Discover datasets

BigP3BCI Study S2 is a P300-based brain-computer interface dataset comprising EEG recordings from 24 healthy subjects performing a 9x8 house/tool visual speller paradigm. This derivative dataset, part of the larger bigP3BCI collection, contains standardized EEG data acquired at 256 Hz from 32 channels using g.USBamp hardware. The dataset is designed for machine learning applications and BCI research, with target and non-target event classifications annotated using HED 8.4.0 standards.

Participants

24

Size

1.11 GB

Version

v1.0.1

Updated

Jun 3, 2026

BigP3BCI Study O is a P300-based brain-computer interface dataset comprising EEG recordings from 18 ALS subjects across 2 sessions each. Participants performed a visual speller task using a 9x8 character grid with supervised/checkerboard stimulus paradigm. The dataset contains 32-channel EEG data sampled at 256 Hz, annotated with target and non-target event labels, and is part of the larger BigP3BCI collection representing one of the most comprehensive public P300 BCI datasets.

Participants

18

Size

981 MB

Version

v1.0.1

Updated

Jun 3, 2026

BigP3BCI Study D is a P300-based brain-computer interface dataset comprising EEG recordings from 17 healthy subjects performing a 6x6 character grid speller task using dynamic row-column stimulus presentation. The dataset contains 32-channel EEG data sampled at 256 Hz with standardized 10-20 electrode montage, annotated with target and non-target event labels using HED 8.4.0 schema. This derivative dataset is part of the larger BigP3BCI collection, the largest public P300 BCI dataset, and is designed for machine learning model development and benchmarking in brain-computer interface research.

Participants

17

Size

731 MB

Version

v1.0.1

Updated

Jun 3, 2026

BigP3BCI Study G is a P300-based brain-computer interface dataset comprising EEG recordings from 20 healthy subjects performing a 9x8 checkerboard visual speller task. This derivative dataset is Study G of 20 studies in the larger BigP3BCI collection, the largest public P300 BCI dataset with ~267 subjects total. The data were acquired at 256 Hz using 16-channel EEG with a g.USBamp amplifier and include target and non-target event classifications suitable for machine learning applications.

Participants

20

Size

327 MB

Version

v1.0.1

Updated

Jun 3, 2026

This dataset comprises electroencephalographic recordings from 24 healthy subjects performing a visual P300 brain-computer interface task based on the Brain Invaders video game paradigm. The study compares adaptive versus non-adaptive calibration approaches using Riemannian geometry for classifier training. EEG data were acquired using 16 electrodes at 512 Hz sampling rate during sessions conducted at GIPSA-lab, Grenoble, France in 2013.

Participants

24

Size

1.71 GB

Version

v1.0.1

Updated

Jun 3, 2026

This dataset comprises electroencephalographic recordings from 25 healthy subjects performing a visual P300 brain-computer interface task based on the Brain Invaders paradigm. Participants viewed flashing alien groups and silently counted target stimuli while 16-channel EEG was recorded at 128 Hz. The dataset includes both training and online sessions with performance feedback, providing a benchmark resource for P300-based BCI research and algorithm validation.

Participants

23

Size

163 MB

Version

v1.0.1

Updated

Jun 3, 2026

A comprehensive EEG database containing electroencephalographic signals from 87 healthy participants performing motor imagery brain-computer interface tasks. The dataset comprises over 20,800 trials (~70 hours of recording) organized into three datasets (A, B, C) using a standardized Graz protocol for right and left hand motor imagery. In addition to raw EEG signals, the database includes detailed participant demographics, personality profiles, cognitive traits, and BCI performance metrics, enabling research on user-performance relationships, cross-user machine learning algorithms, and profile-informed signal classification.

Participants

87

Size

8.83 GB

Version

v1.0.1

Updated

Jun 3, 2026

This dataset comprises EEG recordings from 13 healthy subjects performing a visuomotor learning task involving simulated drone piloting through waypoints of varying difficulty levels. The study investigates real-time decoding of subjective difficulty from EEG signals to enable adaptive closed-loop learning, comparing algorithmic difficulty adjustment with subject-controlled progression. Data include offline and online sessions with preprocessed EEG recordings sampled at 256 Hz. Original 64-channel recordings were preprocessed and reduced to 25 central channels (peripheral electrodes removed) for analysis, along with behavioral markers of task performance.

Participants

13

Size

2.96 GB

Version

v1.0.0

Updated

Jun 3, 2026

This dataset comprises EEG recordings from 11 ALS patients performing a P300-based brain-computer interface speller task using a 6x6 character grid. The study is part of the BigP3BCI project, the largest public P300 BCI dataset, and includes 16-channel EEG data sampled at 256 Hz with visual stimulus presentations and target/non-target event classifications. Data were acquired using g.USBamp hardware and annotated with HED 8.4.0 event tags for standardized analysis.

Participants

11

Size

770 MB

Version

v1.0.0

Updated

Jun 3, 2026

This dataset comprises EEG recordings from 10 healthy subjects performing a P300-based brain-computer interface task using a 9x8 character grid with face/house visual stimuli. The study is part of the BigP3BCI project, the largest public P300 BCI dataset, and includes single-session recordings with 32-channel EEG data sampled at 256 Hz. Data were collected for speller application development with online feedback in a laboratory setting.

Participants

10

Size

474 MB

Version

v1.0.0

Updated

Jun 3, 2026