Related Experiment Video
Updated: May 15, 2025
Using Electroencephalography Measurements and High-quality Video Recording for Analyzing Visual Perception of Media Content
Published on: May 26, 2018
Dataset combining EEG, eye-tracking, and high-speed video for ocular activity analysis across BCI paradigms
Eva Guttmann-Flury1, Xinjun Sheng2, Xiangyang Zhu2
1State Key Laboratory of Mechanical System and Vibration, School of Mechanical Engineering, Shanghai Jiao Tong University, 800 Dongchuan Road, Minhang District, Shanghai, 200240, P. R. China. eva.guttmann.flury@gmail.com.
This study introduces a large multimodal dataset for Brain-Computer Interface (BCI) research, analyzing eye blinks as both noise and features. The dataset aids in developing algorithms to handle eye artifacts and improve cognitive state decoding.
Area of Science:
- Neuroscience
- Computer Science
- Human-Computer Interaction
Background:
- Eye blinks in Brain-Computer Interface (BCI) research present a dual role: as noise that degrades signal quality and as valuable behavioral indicators.
- Accurate decoding of user cognitive states and intentions is often hindered by ocular artifacts in electroencephalogram (EEG) data.
Purpose of the Study:
- To introduce a comprehensive, multimodal dataset for the multifactorial analysis of eye-related movements in BCI.
- To facilitate the development of advanced algorithms for mitigating eye-induced artifacts and enhancing BCI performance.
- To enable the evaluation of cross-paradigm robustness in BCI using data from the same participants.
Main Methods:
- Collection of a large-scale dataset including electroencephalogram (EEG) signals, eye-tracking, and high-speed camera recordings.
- Inclusion of subject mental states and characteristics for a holistic analysis.
- Utilization of four distinct BCI paradigms: motor imagery, motor execution, steady-state visually evoked potentials, and P300 spellers.
Main Results:
- The dataset comprises over 46 hours of data from 31 subjects across 63 sessions.
- Includes substantial trial counts for each paradigm: 2520 for the first three and 5670 for P300 spellers.
- Provides a rich resource for multimodal analysis of eye movements and their impact on BCI.
Conclusions:
- The presented dataset is expected to significantly advance BCI research by enabling the development of robust artifact handling techniques.
- It will support the creation of algorithms that improve the accuracy of cognitive state decoding in the presence of eye artifacts.
- Offers a unique opportunity to assess the generalizability of BCI algorithms across different experimental paradigms.
