 6-Month-Old Infants Later Diagnosed with ASD During a Face-to-Face Dyadic Interaction

Background: Deficits in spontaneous orienting to naturally occurring social stimuli are one of the defining features of autism in early childhood.  Toddlers with autism spectrum disorders (ASD) have been consistently found to show impairments in social attention, but it is not clear if these deficits are already present in the first months of life of those affected by the disorder. 

Objectives: The main objective was to examine patterns of social attention during a face-to-face adult-child dyadic interaction in 6-month-old infants who were later diagnosed with ASD compared to infants who did not develop ASD.

Methods: Six-month-old infants (N=94) both at high risk (HR; N=61) and low risk (LR; N=33) for ASD participated in a standardized, live dyadic interaction as part of their participation in a prospective study. Children were assigned a clinical best estimate diagnosis of ASD (11 HR, 2 LR) or non-ASD (including typical development, 25 HR and 31 LR; or non-spectrum delays such as language delays, global developmental delays, 25 HR) in the 3^rdyear. The face-to-face interaction consisted of five one-minute episodes, graded from purely social interactions (experimenter talking to the infant using motherese, singing a nursery rhyme) to more physical interactions (peek-a-boo, showing a colorful toy, and a tickling game). Attention to various targets (e.g., experimenter’s face, experimenter’s body, parent, and objects) was coded offline by coders blind to group membership. Patterns of attention were compared using generalized linear mixed models (SPSS GLMM) to examine the effects of diagnostic group and episode.

Results: Infants’ targets of attention varied across the five episodes; all infants were able to regulate their attention according to the type of interaction presented. For example, all infants looked at the experimenter’s body (her hands) significantly longer during the peek-a-boo episode than during any other episode. However, across all episodes, infants later diagnosed with ASD looked less at the experimenter’s face (F=4.194, p<0.05) and at their parent (F=4.152, p<0.05), and more at objects (F=7.216, p<0.01) compared to infants without ASD. Pairwise comparisons revealed main effects of group for attention to face and attention to parent across all episodes. The ASD group attended more to objects than the non-ASD group during the motherese episode only.

Conclusions:   At 6 months, infants who were later diagnosed with ASD exhibited a unique pattern of social attention during face-to-face interactions with a stranger.  Compared to other non-affected HR and LR infants, 6-month-olds later diagnosed with ASD showed diminished attention to the social partner as well as to their own parent over the course of the 5-minute probe, and increased attention to objects in the environment, particularly while the examiner spoke to them using motherese.  Further investigation into the underlying mechanisms of these prodromal symptoms and their role in the etiology and manifestation of ASD symptoms in the first year of life is needed.