Developmental Trajectories in Attention to Socially Relevant Information Differs for Infants at Risk for ASD

Background:   Preferential attention to biological motion and faces are socially adaptive, phylogenetically preserved behaviors that appear early in typical development. Attention to these types of socially relevant information is not only compromised among individuals with autism spectrum disorders (ASD), but also independently associated with autism symptomatology. One theory postulates that communicative cues provided by the face (e.g., gaze, facial expressions, and visual speech) constitute a special class of biological motion. Here, we examine that possibility by evaluating the developmental trajectories of visual social attention to biological motion and faces among infants at high (HR) and low risk (LR) for ASD.

Objectives:   This study investigates the developmental progression of visual attention to biological motion and faces of people engaging in social interactions among HR and LR infants. We aim to quantify the extent to which visual attention to various types of socially relevant information serve as converging indices of social functioning.

Methods:   Eye movements of HR and LR infants (HR: n = 17; LR: n = 10) were recorded at 3, 6, 9, and 12 months of age while infants engaged in two free-viewing tasks. The stimuli for the biological motion task consisted of side-by-side panels featuring colored circles moving in what appears to be a chase in one panel and random movement in the other. Relative time spent looking at the chasing scene was measured. Infants also viewed a video excerpt of Sesame Street. Visual social attention for communicative cues was operationalized as relative time spent looking at faces during social interactions versus faces in a crowd. At 12 months, the Autism Observation Scale for Infants (AOSI) was administered.

Results:   Linear mixed models evaluating the effects of risk and age for each task were conducted. While the developmental trajectories in attention for the two tasks did not differ between LR and HR infants, the first derivative of best fit lines revealed that the rate of developmental change for face-looking was significantly slower for HR infants (F_1,64.47 = 3092.32, p < 0.001). An additional model including attention to biological motion as a covariate for predicting developmental changes in face-looking revealed a significant interaction. Visual attention for socially relevant information was only associated among LR infants (F_1,24.79 = 4.36, p = .047). Bivariate correlations revealed that attention to biological motion was only associated with AOSI scores for HR infants (r = .53, p= .04).  

Conclusions:   The data suggest that HR infants process socially relevant information differently than LR infants. The stunted developmental rate of change in monitoring social engagements may cascade into social difficulties for infants who develop ASD. The association between AOSI scores and attention to biological motion found only among HR infants may be attributed to a greater heterogeneity in social and cognitive development within this group. Some HR infants may not glean the social relevance behind biological motion. The data inform the utility of examining associations between types of social attention to identify endophenotypes of autism symptomatology specific to social communication.