Analyzing Social Communication Deficits in Autism Using Wearable Sensors and Real-Time Affective Computing Systems

Social communication deficits are a hallmark characteristic of Autism Spectrum Disorder (ASD), often manifesting as challenges in interpreting and expressing emotions, maintaining eye contact, and engaging in reciprocal interactions. Traditional diagnostic and intervention methods, while valuable, can be limited by observer bias and lack of continuous monitoring. This review explores the emerging role of wearable sensors and real-time affective computing systems in assessing and addressing these deficits. By leveraging physiological signals (e.g., heart rate variability, skin conductance, and movement patterns) and behavioral data (e.g., gaze tracking, facial expressions, and speech prosody), wearable devices offer objective, non-intrusive insights into the emotional and communicative states of individuals with ASD. Affective computing systems, powered by machine learning and signal processing techniques, enable real-time analysis and adaptive feedback, potentially enhancing social skills training and personalized intervention. The paper examines current technologies, evaluation metrics, and case studies, while also addressing challenges such as data privacy, model generalizability, and user compliance. Finally, the review highlights future directions for integrating these technologies into clinical practice and educational environments to support early diagnosis, continuous monitoring, and intervention personalization in ASD care.