Related Experiment Video
Updated: May 10, 2025
09:42
Stimulus-specific Cortical Visual Evoked Potential Morphological Patterns
Published on: May 12, 2019
5.9K
A multi-subject and multi-session EEG dataset for modelling human visual object recognition.
Shuning Xue1,2, Bu Jin2, Jie Jiang2
1School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing, China.
Scientific Data
|April 19, 2025
Summary
This study introduces a large electroencephalographic (EEG) dataset from 32 participants across multiple sessions. This multi-subject, multi-session EEG data supports research in visual responses and brain-computer interfaces.
Area of Science:
- Neuroscience
- Cognitive Science
- Machine Learning
Background:
- Electroencephalography (EEG) is a non-invasive neuroimaging technique widely used to study brain activity.
- Visual perception research often utilizes controlled stimulus presentation to understand neural processing.
- Developing robust brain-computer interfaces (BCIs) requires diverse datasets encompassing inter-subject and inter-session variability.
Purpose of the Study:
- To introduce and describe a novel multi-subject and multi-session (MSS) dataset of 122-channel electroencephalographic (EEG) signals.
- To provide a valuable resource for researchers investigating visual processing using EEG.
- To facilitate the development of advanced machine learning algorithms for cross-subject and cross-session BCIs.
Main Methods:
- Collected 122-channel EEG data from 32 healthy human participants over multiple sessions.
- Experiments involved two visual presentation paradigms: rapid serial visual presentation (stimuli at 5 Hz) and single image presentation (1 second duration).
- Utilized a diverse stimulus set of 10,000 images sourced from PASCAL and ImageNet databases.
Main Results:
- The dataset comprises approximately 800,000 trials for the rapid presentation paradigm and 40,000 trials for the single image paradigm.
- Each participant contributed 1 to 5 sessions, with each session lasting approximately 1.5 hours.
- The dataset captures significant inter-subject and inter-session variability in EEG responses.
Conclusions:
- The MSS dataset offers a comprehensive resource for studying EEG-based visual responses.
- It enables comparative analyses of EEG responses across different visual paradigms.
- The dataset is suitable for training and validating machine learning models for robust, generalizable BCIs.
