A transdiagnostic AI-based measure of interpersonal coordination in autism and other conditions.
Lisa D Yankowitz, Casey J Zampella, Juhi Pandey and 7 others
Autistic adolescents coordinated their facial expressions and head movements with a conversation partner significantly less than neurotypical peers, and also less than peers with other psychiatric conditions.
The pattern held in both video and in-person conversations.
Key findings
1In the videoconferencing sample, the autistic group showed significantly lower nonverbal coordination than the neurotypical group (B = 0.14, p < 0.0001) and the mixed psychiatric group (B = 0.071, p < 0.001), giving the ranking AUT < PSY < NT.
2The reduced coordination in autism relative to neurotypical peers was replicated in a separate face-to-face sample of adolescents (B = 0.12, p < 0.05, d = 0.73), and the group-by-context interaction was not significant, meaning the size of the difference did not depend on whether the conversation was on video or in person.
3Facial expression coordination closely matched the overall group differences, while head-pose coordination differences were smaller and less consistent. A Shapley analysis attributed 88.6% of the explained variance in overall coordination to expression coordination and 11.4% to head-pose coordination.
Still to come
How it was doneWhat they foundWhat it means for SLPs
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
Only people who could sustain a 3 to 5 minute spoken conversation were included, so this does not represent the full autism spectrum, including non-speaking or minimally speaking individuals. The sample was primarily adolescents (12-18 years). The full age range analysis (5-52 years) showed a significant age effect, so the findings may not generalize to young children or adults. All conversation partners were non-autistic research staff, so the findings reflect autistic/non-autistic interactions specifically, not autistic/autistic interactions. The context comparison (video versus face-to-face) was between-subjects, not within-subjects, so it cannot directly compare coordination across the two settings for the same person. The reliability of a single conversation is 0.55, which is modest. Averaging over two conversations improves it to 0.71. Sex effects were inconsistent across samples, and the authors note the samples are too small to draw conclusions about sex differences.
Declared interests
Funded by the National Institute of Mental Health, the National Institute on Deafness and Other Communication Disorders, and the Eunice Kennedy Shriver National Institute of Child Health and Human Development. No industry funding or conflicts of interest declared.
The easy way to misread this
Do not read this as a new clinical assessment you can use with patients. It is a research measurement tool that requires specific video recording and AI processing. The reliability of a single conversation is 0.55, and the sample only included people who could sustain a 3 to 5 minute spoken conversation, so it does not represent the full autism spectrum.
Summarised by AI from the full paper, without a clinician reviewing it. Check it against the source before it changes what you do. Read it on PubMed →