Sensitivity to Social Contingency in High- and Low-Risk Infants During the First Six Months of Life

Background: Typically-developing infants are attuned to the social aspects of their environment, even during the first few months of life. From birth, infants look more at faces and preferentially engage with faces that gaze directly at them.  By 3 months, infants are sensitive to the gaze direction of others. In previous research, between 2 and 3 months of age, typically-developing infants also develop sensitivity to social contingencies. Reciprocal social interactions, such as mother-infant face-to-face interactions, play a key role in development of communication and social understanding. Detecting social contingencies is a key milestone in furthering social competency.  

Objectives: This experiment is intended to test the hypothesis that changes in visual scanning of infants during face-to-face interactions can act as an indicator of an infants’ level of sensitivity to social contingencies. We aim to investigate if differences exist in the level of sensitivity and/or timing of the development of sensitivity to social contingency in typically-developing infants and infants who at high-risk for ASD.  

Methods: Using eye-tracking technology, we compared the visual scanning of infants enrolled in a longitudinal prospective study of infant siblings of children with autism spectrum disorder (ASD). Infants at high-risk for ASD (HR-ASD) had a full sibling with a confirmed diagnosis of ASD, whereas infants at low-risk (LR-TDx) had no siblings with, or family history of, ASD. Between ages 2 to 6 months, visual scanning was compared between the two groups during three conditions: watching a videotaped actress (condition 1), participating in face-to-face interaction with their caregiver (condition 2), and a pre-recorded, thus non-contingent, video of the infant’s caregiver recorded during a previous session (condition 3).  Dependent measures included each infant’s level of fixation on different regions of interest, as well as the timing and duration of those fixations during each experimental condition.  

Results: Preliminary data analyzed using a sub-sample of 20 infants (8 LR-TDx and 12 HR-ASD) with a mean age of 3.46 (1.17) months, indicate that during the first 6 months of life infants are sensitive to social contingencies. Infants showed increased levels of fixation during contingent videos (condition 2) as opposed to non-contingent videos (conditions 1 and 3), t(19) = 3.98, p = .001. In addition, infants responded differently while participating in face-to-face interactions than while watching pre-recorded videos of an actress or their own caregiver. Our results demonstrate a significant interaction between video type and region of interest (p = .005). Post hoc analyses indicate that this effect is driven by increased eye vs. mouth fixation (F(1,19) = 13.69, p = .002) and decreased mouth vs. body fixations (F(1,19) = 5.18, p = .035) during the face-to-face interactions.  Preliminary results showed no difference in our LR-TDx and HR-ASD sub-sample.

Conclusions: Preliminary results suggest that changes in the visual scanning patterns of infants can effectively index infant sensitivity to social contingencies. This experimental paradigm is likely to potentiate between-group differences relative to infants at-risk for autism, thus increasing the utility in detection of early deviations from the normative course of social development.