A human-centered approach to classroom analysis that combines artificial intelligence (AI), multimodal teacher biometrics (such as heart rate, electrodermal activity, and nonverbal cues), and student sensor data (including engagement, emotions, and environmental factors) can offer formative insights for reflective coaching. This paper reviews recent literature and highlights applications of the use of sensors in special education with a specific focus on coaching. Challenges and limitations are addressed, including signal noise, ecological validity in live classrooms, wearability and accessibility issues, potential bias, and limited generalization without local calibration. Additional challenges involve governance burdens, such as consent, retention, FERPA/IDEA compliance, infrastructure and training needs, and reliance on short-term, quasi-experimental evidence. Best practices for responsibly integrating multimodal sensor data streams across teachers and some initial findings in student use aligned with supporting teachers in coaching are suggested, representing high-impact next steps for special education and teacher preparation.
Dieker et al. (Fri,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: