Discover datasets

Preprocessed 64-channel EEG recordings from six healthy participants performing imagined speech of a short word ('in') versus a long word ('cooperate'). The data were band-pass and notch filtered and cleaned of EOG artifacts. They were used to classify imagined speech with Riemannian manifold covariance features and relevance vector machines for brain-computer interface applications. The dataset is re-hosted in BIDS format through the MOABB benchmark collection.

Participants

6

Channels

60 (10-10)

Citations

99

HED

v8.4.0

Size

1.15 GB

Version

v1.0.3

Updated

Aug 18, 2026

THINGS-EEG2 is a large EEG dataset from 10 participants who viewed 16,540 training images and 200 repeatedly presented test images from the THINGS database. Images were shown in rapid serial visual presentation at 5 Hz while 63-channel EEG was recorded at 1000 Hz across four sessions per subject. The dataset supports modeling of human visual object recognition and includes resting-state recordings, the authors' preprocessed epochs, and stimulus images.

Participants

10

Channels

63 (10-10)

Citations

88

Size

241 GB

Version

v1.1.0

Updated

Jul 5, 2026

TMNRED is an EEG dataset recorded from 30 native Chinese speakers during natural reading of news headlines and short sentences. Participants identified fuzzy semantic targets among target and non-target items. The dataset supports research on semantic processing of Chinese in naturalistic reading and includes raw and preprocessed data in BIDS format.

Participants

30

Channels

30 (10-10)

Citations

4

Size

17.4 GB

Version

v1.0.0

Updated

Jun 26, 2026

Stereotactic EEG recordings of 78 seizures (44 spontaneous, 34 stimulation-induced) from 29 patients with drug-resistant epilepsy, recruited at the University of Pennsylvania and the Children's Hospital of Philadelphia. Three or five board-certified epileptologists independently annotated each seizure for onset time, onset channels, and channels seizing 10 seconds after onset. The BIDS-formatted data include electrode localizations, patient demographics, and clinical outcomes. They support benchmarking of automated seizure onset and spread detection against expert performance and comparison of spontaneous and stimulation-induced seizure networks.

Participants

29

Size

7.95 GB

Version

v1.0.0

Updated

Oct 8, 2026

This dataset contains magnetoencephalography (MEG) recordings from participants performing a cued recall task involving personally familiar people and places. It was collected to examine sustained neural representations of familiar people and places during recall. The data are organized in BIDS format.

Participants

30

Size

76.7 GB

Version

v1.0.0

Updated

Jun 23, 2026

This dataset contains EEG recordings acquired with a BioSemi system in an implicit learning paradigm. It is organized according to the EEG-BIDS specification and was prepared with MNE-BIDS. The data are shared as a raw BIDS dataset under a CC0 license.

Participants

61

Size

114 GB

Version

v1.0.0

Updated

Jun 20, 2026