{"success":true,"data":{"author":"Paras Qadir Memon, Shoaib Memon, Walter G. Besio, Charles W. Anderson","bucketAccessType":"public","bucketListable":true,"bucketName":"openneuro.org","bucketPrefix":"ds008895","category":"Neuroscience","cloudProvider":"aws","createdAt":"2026-11-19T15:09:22.776Z","description":"EEG from 16 adults recorded with tripolar concentric ring electrodes (TCRE) while they named pictures aloud (overt speech) or silently (covert, imagined speech). Each TCRE gives two co-located signals, a tripolar estimate of the surface Laplacian and an emulated conventional disc signal from the outer ring, so Laplacian and conventional EEG can be compared at the same scalp sites. Electrodes form four arrays centered on F4, P6, F3 and P5 (10-10 system), with 5 to 8 electrodes per array depending on the participant, plus EOG on some participants. Each participant did 4 runs of 50 trials (2 overt, 2 covert): a picture for 3.5 s, then 2.5 s of rest. Recording used a BrainAmp amplifier and BrainVision Recorder at 1000 or 2500 Hz. The BIDS release (133 files, about 10 GB) has BrainVision .eeg/.vhdr/.vmrk files, events.tsv, participants.tsv with TCRE size and run order, code/ scripts, and derivatives with cleaned epochs and 500 Hz resampled recordings in MNE .fif. Load with mne_bids.read_raw_bids. For speech BCI, imagined speech decoding and Laplacian EEG methods. CC0 on OpenNeuro; cite Memon et al. (Sensors, doi 10.3390/s26134084) and the dataset DOI.","displayTitle":"Overt and Covert Speech EEG with Tripolar Concentric Ring Electrodes (TCRE Laplacian EEG)","doi":"10.18112/openneuro.ds008895.v1.1.0","fuseMountBrowsable":true,"id":"fa5a2bb1-f1e5-5c21-b7f6-546068c9970b","license":"CC0-1.0","licenseEvidenceUrl":"https://openneuro.org/datasets/ds008895","objectCountEstimate":19,"probeStatus":"ok","referenceUrl":"https://doi.org/10.3390/s26134084","region":"us-east-1","schemaCard":{"accessHints":[{"language":"python-anonymous","snippet":"import boto3\nfrom botocore import UNSIGNED\nfrom botocore.client import Config\ns3 = boto3.client('s3', region_name='us-east-1', config=Config(signature_version=UNSIGNED))\ns3.list_objects_v2(Bucket='openneuro.org', Prefix='ds008895/', Delimiter='/')"},{"language":"python-local","snippet":"open('/mnt/archive/.datalad/config', 'rb')"}],"caveats":["Files are binary — use a domain-specific parser (e.g. metpy / pynexrad / pydicom) rather than text I/O."],"format":"binary","generatedAt":"2026-11-19T15:21:46.662Z","layout":{"example":"ds008895/.datalad/config","pattern":"{prefix}/{seg1}/{object}"},"mountPath":"/mnt/archive","sampleHeaders":[{"bytes":"[datalad \"dataset\"]\n\tid = 14ecd8a7-64c7-4549-8c8f-678a6ead2b99\n","encoding":"utf8","key":"ds008895/.datalad/config","truncatedAt":512}],"sampleObjects":["ds008895/.datalad/config"],"sidecars":[{"bytes":"An EEG Dataset of Overt and Covert Speech Recorded with Tripolar Concentric Ring Electrodes\n============================================================================================\n\nOverview\n--------\nEEG from 16 adult participants recorded with tripolar concentric ring electrodes (TCRE) while they\nnamed pictures aloud (overt speech) or silently (covert speech). Every TCRE provides two co-located\nsignals: a tripolar signal (an estimate of the surface Laplacian) and an emulated conventional disc\nsignal (recorded from the outer ring). The dataset therefore allows a direct comparison of Laplacian\nand conventional EEG recorded at the same scalp positions.\n\nParticipants\n------------\n16 adults (sub-01 to sub-16). participants.tsv lists gender, handedness,\nlanguage background, TCRE size (4 or 10 mm outer diameter), electrodes per array, EOG channels\nand run order. Participants are identified only by their BIDS labels.\n\nTask\n----\n4 runs of 50 trials per participant (200 trials), one condition per run: 2 overt runs and 2 covert\nruns; the run order differs between participants (participants.tsv, run_order).\nEach trial: a picture shown for 3.5 s, during which the participant named it aloud (overt) or\nsilently (covert), followed by a 2.5 s grey screen (rest).\nEvents: one marker at each picture onset, labelled Stimulus/Overt or Stimulus/Covert\n(events.tsv). The grey screen began 3.5 s after each marker; it has no marker of its own.\n\nRecording\n---------\nBrainAmp amplifier and BrainVision Recorder (Brain Products GmbH). Reference and ground on the right\nmastoid. Sampling rate 1000 Hz (sub-01) or 2500 Hz (all others); resolution 0.1 uV.\nHardware filters: high-pass 10 s time constant (0.016 Hz); low-pass 1000 Hz (40/42-channel recordings)\nor 250 Hz (56/64-channel recordings). No notch or software filter during recording.\nPower-line frequency 60 Hz.\n\nChannels\n--------\nTCRE k is recorded on channels 2k-1 and 2k:\n  odd channel  (2k-1) = tripolar (TCRE Laplacian) signal\n  even channel (2k)   = emulated disc signal of the same electrode\nThe channel names are placeholder labels from the recording software; they do NOT give the\nelectrode positions. Positions are given by the array layout below.\n\nThe electrodes form four arrays centred on F4, P6, F3 and P5 (10-10 system). Channels are ordered\nby array (F4, P6, F3, P5) and, within an array, from the lowest (most inferior) electrode upward.\n\n  Participants                       Electrodes per array   Array channels   EOG channels\n                                     (F4/P6/F3/P5)\n  sub-01                             5/5/5/5                1-40             none\n  sub-07, 08, 09, 14, 15, 16         5/5/5/5                1-40             41-42\n  sub-05, 06                         7/7/7/7                1-56             none\n  sub-02, 03, 04, 10                 8/8/8/8                1-64             none\n  sub-11, 12, 13                     8/7/8/8                1-62             63-64\n\nEOG channels: two standard disc electrodes (not TCREs), placed above and to the right of one eye,\nrecorded as the last two channels. They are labelled EOG in channels.tsv, so\nmne_bids.read_raw_bids loads them as EOG and picks=\"eeg\" excludes them.\n\nCode\n----\ncode/make_cleaned_epochs.py (with code/channel_sets.py) generates derivatives/cleaned-epochs and\ncode/make_resampled.py generates derivatives/resampled-500hz from the raw data.\ncode/artifact_ica_ssp.py applies optional ICA or SSP artifact removal to the resampled recordings,\nseparately for the TCRE and disc signals. The analysis code of the data descriptor is available at\nhttps://github.com/parasmemon/tcre-overt-covert-speech-eeg\n\nDerivatives\n-----------\nderivatives/cleaned-epochs: epoched, cleaned recordings generated by code/make_cleaned_epochs.py\n(MNE .fif).\n\nderivatives/resampled-500hz: continuous recordings of all channels, including EOG, resampled to\n500 Hz (MNE .fif, polyphase resampling with anti-aliasing filter). No notch or band-pass filter\nand no rejection. See derivatives/resampled-500hz/README.\n\n\nLoading\n-------\n  from mne_bids import BIDSPath, rea","encoding":"utf8","key":"ds008895/README","name":"README","size":5608,"truncated":true}],"sourceObjectKey":"ds008895/.datalad/config","starterCell":"with open('/mnt/archive/.datalad/config', 'rb') as fh:\n    head = fh.read(1024)\nprint(len(head), head[:64])","totalObjects":"19 top-level prefixes"},"summary":"EEG from 16 adults recorded with tripolar concentric ring electrodes (TCRE) while they named pictures aloud (overt speech) or silently (covert, imagined speech).","title":"Overt and Covert Speech EEG with Tripolar Concentric Ring Electrodes (TCRE Laplacian EEG)","updatedAt":"2026-11-19T18:52:24.778Z","visibility":"public","websiteUrl":"https://openneuro.org/datasets/ds008895"}}