Cognitive and Behavioral Modeling in Digital Learning Systems

Digital learning systems are evolving well beyond static content delivery toward dynamic, context-aware environments capable of modeling the whole learner. This chapter presents a Digital Twin–Cognitive-Behavioral Modeling (DT-CBM) framework that integrates knowledge tracing, affective state recognition, self-regulated learning (SRL) detection, and reinforcement-based pathway optimization within a unified digital twin architecture. Drawing on multimodal data—clickstreams, physiological signals, facial expressions, and assessment logs—the framework constructs a living computational mirror of each learner's cognitive and behavioral state. Explainable AI (XAI) mechanisms translate opaque model outputs into actionable, transparent feedback for learners and educators alike. The chapter reviews foundational theories, examines current modeling techniques, identifies persistent limitations, and proposes a forward-looking architecture suited to lifelong adaptive learning.

Перевод пока недоступен