Discover datasets

EEG dataset from 11 healthy participants performing a 3-class auditory brain-computer interface task based on auditory stream segregation. Musical tones forming three tone streams were presented to the right ear, and participants attended to one stream while counting target stimuli, which elicited P300 activity. Recordings comprise 64-channel EEG and 2-channel EOG at 1000 Hz. Classification based on Riemannian geometry achieved over 80% accuracy for five subjects and over 75% for nine subjects in 10-fold cross-validation.

Participants

11

Channels

64 (10-10)

Citations

10

HED

v8.4.0

Size

6.31 GB

Version

v1.0.5

Updated

Oct 8, 2026

…Large-scale replication study reveals a limit on probabilistic prediction in language…

Participants

356

Channels

63 (10-10)

Citations

98

Size

103 GB

Version

v1.1.0

Updated

Jul 5, 2026

Study P of the bigP3BCI collection, an open, machine-learning-ready P300-based brain-computer interface dataset. It contains 32-channel EEG recorded at 256 Hz from 19 participants with ALS, each over two sessions, while they used a 9x8 character grid speller with predictive and non-predictive spelling paradigms. Target and non-target flash events are annotated with HED tags, and the data are reformatted from the original PhysioNet release into BIDS.

Participants

19

Channels

32 (10-10)

Citations

2

HED

v8.4.0

Size

4.96 GB

Version

v1.0.4

Updated

Oct 8, 2026

[![DOI](https://img.shields.io/badge/DOI-10.82901%2Fnemar.on007338-blue…

Participants

1

Channels

129 (egi-geodesic)

Citations

7

Size

45.0 MB

Version

v1.0.0

Updated

Jun 30, 2026

EEG recordings from a P300 speller brain-computer interface study comparing flashing paradigms for online typing with language model support. The dataset was collected with BCI2000 to evaluate typing performance across different stimulus presentation schemes. It supports research on P300-based communication interfaces.

Participants

11

Citations

1

Size

1.46 GB

Version

v1.0.0

Updated

May 13, 2026

EEG recordings from 13 healthy participants performing an auditory oddball (P300) paradigm with target (1000 Hz) and non-target (500 Hz) tones at stimulus onset asynchronies between 60 and 600 ms. The study evaluated closed-loop Bayesian optimization and random search to find each individual's best stimulation speed for maximizing classification performance of a regularized LDA. The data comprise 31-channel EEG sampled at 1000 Hz, with an optimization part and a validation part.

Participants

13

Channels

31 (10-10)

Citations

2

HED

v8.4.0

Size

3.68 GB

Version

v1.0.7

Updated

Oct 9, 2026

…TASK Participants passively listened to 30-45s passages of conversational speech and…

Participants

10

Size

13.2 GB

Version

v1.0.0

Updated

Jun 25, 2026

This dataset contains functional near-infrared spectroscopy (fNIRS) recordings collected during a word-color Stroop task using a sparse fNIRS montage. It accompanies a study comparing high-density, multi-distance fNIRS with sparse configurations for detecting task-related brain activity. The data are organized in BIDS format and are mirrored on NEMAR from OpenNeuro.

Participants

17

Size

169 MB

Version

v1.0.0

Updated

Jun 18, 2026

…If you want the speech waveform aligned to the time the subject…

Participants

32

Channels

62 (10-10)

Size

136 GB

Version

v1.0.0

Updated

Jun 27, 2026

NeuroMorph is a magnetoencephalography (MEG) dataset with high temporal resolution, designed for morpheme-based linguistic analysis. It contains recordings from 24 participants, totaling over 17 hours of data, acquired with a 208-channel KIT/Yokogawa magnetometer system. The dataset supports research on the neural processing of morphology during language comprehension.

Participants

24

Size

140 GB

Version

v1.0.0

Updated

Jun 26, 2026