Enhancing Action Recognition in Educational Settings Through Exercise-Induced Neuroplasticity

Mingyang Sun^*Shuangyi Feng

• Catholic University of Korea, Seoul, Republic of Korea

In educational settings, the role of neuroplasticity in shaping cognitive development has gained increasing attention. Traditional pedagogical models often fail to capture the dynamic neural adaptations that underlie effective learning. To bridge this gap, we propose a novel approach that integrates neuroplastic principles into action recognition for educational applications. Existing models primarily rely on behavioral metrics, neglecting the underlying neural mechanisms that drive skill acquisition. Our method introduces a Neuroplastic Learning Dynamics Model (NLDM), a computational framework designed to simulate the synaptic modifications, cortical reorganization, and learning-induced connectivity changes that occur during educational engagement. By leveraging a mathematical formulation of neuroplastic adaptation, NLDM enables a dynamic representation of cognitive transformation. Furthermore, we introduce Neuroplasticity-Driven Learning Optimization (NDLO), a strategic framework that adapts pedagogical interventions based on real-time neural responses. Our approach enhances action recognition by aligning educational content with the learner's neuroplastic profile, optimizing instructional strategies, and leveraging cognitive load modulation. NDLO integrates multimodal data sources, including neurophysiological signals and behavioral feedback, to refine personalized learning pathways. By dynamically adjusting educational interventions, our framework fosters deeper engagement, accelerates skill acquisition, and enhances cognitive flexibility. Experimental results demonstrate that our neuroplasticity-based framework significantly improves action recognition accuracy, learning efficiency, and long-term knowledge retention. This study establishes a direct link between neural adaptability and educational performance, providing a foundation for future advancements in neuroeducation, AI-assisted learning environments, and the development of highly adaptive intelligent tutoring systems.