2026-09-01 00:00 France United States Papers Foundations & Methods Translated from EN

NETBCI Dataset Pairs MEG and EEG from 19 Users across Four BCI Training Sessions

Summary A team in France and the United States has released NETBCI, a longitudinal multimodal dataset pairing magnetoencephalography (MEG) and electroencephalography (EEG) from 19 healthy subjects across 4 sessions performed on 4 different days, built to study how brain networks reorganize during brain-computer interface training. Controlling a BCI remains a learned skill that a non-negligible proportion of users never acquire even after several sessions, the authors write, and the causes of that inter-individual variability remain an open question. Each session comprises 2 eyes-open resting-state recordings of 3 minutes each plus 6 runs in which participants either sustained right-hand motor imagery or stayed at rest to control the position of a virtual cursor, and the release also includes anonymized MRI scans and behavioral scores; the authors say they hope the sample size and range of modalities will support analyses beyond brain network reorganization.

Why it matters BCI illiteracy is usually reported as a rate, rarely explained, partly because few datasets follow the same users across training with more than one imaging modality. Pairing MEG and EEG with MRI and behavioral scores over 4 sessions makes the reorganization question tractable rather than merely observable. Sample size is modest at 19 subjects, and all participants are healthy, so clinical populations remain out of scope.

BCIwiki (bciwiki.com) – The paper was published in Scientific Data on September 1, 2026, by researchers at Sorbonne Universite and the Paris Brain Institute in France and the University of Pennsylvania in the United States. The team is releasing the Networks for BCI (NETBCI) dataset to help identify the brain network reorganization that underlies BCI training and to support better BCI systems. Controlling a BCI remains a learned skill that a non-negligible proportion of people cannot acquire after several sessions, the authors write, and identifying the causes of that inter-individual variability is still an open avenue.

NETBCI contains magnetoencephalographic (MEG) and electroencephalographic (EEG) recordings from 19 healthy subjects across 4 sessions performed on 4 different days, according to the paper. Each session comprises 2 resting-state recordings of 3 minutes each with eyes open, and 6 runs of a BCI experiment in which the participant either performs sustained right-hand motor imagery or remains at rest to control the position of a virtual cursor. The dataset also comprises anonymized MRI and behavioral scores.

The authors say they hope NETBCI can be used for analysis beyond the investigation of brain network reorganization during BCI training, and that the sample size and number of modalities should be useful to the research community. The recordings come from healthy subjects and do not cover patient populations.

Compiled by BCIwiki from public sources

Sources · 1

doi.org 2026-09-01

Inserm Paris Brain Institute Sorbonne Université University of Pennsylvania EEG Motor Decoding Non-invasive BCI France United States

From these organizations

2026-08 Inserm Stanford's Palanker Wins Defense Health Agency Award for PRIMA Retinal Implant

Inserm → Paris Brain Institute → Sorbonne Université → University of Pennsylvania →

Read original ↗

Suggest a correction Revisions · none 中 / EN