2026-10-07T06:30:25Z https://api.figshare.com/v2/oai

oai:figshare.com:article/33404251 2026-09-24T10:02:18Z category_26299 category_24748 category_26407 category_23848 item_type_3 month_year_09_2026

High-Quality Rapid Serial Visual Presentation (RSVP) EEG Dataset ziyuan zhang (23862255) Biomedical engineering not elsewhere classified Neurosciences not elsewhere classified Signal processing Agricultural biotechnology not elsewhere classified Brain-Computer-Interfaces (BCI) rapid serial visual presentation (RSVP) event related potential (ERP) electroencephalogram (EEG) biological engineering If you use this dataset, please cite both of the following publications: Zhang Z, Wang Z, Guo K, et al. Boosting brain-computer interface performance through cognitive training: A brain-centric approach [J]. Journal of Information and Intelligence, 2025, 3(1): 19–35. Zhang Z, Zheng Y, Guo K, et al. A Few-Layer Multilayer Perceptron is Worth Attention for EEG Classification in Rapid Serial Visual Presentation Task [J]. International Journal of Neural Systems, 2026, 36(09): 2650030. <h4 dir="ltr">@article{zhang2025boosting, title={Boosting brain-computer interface performance through cognitive training: A brain-centric approach}, author={Zhang, Ziyuan and Wang, Ziyu and Guo, Kaitai and Zheng, Yang and Dong, Minghao and Liang, Jimin}, journal={Journal of Information and Intelligence}, volume={3}, number={1}, pages={19--35}, year={2025}, publisher={Elsevier} }</h4><h4 dir="ltr"> @article{zhang2026few, title={A Few-Layer Multilayer Perceptron is Worth Attention for EEG Classification in Rapid Serial Visual Presentation Task}, author={Zhang, Ziyuan and Zheng, Yang and Guo, Kaitai and Liang, Jimin and Dong, Minghao}, journal={International Journal of Neural Systems}, volume={36}, number={09}, pages={2650030}, year={2026}, publisher={World Scientific} } The dataset described in the original publication includes two participant groups: an experimental group (Group A) and a control group (Group B). EEG recordings were collected during both high-target-proportion RSVP tasks (20% targets) and low-target-proportion RSVP tasks (4% targets), with each participant completing a pre-test session and a post-test session. Participants in Group A underwent cognitive training between the two sessions, and the post-test session therefore contains EEG data acquired after training. Due to the current total upload size limit of 20 GB, this released dataset temporarily includes only the RSVP EEG recordings from the post-test session of Group A under the two target-proportion conditions.</h4> This dataset comprises EEG recordings from 17 subjects (10 male and 7 female), all of whom underwent cognitive training to provide high-quality EEG data. Detailed experimental protocols and cognitive training procedures are described in the original publication. EEG signals were acquired using a 64-channel ActiCHamp system (Brain Products, Inc.) at a sampling rate of 1000 Hz, following an extended 10–20 electrode montage. FCz served as the reference electrode, resulting in 63 recorded EEG channels. The spatial coordinates of the 63 electrodes are provided in the accompanying <code>.ced</code> channel-location file.Each subject completed four blocks of RSVP tasks, with each block consisting of 60 sequences of 50 images presented at 10 Hz, resulting in 12,000 trials per subject. Depending on the experimental version, target images accounted for either 8% or 20% of the presented stimuli.The EEG recordings are stored in the standard BrainVision format, consisting of <code>.vhdr</code>, <code>.vmrk</code>, and <code>.eeg</code> files. The data can be loaded and processed using either EEGLAB in MATLAB or MNE-Python. In EEGLAB, the recordings can be imported using <code>pop_loadbv</code>, and the electrode locations can be assigned from the provided <code>.ced</code> file using <code>pop_chanedit</code>. In MNE-Python, the recordings can be loaded directly from the <code>.vhdr</code> file using <code>mne.io.read_raw_brainvision</code>, which automatically accesses the associated <code>.eeg</code> and <code>.vmrk</code> files. Users may then perform their own preprocessing and event extraction according to their specific analysis requirements. 2026-09-24T10:02:18Z Dataset Dataset 10.6084/m9.figshare.33404251.v3 https://figshare.com/articles/dataset/_b_High-Quality_Rapid_Serial_Visual_Presentation_RSVP_EEG_Dataset_b_/33404251 CC BY 4.0