Identifying the dynamical trajectories of infant-caregiver interactions in Autism Spectrum Disorder

Autism spectrum disorder (ASD) is a developmental disorder where early identification has the potential to improve the long-term prognosis of patients. ASD associated social deficits, however, have been difficult to measure quantitatively, particularly in nonverbal children younger than six months. We utilize two methodologies to identify the behaviors of caregivers and infants that contribute the most to the success of their social interaction, and to quantify age-related behavioral trajectories during interactions between dyads. We used unsupervised techniques to identify and quantify stereotyped facial expression behaviors in both individuals, and a novel causality method based on nonlinear dynamics to assess the extent and directionality of the individuals' interactions. Our data demonstrate differential developmental trajectories of facial expression behaviors in infants interacting with their caregivers between those who go on to develop ASD, typically developing infants, and infants who are at an elevated genetic risk for developing autism that do not ultimately develop the disorder. Our novel method indicates periods of the interaction with both high and low interaction success, providing a potential for new clinically relevant assessment metrics.