Automated Quantification of Affect Synchrony: Links to Autism Risk and Social Communication in Infancy.

Area of Science:

• Developmental Psychology
• Computational Social Science
• Autism Research

Background:

• Interpersonal affect synchrony (IAS) is crucial for early social development.
• Quantifying IAS and its link to Autism Spectrum Disorder (ASD) presents scalability challenges.
• Objective, behavior-based assessments in infancy are needed for early ASD detection.

Purpose of the Study:

• To evaluate automated, computer-vision-based quantification of IAS.
• To determine if automated IAS can identify early dyadic interaction patterns linked to ASD risk.
• To assess the association between IAS and later ASD outcomes.

Main Methods:

• Analysis of video recordings from 70 mother-infant dyads at elevated ASD risk (6-14 months).
• Quantification of IAS using a validated computer vision pipeline.
• Comparison of classification models with and without dyadic metrics for ASD risk identification.

Main Results:

• Infants later diagnosed with ASD showed significantly lower IAS.
• IAS independently predicted concurrent social-communicative functioning.
• Dyadic metrics improved classification model performance for ASD risk.

Conclusions:

• Automated IAS quantification offers a reliable and scalable method for assessing early interaction patterns.
• Findings support targeting parent-infant synchrony for early ASD screening and intervention.
• This approach provides a promising direction for objective, behavior-based infant assessments.