{"success":true,"data":{"author":"Sun Xinwei, Wang Kun, Pan Lincong, Cao Yupei, Meng Lin (Tianjin University)","bucketAccessType":"public","bucketListable":true,"bucketName":"openneuro.org","bucketPrefix":"ds007221","category":"Neuroscience","cloudProvider":"aws","createdAt":"2026-11-16T14:11:23.396Z","description":"64-channel EEG from 84 healthy participants (ages 20 to 32, 49 female) performing motor imagery (MI) brain-computer interface tasks in two environments: a simulated hospital ward with visual, auditory, olfactory and tactile cues, and an electromagnetically shielded laboratory, in counterbalanced order (ses per environment). Four sub-datasets with separate participants: Graz (14; left hand, right hand, feet MI and rest), SSMVEP-MI (22; hand MI with action observation of periodic hand-action videos), Hybrid (37; hand MI under Graz, SSMVEP, Video and SSVideo stimulus conditions) and Hybrid-online (11; offline calibration plus online feedback). Recorded with a Neuroscan SynAmps2 at 1000 Hz with a 2 s cue, 4 s task, 4 s rest trial structure; raw and preprocessed EEG plus questionnaires. BIDS layout, 10,073 files, about 134 GB; load with MNE-BIDS. Useful for MI decoding, cross-subject and cross-environment transfer learning and SSMVEP hybrid BCI. The OpenNeuro release is CC0; cite Sun et al. and the dataset DOI.","displayTitle":"Cross-Environment Multi-Paradigm Motor Imagery EEG Dataset (84 Participants, Hospital Ward vs Lab)","doi":"10.18112/openneuro.ds007221.v1.0.3","fuseMountBrowsable":true,"id":"a3925368-6b9c-5eb6-a8af-37a3e98fa8b9","license":"CC0-1.0","licenseEvidenceUrl":"https://openneuro.org/datasets/ds007221","objectCountEstimate":86,"probeStatus":"ok","referenceUrl":"https://github.com/boom-sun/Cross-Environment-Multi-Paradigm-Motor-Imagery-EEG-Dataset","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='ds007221/', Delimiter='/')"},{"language":"python-local","snippet":"open('/mnt/archive/.datalad/config', 'rb')"}],"caveats":["Bucket has a large top-level fan-out — never recursively list; scope LIST operations by prefix.","Files are binary — use a domain-specific parser (e.g. metpy / pynexrad / pydicom) rather than text I/O."],"format":"binary","generatedAt":"2026-11-16T14:11:26.766Z","layout":{"example":"ds007221/.datalad/config","pattern":"{prefix}/{seg1}/{object}"},"mountPath":"/mnt/archive","sampleHeaders":[{"bytes":"[datalad \"dataset\"]\n\tid = 23f3170d-63d1-43e9-b09f-cab9735cd66e\n","encoding":"utf8","key":"ds007221/.datalad/config","truncatedAt":512}],"sampleObjects":["ds007221/.datalad/config"],"sidecars":[{"bytes":"# Cross-Environment Multi-Paradigm Motor Imagery EEG Dataset\n\n## Overview\n\nThis dataset contains electroencephalography (EEG) recordings from **84 healthy participants**, each recorded in two environments: a **simulated hospital ward** and a **controlled laboratory**. It comprises four sub-datasets covering multiple motor imagery (MI) paradigms and both offline and online recordings.\n\nThe dataset supports research on MI decoding, cross-subject transfer, cross-environment generalization, and the effects of visual stimulation on MI-based brain-computer interfaces (BCIs). Raw recordings, preprocessed EEG, and supporting questionnaire materials are provided.\n\n- **Data:** [OpenNeuro ds007221](https://openneuro.org/datasets/ds007221)\n- **Dataset DOI (version 1.0.3):** [10.18112/openneuro.ds007221.v1.0.3](https://doi.org/10.18112/openneuro.ds007221.v1.0.3)\n- **Analysis code:** [GitHub](https://github.com/boom-sun/Cross-Environment-Multi-Paradigm-Motor-Imagery-EEG-Dataset)\n\n## Participants and experimental design\n\nParticipants were aged 20-32 years (mean 23.18 years; 49 female). Each participant completed recordings in both environments, with environment order counterbalanced across participants. The four sub-datasets contain separate participant groups.\n\n| Sub-dataset | Participants | Tasks and paradigms |\n| --- | ---: | --- |\n| Graz | 14 | Left-hand MI, right-hand MI, feet MI, and rest |\n| SSMVEP-MI | 22 | Left- and right-hand MI and action observation (AO), using periodic hand-action video stimuli |\n| Hybrid | 37 | Left- and right-hand MI under four stimulus conditions: Graz, SSMVEP, Video, and SSVideo |\n| Hybrid-online | 11 | The same four stimulus conditions, with offline calibration and online feedback |\n\n**Scene 1** was a simulated hospital ward incorporating visual, auditory, olfactory, and tactile contextual cues. **Scene 2** was a laboratory within an electromagnetically shielded enclosure. These recordings were collected from healthy volunteers, not a clinical patient cohort.\n\nThe manuscript describes a trial sequence of a 2-second cue, a 4-second task, and a 4-second rest period. Use recording-specific event files to determine marker timing and class labels.\n\nThe study was approved by the ethics committee of Tianjin University. All participants provided written informed consent, including consent to share de-identified data.\n\n## EEG acquisition and preprocessing\n\nThe acquisition protocol used a SynAmps2 system with a 64-channel Quick-Cap, with 60 scalp channels selected for analysis. Raw EEG was sampled at **1000 Hz**, referenced to the **tip of the nose**, with a forehead ground and a 50 Hz acquisition notch filter.\n\nThe preprocessed derivatives contain 60-channel EEG at **250 Hz**. Processing included a 49-51 Hz notch filter, common-average re-referencing, an approximately 4 Hz high-pass filter, downsampling, and epoching from **-2 to 4 seconds relative to task onset**.\n\nSome raw sidecars currently differ from the manuscript in manufacturer and channel-count fields. Check recording-specific channel files and headers before analysis; the number of stored raw channels may differ from the 60-channel analysis montage.\n\n## Data organization\n\nRaw EEG is organized using BIDS conventions, with BrainVision recordings (`.vhdr`, `.vmrk`, `.eeg`) and accompanying event, channel, and acquisition metadata.\n\n| File or directory | Contents |\n| --- | --- |\n| `sub-*/ses-*/eeg/` | Raw EEG and recording-specific metadata |\n| `participants.tsv` | Demographics and sub-dataset membership (`source`) |\n| `recordings.tsv` | Recording inventory, including task, paradigm, run, and environment-order information |\n| `derivatives/` | Preprocessed MATLAB files grouped into Graz, SSMVEP-MI, Hybrid, and Hybrid-online |\n| `Questionnaire.xlsx` | Questionnaire texts and raw STAI responses |\n\nDerivative `.mat` files contain `data` (channels x time points x trials), `labels` (task-trigger labels), and `fs` (250 Hz). Their filenames encode participant and scene indices, with additional paradigm labels for Hybrid and Hybrid-online. Verify participant corr","encoding":"utf8","key":"ds007221/README.md","name":"README.md","size":5997,"truncated":true}],"sourceObjectKey":"ds007221/.datalad/config","starterCell":"with open('/mnt/archive/.datalad/config', 'rb') as fh:\n    head = fh.read(1024)\nprint(len(head), head[:64])","totalObjects":"86 top-level prefixes"},"summary":"64-channel EEG from 84 healthy participants (ages 20 to 32, 49 female) performing motor imagery (MI) brain-computer interface tasks in two environments: a simulated hospital ward…","title":"Cross-Environment Multi-Paradigm Motor Imagery EEG Dataset (84 Participants, Hospital Ward vs Lab)","updatedAt":"2026-11-16T16:20:08.312Z","visibility":"public","websiteUrl":"https://openneuro.org/datasets/ds007221"}}