Social Visual Engagement during Dyadic Interaction in Infants with and without ASD

Thursday, May 11, 2017: 12:00 PM-1:40 PM
Golden Gate Ballroom (Marriott Marquis Hotel)
I. Stallworthy1, S. Glazer2, P. Lewis1, A. Klin3, S. Shultz4 and W. Jones5, (1)Marcus Autism Center, Atlanta, GA, (2)University of Texas M.D. Anderson Cancer Center, Houston, TX, (3)Marcus Autism Center, Children's Healthcare of Atlanta & Emory University School of Medicine, Atlanta, GA, (4)Marcus Autism Center, Children's Healthcare of Atlanta, Emory University, Atlanta, GA, (5)Marcus Autism Center, Children's Healthcare of Atlanta and Emory University School of Medicine, Atlanta, GA

Infant-caregiver interactions are both the platform and the catalyst for subsequent development, as mutually-adapted infant and caregiver behaviors create cycles of contingency that scaffold infants’ emerging abilities and facilitate further interaction (Fogel & Thelen, 1987). Given the importance of contingent social interaction in guiding typical development, this study investigates whether infants with Autism Spectrum Disorder (ASD) are sensitive to two critical aspects of social experience—contingency and familiarity—within the context of face-to-face dyadic interactions. Identifying aspects of early social experience that are disrupted in ASD may provide an early diagnostic marker of the condition, and may also inform our understanding of the developmental processes through which early atypical experiences culminate in symptoms of ASD.


This study examines sensitivity to social contingency and familiarity in typically developing (TD) infants and those with ASD by measuring infant social visual engagement with: (1) a familiar social partner interacting with them contingently, (2) a familiar social partner interacting with them non-contingently, and (3) an unfamiliar social partner engaging with them non-contingently.


Participants were 2-5-month-old typically developing infants (typical outcomes confirmed at 24 months; n=14, mean age=4.40(0.23) months, 7 female), and 2-5-month-old infants with ASD (data collected prospectively, prior to diagnosis; diagnostic ascertainment performed at 24 months; n=8, mean age=4.33(0.20) months, 1 female). Eye-tracking and video data were collected monthly between 2 and 6 months while infants viewed: (1) a closed-circuit live video feed of their caregiver (familiar/contingent, Figure 1A); (2) pre-recorded clips of their caregiver (familiar/non-contingent); and (3) pre-recoded clips of unfamiliar actresses (unfamiliar/non-contingent) (Figure 1B). Regions of interest were coded in each video, and percentage of fixation time on eyes was compared between groups and conditions.


Consistent with prior analyses (Jones & Klin, 2013), eye-looking at this cross-sectional age point is present at levels that are similar between groups (p<0.51). In TD infants, eye-looking showed a trend towards increasing as a function of both contingency and familiarity: TD infants look most at the eyes when interacting contingently with their caregiver, less when viewing a non-contingent recording of their caregiver, and least when viewing pre-recorded non-contingent strangers (Figure 2A). By contrast, infants with ASD showed no modulation of eye-looking as a function of contingency (Figure 2B). They instead demonstrated a trend towards increased eye-looking when viewing familiar versus unfamiliar caregivers. Immediate next steps include further examination of these trends in a larger sample.


Preliminary findings suggest that, from the earliest months of life, sensitivity to social contingency may be disrupted in infants with ASD. While TD infants look more at the eyes of those who are engaging with them contingently, levels of eye-looking were not modulated by contingency in infants with ASD. Instead, visual engagement in infants with ASD may be more strongly influenced by the familiarity of their social partner. Future analyses will examine the emergence and refinement of sensitivity to social contingency longitudinally, and will include measures of reciprocal social smiling and time-varying mutual gaze as further indices of sensitivity to social contingency.