Related Experiment Video
Updated: May 28, 2026
10:43
Eye-tracking Technology and Data-mining Techniques used for a Behavioral Analysis of Adults engaged in Learning Processes
Published on: June 10, 2021
Quantification of Cognitive States via Eye Tracking and Using Artificial Intelligence to Analyze Virtual Reality
1Department of Culture and Technology Convergence, Changwon National University, Changwon 51140, Republic of Korea.
Journal of Eye Movement Research
|May 27, 2026
Summary
Virtual reality (VR) learning can cause cognitive overload due to excessive stimuli. This study uses eye-tracking data and AI to classify cognitive states like distraction and immersion, achieving 75.60% accuracy.
Area of Science:
- Educational Technology
- Human-Computer Interaction
- Cognitive Science
Background:
- Virtual reality (VR) enhances learning engagement but presents complex stimuli, potentially causing cognitive overload and distraction.
- Evaluating learners' cognitive states in VR is crucial for optimizing the learning environment.
- Traditional methods for assessing cognitive states (subjective and objective) have limitations, necessitating advanced techniques.
Purpose of the Study:
- To develop an evaluation system for VR learning experiences using eye-tracking data.
- To classify cognitive states during VR learning, including cognitive overload, immersion, and distraction.
- To explore the correlation between eye-tracking metrics and cognitive states for potential objective quantification.
Main Methods:
- Development of a VR learning experience evaluation system.
- Classification of cognitive states (cognitive overload, immersion, distraction) using eye-tracking data.
- Application of an LSTM-based model for cognitive state classification.
- Subject-independent validation of the model's accuracy.
Main Results:
- An LSTM-based model was developed to classify cognitive states from eye-tracking data.
- The model achieved a moderate accuracy of 75.60% in a subject-independent validation setting.
- Correlations between eye-tracking metrics and specific cognitive states were evaluated.
Conclusions:
- Eye-tracking data, analyzed with AI, shows potential for evaluating cognitive states in VR learning.
- The developed system offers a pathway towards objective quantification of cognitive states, aiding in the design of more effective VR learning environments.
- Further research is needed to improve accuracy and explore broader applications of this technology.
