Caregiver Viewing Patterns during Infant-Caregiver Dyadic Interactions

Background: The first six months of life constitute a uniquely important period of growth and development for human infants, during which normative social development in typically developing (TD) infants is scaffolded and facilitated by highly contingent and mutually-reinforcing interaction with caregivers. In order to create and sustain high degrees of contingency, caregivers and infants must be closely attuned to one another’s behavior on a moment-by-moment timescale. Caregivers, for example, must allocate their visual resources to maximize perception of developmentally important infant cues, such as smiling or the initiation of eye contact, in order to optimally respond to these signals and thereby elicit further signaling. In doing so, caregivers promote future opportunities for learning via continued social interaction. Infants later diagnosed with autism spectrum disorder (ASD), however, show a reduction in shared smiles and mutual eye gaze, behaviors that would otherwise serve as important signals to caregivers (Stallworthy et al., 2016; Ozonoff et al., 2009). These differences in infant signaling may then alter or disrupt social contingency in the infant-caregiver dyad, in turn influencing caregiver behavior. This study seeks to investigate whether subtle differences in the behavior of infants later diagnosed with ASD are reflected in changes in caregiver viewing patterns.

Objectives: To examine and quantify caregivers’ visual scanning of infant faces during dyadic interactions with TD infants and infants with ASD.

Methods: Eye-tracking data were collected from caregivers and their 2- to 6-month-old infants during three 30-second interactions via closed-circuit live video feed. Diagnostic evaluations at 24 and 36 months were used to separate participants into two groups: caregivers whose infants received an ASD diagnosis at outcome (n=6), and those whose infants were TD and had no family history of ASD (n=6) (Table 1). Caregiver eye-tracking fixation locations were coded relative to four mutually exclusive regions of interest (ROIs): eyes, mouth, nose/body, and background.

Results: To compare caregiver fixation patterns across groups, we performed a repeated-measures ANOVA with diagnosis as a between-subjects factor and ROI as a within-subjects factor. We observed a significant main effect of ROI (p<0.001), with caregivers fixating more on their infant’s eyes than mouths, and more on infant’s mouths than bodies (all p’s<.05). The diagnosis by ROI interaction trended towards significance (p=0.096). Exploratory independent-samples t-tests revealed a trend towards increased mouth-looking in caregivers of TD infants compared to caregivers of infants with ASD (p=0.067) (Figure 1).

Conclusions: Preliminary analyses provide the first exploration of caregiver viewing patterns during infant-caregiver dyadic interactions, revealing that caregivers spend more time fixating on their infant’s eyes, followed by mouth, and body. Immediate next steps include increasing the sample size to further investigate the observed trend towards increased mouth-looking in caregivers of TD infants. Additionally, longitudinal analyses of infant affect, facial cues, and vocalizations will be completed to identify specific infant signals that influence moment-by-moment shifting of caregiver visual attention. These analyses have important implications for understanding how early disruptions to interactive contingency within the infant-caregiver dyad may impact subsequent development in infants with ASD.