Autism spectrum disorder (ASD) is a neurodevelopmental condition characterized in part by difficulties in verbal and nonverbal social communication. Evidence indicates that autistic people, compared to neurotypical peers, exhibit differences in head movements, a key form of nonverbal communication. Despite the crucial role of head movements in social communication, research on this nonverbal cue is relatively scarce compared to other forms of nonverbal communication, such as facial expressions and gestures. There is a need for scalable, reliable, and accurate instruments for measuring head movements directly within the context of social interactions. In this study, we used computer vision and machine learning to examine the head movement patterns of neurotypical and autistic individuals during naturalistic, face–to–face conversations, at both the individual (monadic) and interpersonal (dyadic) levels. Our model predicts diagnostic status using dyadic head movement data with an accuracy of , highlighting the value of head movement as a marker of social communication. The monadic data pipeline had lower accuracy () compared to the dyadic approach, emphasizing the importance of studying back-and-forth social communication cues within a true social context. The proposed classifier is not intended for diagnostic purposes, and future research should replicate our findings in larger, more representative samples.
Advances in computational behavior analysis have the potential to increase our understanding of behavioral patterns and developmental trajectories in neurotypical individuals, as well as in individuals with mental health conditions marked by motor, social, and emotional difficulties. This study focuses on investigating how head movement patterns during face–to–face conversations vary with age from childhood through adulthood. We rely on computer vision techniques due to their suitability for analysis of social behaviors in naturalistic settings, since video data capture can be unobtrusively embedded within conversations between two social partners. The methods in this work include unsupervised learning for movement pattern clustering, and supervised classification and regression as a function of age. The results demonstrate that 3–minute video recordings of head movements during conversations show patterns that distinguish between participants that are younger vs. older than 12 years with accuracy. Additionally, we extract relevant patterns of head movement upon which the age distinction was determined by our models.
Recent work has demonstrated the feasibility of producing knitted capacitive touch sensors through digital fabrication, which rely on a single conductive yarn and two external connections. This technique increases these sensors' robustness and usability, while shifting the complexity of enabling interactivity from the hardware to computational models. The application of algorithmic and artificial intelligence models to these novel pervasive technologies is key to unfolding their potential, particularly when real-world and user experience considerations are also included. To bring this technology closer to real-world use, this dissertation goes beyond previous work on coarse touch discrimination, to enable fine, accurate touch localization and complex gesture recognition on these low-profile knitted sensors. Deep learning and algorithmic models are presented to analyze noisy time-series signal data, which are able to capture the temporal behavior of the sensors and extract relevant local features. Furthermore, several user studies are conducted to train these models, demonstrate their generalizability with new users, and investigate their robustness when exposed to everyday use events. To start shaping the future of touch-sensitive fabric technology according to user expectations and everyday use scenarios, through a formative focus group study, users' views of these fabrics are also explored in different contexts. The contributions of this work set the foundations for creating pervasive interactive systems powered by artificial intelligence models that use minimalistically-designed knitted sensors as an input medium.