Optimizing VR Learning with Eye-Tracking and Deep Learning: Enhancing Concentration and Memory

Virtual Reality (VR) technology offers immersive learning experiences by simulating realistic environments. However, traditional VR systems often lack personalized adaptation to learners’ cognitive states. This study introduces a VR teaching system integrating eye-tracking and Convolutional Neural Networks (CNNs) to dynamically adjust instructional content based on learners’ visual attention and cognitive load. By analyzing gaze behavior and cognitive responses, the system provides adaptive feedback to enhance learning outcomes. Experimental results demonstrate a significant positive correlation ([Formula: see text], [Formula: see text]) between attention concentration and memory accuracy, suggesting that adaptive feedback based on eye-tracking enhances learning. Structural equation modeling validates the system’s predictive mechanisms and instructional adaptability, offering empirical support for integrating eye-tracking and deep learning in VR-based instructional design.