Complexity analysis of head movements in autistic toddlers

Abstract

Background: Early differences in sensorimotor functioning have been documented in young autistic children and infants who are later diagnosed with autism. Previous research has demonstrated that autistic toddlers exhibit more frequent head movement when viewing dynamic audiovisual stimuli, compared to neurotypical toddlers. To further explore this behavioral characteristic, in this study, computer vision (CV) analysis was used to measure several aspects of head movement dynamics of autistic and neurotypical toddlers while they watched a set of brief movies with social and nonsocial content presented on a tablet. Methods: Data were collected from 457 toddlers, 17-36 months old, during their well-child visit to four pediatric primary care clinics. Forty-one toddlers were subsequently diagnosed with autism. An application (app) displayed several brief movies on a tablet, and the toddlers watched these movies while sitting on their caregiver's lap. The front-facing camera in the tablet recorded the toddlers' behavioral responses. CV was used to measure the participants' head movement rate, movement acceleration, and complexity using multiscale entropy. Results: Autistic toddlers exhibited significantly higher rate, acceleration, and complexity in their head movements while watching the movies compared to neurotypical toddlers, regardless of the type of movie content (social vs. nonsocial). The combined features of head movement acceleration and complexity reliably distinguished the autistic and neurotypical toddlers. Conclusions: Autistic toddlers exhibit differences in their head movement dynamics when viewing audiovisual stimuli. Higher complexity of their head movements suggests that their movements were less predictable and less stable compared to neurotypical toddlers. CV offers a scalable means of detecting subtle differences in head movement dynamics, which may be helpful in identifying early behaviors associated with autism and providing insight into the nature of sensorimotor differences associated with autism.

Relationship between quantitative digital behavioral features and clinical profiles in young autistic children

“…Expressive language skills were associated with greater head movements, particularly when viewing nonsocial stimuli. Our earlier work on head movements during SenseToKnow found greater variability, or complexity, in head movements in the autism group compared to the non‐autism group (Krishnappa Babu et al, 2022).…”

“…(a) Features based on the name call: Response to Name (RN), and Response to Name Delay (RND) (Perochon et al, 2021) (b) Features based on gaze information: Gaze Percent Social (GPS), Gaze Silhouette Score (GSS), Gaze Speech Correlation (GSC) (Chang et al, 2021). (c) Features based on the head movement: Head Movement during Social stimuli (HMS), Head Movement during non‐Social stimuli (HMnS) (Dawson et al, 2018; Krishnappa Babu et al, 2022). (d) Features based on the performance during the Bubble Popping game: Touch Popping Ratio (TPR), Touch screen Exploration Percentage (TEP), Touch Average Applied Force (TAAF), Touch Average Length (TAL).…”

“…The proxy for attention defined above was used to select the frames of interest, and the change in the distance between the eyes was used to account for invariance in the distance between a child and the screen. We computed the average rate of head movement across each stimuli according to previously established procedures (Dawson et al, 2018; Krishnappa Babu et al, 2022). Dawson et al (2018) reported that autistic toddlers showed more frequent head movements while watching the movies, accentuated during movies of high social content.…”

“…At the end of the movies, a brief bubble popping game is shown on the device, and the child is encouraged to pop the bubbles with their finger while all touch related features are recorded. The movies and game shown on the device are designed to elicit behaviors known to be associated with autism, including response to name, social attention (including attention to speech), frequency of head movements (Dawson et al, 2018; Krishnappa Babu et al, 2022), and visual motor abilities (Campbell et al, 2019; Chang et al, 2021; Dawson et al, 2018; Krishnappa Babu et al, 2022; Perochon et al, 2021). Potential strengths of this approach include scalability, improved measurement and temporal precision, and objective, reliable measures that are not subject to reporter expectancy effects and do not require literacy on the part of the reporter.…”

Blink rate and facial orientation reveal distinctive patterns of attentional engagement in autistic toddlers: a digital phenotyping approach

“…By combining these novel indices of attention with other digital phenotypic features, such as facial dynamics 24 , 25 , orienting 26 , and head movements 27 , 28 , in the future, it may be possible to develop a scalable robust phenotyping tool to detect autism in toddlers, as well as monitor longitudinal development and response to early intervention.…”