Discover datasets

EEG recordings from 18 participants who separately listened to auditory rhythms and watched visual flashing rhythms. The stimulus trains contained occasional omissions, which allowed predictive neural activity to be separated from sensory-evoked activity. The dataset supports study of the predictive mechanisms of rhythm perception. Code for preprocessing, time-frequency computation, frequency band extraction and statistics is provided.

Participants

18

Channels

32 (10-10)

Size

1.95 GB

Version

v1.0.0

Updated

Jun 21, 2026

This BIDS derivative dataset provides single-unit (microwire) recordings from the medial temporal lobe of nine patients with drug-resistant focal epilepsy performing a modified Sternberg verbal working memory task. It contains 1526 sorted units from the hippocampus, entorhinal cortex and amygdala, with spike times per trial, mean and standard deviation waveforms, unit-to-electrode anatomical mapping, and trial properties. Subject, session and trial labels are aligned to OpenNeuro ds004752 (NEMAR on004752), which holds the matching scalp and intracranial EEG. The data originate from the Boran et al. release and were re-packaged without changing values.

Participants

9

Citations

118

Size

16.2 GB

Version

v1.0.2

Updated

Oct 16, 2026

This dataset pools cEEGrid ear-EEG recordings from two earlier studies, with 15 participants from Jaeger et al. (2020) and 21 from Holtze et al. (2021). Participants performed a two-competing-speaker paradigm in which they attended to either the left or right of two simultaneously presented audio books. Blocks were presented in omnidirectional or beamforming conditions, so the data can be used to study auditory attention decoding to continuous speech with ear-EEG.

Participants

36

Channels

18 (other)

Citations

47

Size

6.03 GB

Version

v1.0.0

Updated

Jun 23, 2026

Preprocessed 64-channel EEG recordings from 6 healthy participants performing imagined speech of three short words (out, in, up), cued by auditory and visual stimuli. The dataset contains 1800 trials of a single session, analyzed in the source study with Riemannian manifold covariance features and relevance vector machines for brain-computer interface applications. It is re-hosted in BIDS format from the MOABB benchmark collection.

Participants

6

Channels

60 (10-10)

Citations

99

HED

v8.4.0

Size

1.90 GB

Version

v1.0.3

Updated

Aug 18, 2026

Resting-state EEG recordings from 19 healthy participants who alternated between eyes-closed and eyes-open conditions in 10-second trials. Data were acquired with a 16-channel g.tec g.USBamp system at 512 Hz to study posterior alpha-band (8-12 Hz) modulation. The dataset is distributed in BIDS format via MOABB and originates from the GIPSA-lab Alpha Waves dataset.

Participants

19

Channels

16 (10-10)

Citations

2

HED

v8.4.0

Size

178 MB

Version

v1.0.3

Updated

Aug 18, 2026

Stereo-EEG (SEEG) recordings from 37 patients with refractory epilepsy viewing fast periodic visual stimulation, in which real words appeared at 2 Hz among pseudofonts, nonwords, or pseudowords presented at 10 Hz. The dataset probes letter- and word-selective responses in the ventral occipito-temporal cortex, measured at 2 Hz and its harmonics. This is a BIDS derivative of the authors' Letswave files, low-pass filtered at 40 Hz and cut into sequences, with electrode coordinates in Talairach and MNI spaces.

Participants

37

Citations

93

Size

14.5 GB

Version

v1.0.2

Updated

Oct 16, 2026

The Bern-Barcelona EEG database contains 7500 pairs of intracranial EEG signals, 3750 focal and 3750 non-focal, drawn from five patients with pharmacoresistant temporal lobe epilepsy. Each pair is a 20-second segment sampled at 512 Hz and recorded during seizure-free intervals. The dataset supports nonlinear time series analysis of focal versus non-focal activity. This BIDS version is a lossless repackaging of the published source text files, with patient, channel and time of origin not recorded.

Participants

1

Citations

87

Size

1.38 GB

Version

v1.0.2

Updated

Oct 16, 2026

A BIDS-formatted subset of Kai Miller's library of human electrocorticographic (ECoG) data, covering six calibrated tasks: finger flexion, hand gestures, motor imagery, basic motor movement, and speech (verb generation and noun reading). It contains 111 recordings from 24 epilepsy patients with subdural electrode grids, recorded at 1 kHz at Harborview Hospital. The data are scaled to physical units and include events, channel metadata and electrode positions.

Participants

24

Citations

84

Size

14.0 GB

Version

v1.0.2

Updated

Oct 16, 2026

A large, longitudinal EEG dataset collected while 62 healthy adults learned to control a sensorimotor rhythm-based brain-computer interface. Participants completed 7-11 online training sessions of continuous cursor control using motor imagery of the left hand, right hand, both hands, and rest. In total the dataset comprises 598 sessions and over 250,000 trials, offering a resource for designing and evaluating BCI algorithms and for studying BCI learning.

Participants

62

Channels

60 (10-10)

Citations

68

HED

v8.4.0

Size

722 GB

Version

v1.0.4

Updated

Aug 18, 2026

This dataset contains behavioral events and intracranial EEG recordings from a delayed free recall task in which participants studied visually presented word lists, solved arithmetic distractor problems, and then freely recalled the words. Data were collected at multiple clinical sites in collaboration with the Computational Memory Lab at the University of Pennsylvania. Monopolar and bipolar recordings are provided with subject-specific electrode montages, MNI/Talairach coordinates, and brain region annotations.

Participants

280

Citations

2

Size

576 GB

Version

v1.0.0

Updated

Jun 25, 2026