Discover datasets

A preprocessed EEG dataset comprising motor imagery recordings from 14 healthy participants performing imagined right-hand tapping, left-hand tapping, and rest tasks. The dataset contains 1,665 trials acquired at 1024 Hz using a 19-channel montage with standard 10-20 electrode placement. Data have been preprocessed with ICA artifact removal and frequency filtering, and include annotations for motor imagery classification and brain-computer interface applications.

Participants

14

Channels

19 (10-20)

Citations

9

HED

v8.4.0

Size

1.23 GB

Version

v1.0.2

Updated

Aug 18, 2026

A classical motor imagery EEG dataset comprising recordings from 7 healthy subjects performing left hand, right hand, and passive rest tasks. The dataset contains 19-channel EEG data sampled at 200 Hz with visual arrow cues guiding motor imagery tasks. This benchmark dataset is designed for evaluating brain-computer interface algorithms and motor imagery classification methods.

Participants

7

Channels

19 (10-20)

Citations

81

HED

v8.4.0

Size

1.20 GB

Version

v1.0.2

Updated

Aug 18, 2026

Participants

40

Channels

30 (10-10)

Citations

92

HED

v8.4.0

Size

17.5 GB

Version

v1.1.1

Updated

Jul 10, 2026

A derivative SSVEP dataset comprising EEG recordings from 8 healthy subjects performing a combined brain-computer interface task with overlapping visual stimuli at a single location. Participants attended to one of four arrow directions (up, down, left, right), each flickering at distinct frequencies (14.17, 12.14, 9.44, 7.73 Hz) derived from an 85 Hz CRT refresh rate. Data were acquired at 1000 Hz using a 32-channel ANT Neuro system across two experimental schemes examining moving dot-formed arrows and space/object-based attention.

Participants

8

Channels

31 (10-10)

Citations

3

HED

v8.4.0

Size

5.30 GB

Version

v1.0.2

Updated

Aug 18, 2026

This dataset (eldBETA) provides 64-channel SSVEP-BCI EEG recordings from 100 elderly participants (aged 51-81, mean 63.17 years) performing a 9-target speller task using joint frequency and phase modulation (JFPM) stimuli. Each subject completed 7 sessions of 7 blocks with 9 trials each, recorded at 1000 Hz using a Synamps2 (Neuroscan) amplifier with a Cz reference. The dataset is intended as a benchmark for SSVEP-BCI algorithms in aging populations and has been converted to BIDS format with HED event annotations for use in the MOABB benchmarking framework.

Participants

100

Channels

64 (10-10)

Citations

29

HED

v8.4.0

Size

34.8 GB

Version

v1.0.3

Updated

Aug 20, 2026

A 40-class steady-state visually evoked potential (SSVEP) brain-computer interface speller dataset acquired from 40 healthy subjects using beta-range stimulation frequencies (14.0–21.8 Hz) to reduce visual fatigue. The dataset comprises 33-channel EEG recordings (31 scalp + 2 mastoid references) sampled at 1024 Hz across 6 sessions per subject, with 240 trials total using the joint frequency-phase modulation (JFPM) approach. This resource supports the development and benchmarking of low-fatigue BCI applications for communication interfaces.

Participants

40

Channels

31 (10-10)

Citations

4

HED

v8.4.0

Size

16.5 GB

Version

v1.0.2

Updated

Aug 18, 2026

A benchmark steady-state visual evoked potential (SSVEP) dataset comprising EEG recordings from 34 healthy subjects performing a 40-target brain-computer interface speller task using joint frequency and phase modulation. The dataset includes 64-channel EEG data sampled at 250 Hz with visual stimuli flickering at frequencies ranging from 8 to 15.8 Hz, designed to facilitate development and evaluation of SSVEP-based BCI algorithms.

Participants

34

Channels

64 (10-10)

Citations

98

HED

v8.4.0

Size

6.48 GB

Version

v1.0.2

Updated

Aug 18, 2026

This dataset comprises steady-state visually evoked potential (SSVEP) recordings from 23 healthy participants performing a brain-computer interface task during various locomotor states (standing, walking, running). The study includes 73-channel EEG data acquired at 100 Hz across 4 sessions per subject, with visual stimuli presented at three frequencies (5.45, 8.57, and 12.0 Hz). The dataset is designed to evaluate mobile BCI performance and the feasibility of SSVEP-based control during dynamic physical activity. This is a BIDS-formatted derivative of the original dataset published by Lee et al. (2021) and available at https://osf.io/r7s9b/.

Participants

23

Channels

73 (other)

Citations

34

HED

v8.4.0

Size

1.31 GB

Version

v1.0.2

Updated

Aug 18, 2026

This dataset contains 64-channel EEG recordings from 24 healthy subjects performing two SSVEP-BCI paradigms (low-frequency 8-15.5 Hz and high-frequency 25.5-33 Hz, 16 targets each) using JFPM visual flicker stimulation. Each subject completed training and fatigue-inducing sessions, enabling study of how mental fatigue affects SSVEP-BCI performance and classification accuracy. The dataset was converted to BIDS format using MOABB and is intended to support research on dynamic stopping strategies and fatigue-aware BCI algorithms.

Participants

24

Channels

64 (10-10)

Citations

4

HED

v8.4.0

Size

33.1 GB

Version

v1.0.1

Updated

Aug 18, 2026

This dataset contains SSVEP (Steady-State Visually Evoked Potential) EEG recordings from 12 healthy subjects seated in an exoskeleton-equipped wheelchair, collected as part of E. Kalunga's PhD research at the University of Versailles. Subjects focused on LED panels flickering at 13, 17, and 21 Hz, or on a fixation point (reject/rest class), across 1–5 sessions per subject. The dataset was designed for developing online SSVEP-based brain-computer interface (BCI) systems using Riemannian geometry, with applications toward assistive robotics.

Participants

12

Channels

8 (10-10)

Citations

74

HED

v8.4.0

Size

160 MB

Version

v1.0.1

Updated

Aug 18, 2026