Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Focused State Recognition Using EEG with Eye Movement-Assisted Annotation

Tian-Hua Li, Tian-Fang Ma, Dan Peng, Wei-Long Zheng, Bao-Liang Lu

TL;DR

This work targets focused-state recognition by pairing eye-movement-derived binocular disparity with EEG features to annotate focused versus unfocused periods. It builds a dataset of 62-channel EEG data annotated via eye-tracking and evaluates multiple classifiers, with the Transformer achieving 90.16% accuracy, 92.59% F1, and 94.09% AUC in subject-dependent tests. Differential Entropy features across five EEG bands are computed per time slice, and a rigorous preprocessing and balancing strategy supports robust evaluation. Cross-subject tests show strong generalizability (≈87.65% accuracy), and topographic analyses reveal gamma-band activations in focused states, underscoring the method's physiological relevance and potential for application in safety and mental-state monitoring.

Abstract

With the rapid advancement in machine learning, the recognition and analysis of brain activity based on EEG and eye movement signals have attained a high level of sophistication. Utilizing deep learning models for learning EEG and eye movement features proves effective in classifying brain activities. A focused state indicates intense concentration on a task or thought. Distinguishing focused and unfocused states can be achieved through eye movement behaviors, reflecting variations in brain activities. By calculating binocular focusing point disparity in eye movement signals and integrating relevant EEG features, we propose an annotation method for focused states. The resulting comprehensive dataset, derived from raw data processed through a bio-acquisition device, includes both EEG features and focused labels annotated by eye movements. Extensive training and testing on several deep learning models, particularly the Transformer, yielded a 90.16% accuracy on the subject-dependent experiments. The validity of this approach was demonstrated, with cross-subject experiments, key frequency band and brain region analyses confirming its generalizability and providing physiological explanations.

Focused State Recognition Using EEG with Eye Movement-Assisted Annotation

TL;DR

This work targets focused-state recognition by pairing eye-movement-derived binocular disparity with EEG features to annotate focused versus unfocused periods. It builds a dataset of 62-channel EEG data annotated via eye-tracking and evaluates multiple classifiers, with the Transformer achieving 90.16% accuracy, 92.59% F1, and 94.09% AUC in subject-dependent tests. Differential Entropy features across five EEG bands are computed per time slice, and a rigorous preprocessing and balancing strategy supports robust evaluation. Cross-subject tests show strong generalizability (≈87.65% accuracy), and topographic analyses reveal gamma-band activations in focused states, underscoring the method's physiological relevance and potential for application in safety and mental-state monitoring.

Abstract

With the rapid advancement in machine learning, the recognition and analysis of brain activity based on EEG and eye movement signals have attained a high level of sophistication. Utilizing deep learning models for learning EEG and eye movement features proves effective in classifying brain activities. A focused state indicates intense concentration on a task or thought. Distinguishing focused and unfocused states can be achieved through eye movement behaviors, reflecting variations in brain activities. By calculating binocular focusing point disparity in eye movement signals and integrating relevant EEG features, we propose an annotation method for focused states. The resulting comprehensive dataset, derived from raw data processed through a bio-acquisition device, includes both EEG features and focused labels annotated by eye movements. Extensive training and testing on several deep learning models, particularly the Transformer, yielded a 90.16% accuracy on the subject-dependent experiments. The validity of this approach was demonstrated, with cross-subject experiments, key frequency band and brain region analyses confirming its generalizability and providing physiological explanations.
Paper Structure (10 sections, 2 equations, 3 figures, 2 tables)

This paper contains 10 sections, 2 equations, 3 figures, 2 tables.

Table of Contents

  1. INTRODUCTION
  2. METHOD
  3. Experiment Setup
  4. Annotation of Focused State
  5. EEG Feature Extraction
  6. Classifier Selection
  7. EXPERIMENT & RESULT DISCUSSION
  8. Results Analysis
  9. Cross-Subject Experiment
  10. CONCLUSION

Figures (3)

  • Figure 1: The pipeline of whole data process.
  • Figure 2: Confusion matrix of SVM and Transformer.
  • Figure 3: Topographic maps of the focused and unfocused states in the five frequency bands over all subjects.