Discover datasets

A preprocessed derivative EEG dataset from BCI Competition 2020 Track 4, containing motor imagery recordings of three right-arm grasping tasks (cylindrical, spherical, lumbrical) from 15 healthy subjects across three sessions. It provides 60-channel EEG at 250 Hz with 450 trials per subject, and is intended for evaluating session-to-session transfer learning in brain-computer interfaces. Data were notch filtered and epoched to the 4 s motor imagery window, and re-hosted in BIDS format via MOABB.

Participants

14

Channels

60 (10-10)

HED

v8.4.0

Size

2.76 GB

Version

v1.0.3

Updated

Aug 18, 2026

A pediatric steady-state visual evoked potential (SSVEP) brain-computer interface dataset recorded from 47 healthy children and adolescents aged 5-18 years. Each participant completed a stimulus personalization and comfort-rating procedure, then played an online four-target SSVEP-controlled game with both a personalized and a standard high-contrast stimulus. The 16-channel EEG recordings are provided in BIDS format with HED event annotations.

Participants

41

Channels

16 (10-10)

HED

v8.4.0

Size

1.73 GB

Version

v1.0.0

Updated

Oct 8, 2026

EEG recordings from 21 healthy participants performing a visual P300 brain-computer interface speller task, with each participant completing one session on a personal computer and one in virtual reality using a passive smartphone-based head-mounted display. Signals were recorded from 16 electrodes at 512 Hz. The dataset supports comparison of P300-based BCI performance, physiological responses, and user experience between the PC and VR conditions.

Participants

21

Channels

16 (10-10)

Citations

5

HED

v8.4.0

Size

999 MB

Version

v1.0.5

Updated

Oct 8, 2026

EEG dataset from the PROTEUS brain-computer interface (BCI) project in Bordeaux, released for the EEG/EMG Foundation Challenge 2026 (Track 02). It contains recordings from 10 participants across 14 sessions (112 runs) performing three cued mental tasks: kinesthetic motor imagery, mental subtraction and word generation. Runs include resting baselines, calibration runs with sham feedback and online runs with real feedback, using Graz and BrainHero interfaces.

Participants

10

Channels

41 (10-10)

Size

1.48 GB

Version

v1.0.0

Updated

Oct 1, 2026

Study D of the bigP3BCI collection, comprising 32-channel EEG recordings from 17 healthy participants using a P300-based speller with a 6x6 character grid and dynamic row-column stimulus presentation. Data were recorded in a single session per subject with online feedback, and events are labeled as target or non-target flashes. The dataset is a BIDS-formatted derivative of the PhysioNet bigP3BCI release, intended for machine learning and BCI benchmarking.

Participants

17

Channels

32 (10-10)

Citations

2

HED

v8.4.0

Size

2.41 GB

Version

v1.0.3

Updated

Aug 18, 2026

Three-class motor imagery EEG dataset (MI-B) from the BCI competition at the 2023 World Robot Contest (WRCC2023). It contains recordings from nine subjects, two of them stroke patients and seven healthy, performing cued imagery of left hand, right hand and feet movements (30 trials per class). Signals were recorded with a 59-channel Neuracle EEG system at 1000 Hz and re-released in BIDS format via MOABB.

Participants

9

Channels

59 (10-10)

HED

v8.4.0

Size

1.80 GB

Version

v1.0.0

Updated

Oct 8, 2026

EEG and behavioral recordings from participants performing an extended simulated driving task with modulated visual complexity and perturbation frequency. The study aimed to identify driver fatigue and task performance changes through EEG-based biomarkers, compared against objective performance measures and a non-fatigued calibration driving session. It extends the Baseline Driving paradigm by adding pedestrians and traffic and varying how often lateral perturbations occurred.

Participants

29

Channels

64 (10-10)

HED

v8.0.0

Size

17.5 GB

Version

v1.0.0

Updated

Jun 23, 2026

EEG recordings from 24 healthy participants performing a visual oddball (P300) event-related potential paradigm during mobile conditions: standing, walking, and running. Data were acquired with scalp and ear-EEG systems (73 channels, plus EOG) across five sessions per participant. This is a BIDS-formatted derivative of the Mobile BCI dataset, intended for benchmarking brain-computer interface decoding under locomotion.

Participants

24

Channels

48 (10-10)

Citations

36

HED

v8.4.0

Size

8.08 GB

Version

v1.0.3

Updated

Oct 8, 2026

A multimodal EEG dataset of motor imagery (MI) recorded from 31 healthy participants over up to three sessions each. Participants imagined left- and right-hand grasping in response to visual cues while 64-channel EEG plus EOG were recorded at 1000 Hz. This is a MOABB-derived version of the Eye-BCI dataset, which also includes eye-tracking and high-speed video for ocular activity analysis across BCI paradigms.

Participants

31

Channels

66 (10-10)

Citations

13

HED

v8.4.0

Size

4.65 GB

Version

v1.0.0

Updated

Jun 3, 2026

Study J of the bigP3BCI collection, which provides EEG recordings from 20 healthy participants using a P300-based speller. The task used a 9x8 character grid with a performance-based row-column paradigm, and 16 EEG channels were recorded at 256 Hz. Target and non-target flash events are annotated with HED tags, and the dataset supports machine-learning benchmarking of P300 brain-computer interfaces.

Participants

20

Channels

16 (10-10)

Citations

2

HED

v8.4.0

Size

1.49 GB

Version

v1.0.3

Updated

Aug 18, 2026