Prosodic Synchrony Profiles in Mandarin-Speaking Children With Autism, Developmental Delay, and Typical Development: A Multi-Timescale Analysis

Purpose: Prosodic synchrony differences between autistic and non-autistic individuals have been increasingly documented, but findings remain heterogeneous. This study delineates a multi-timescale profile of static and dynamic pitch synchrony in autistic children during naturalistic Mandarin caregiver–child interactions to deepen understanding of interactional characteristics in autism. Method: Sixty Mandarin-speaking children with autism spectrum disorder (ASD), developmental delay, or typical development (TD) participated in 10-min caregiver–child free-play interactions (20 per group). Prosodic synchrony was indexed from fundamental frequency ( F 0) using a multi-timescale analytical framework informed by Wynn and Borrie (2022) , with dynamic time warping (DTW) for utterance alignment and pseudo-pair normalization for synchrony estimation. Analyses were performed at three timescales: turn level (adjacent caregiver-to-child utterance pairs), block level (short sequences of alternating turns separated by brief pauses), and conversation level (the full interaction). Static synchrony captured overall similarity averaged within a timescale, while dynamic synchrony captured systematic time-varying changes in synchrony over the interaction. Group effects and temporal structure were evaluated using linear mixed-effects models and generalized additive mixed models (GAMMs). Results: The results indicated that children with ASD demonstrated significantly reduced static prosodic synchrony compared to their TD peers, consistently observed across turn-level ( p = .005), block-level ( p = .001), and conversation-level analyses ( p = .005). Although dynamic synchrony was rarely observed, group-wise comparisons of smooth trajectories revealed that children with developmental disorders showed pronounced deviations in prosodic synchrony relative to their TD peers within specific temporal segments. Moreover, group differences emerged in trajectory variability, with the ASD group showing greater range and standard deviation of dynamic synchrony trajectories ( p < .001). Conclusions: The results offer a multi-scale behavioral profile of prosodic synchrony in neurodevelopmental conditions. Turn-level synchrony may be informative for early screening, and dynamic-trajectory variability may support differential characterization and intervention tracking. Supplemental Material: https://doi.org/10.23641/asha.32867315