{"success":true,"database":"eegdash","data":{"_id":"6953f4239276ef1ee07a329c","dataset_id":"ds002094","associated_paper_doi":null,"authors":[],"bids_version":"1.3.0","contact_info":["Sara J Hussain"],"contributing_labs":null,"data_processed":false,"dataset_doi":null,"datatypes":["eeg"],"demographics":{"subjects_count":20,"ages":[],"age_min":null,"age_max":null,"age_mean":null,"species":null,"sex_distribution":null,"handedness_distribution":null},"experimental_modalities":null,"external_links":{"source_url":"https://openneuro.org/datasets/ds002094","osf_url":null,"github_url":null,"paper_url":null},"funding":[],"ingestion_fingerprint":"c7d13bee3b3e186c2b6aa64c4fa036998a41a6f9bb5c042221213a16fbd93c96","license":"CC0","n_contributing_labs":null,"name":"Single-pulse open-loop TMS-EEG dataset","readme":null,"recording_modality":["eeg"],"senior_author":null,"sessions":[],"size_bytes":42339775606,"source":"openneuro","study_design":null,"study_domain":null,"tasks":["rest","tmseeg1","tmseeg2"],"timestamps":{"digested_at":"2026-04-22T12:25:25.964875+00:00","dataset_created_at":"2019-08-02T12:56:34.579Z","dataset_modified_at":"2019-08-02T13:56:21.000Z"},"total_files":43,"storage":{"backend":"s3","base":"s3://openneuro.org/ds002094","raw_key":"dataset_description.json","dep_keys":["CHANGES","participants.tsv"]},"nemar_citation_count":30,"computed_title":"Single-pulse open-loop TMS-EEG dataset","nchans_counts":[{"val":30,"count":43}],"sfreq_counts":[{"val":5000.0,"count":43}],"stats_computed_at":"2026-04-22T23:16:00.221518+00:00","tags":{"pathology":["Unknown"],"modality":["Other"],"type":["Clinical/Intervention"],"confidence":{"pathology":0.4,"modality":0.6,"type":0.6},"reasoning":{"few_shot_analysis":"No few-shot example directly matches a TMS-EEG (transcranial magnetic stimulation + EEG) single-pulse, open-loop paradigm. The closest convention guidance comes from the Parkinson’s “Cross-modal Oddball Task” example labeled Type=Clinical/Intervention because it is a clinical-study-style dataset focused on a condition and biomarkers, and from the resting-state examples labeled Modality=Resting State and Type=Resting-state when the paradigm is purely rest. Here, the presence of explicit TMS-EEG tasks (beyond rest) suggests the dataset is not primarily a passive resting-state dataset; by convention, stimulation/perturbation paradigms are best mapped to Type=Clinical/Intervention (intervention/perturbation focus), even if the participants are not stated to be clinical.","metadata_analysis":"Key available metadata is sparse but includes: (1) Title explicitly indicating TMS perturbation: \"Single-pulse open-loop TMS-EEG dataset\". (2) Task list includes both rest and TMS-EEG runs: \"tasks\": [\"rest\", \"tmseeg1\", \"tmseeg2\"]. (3) No diagnostic or recruitment criteria are given: \"participants_overview\": \"Subjects: 20\".","paper_abstract_analysis":"No useful paper information.","evidence_alignment_check":"Pathology: Metadata says only \"Subjects: 20\" with no mention of patients/diagnosis; few-shot conventions would label as Healthy only when healthy/non-clinical recruitment is explicit. ALIGNMENT: aligns with choosing Unknown (insufficient evidence).\n\nModality: Metadata indicates TMS-EEG perturbation (\"Single-pulse... TMS-EEG\") plus a \"rest\" task; few-shot conventions for modality are based on stimulus/input channel (auditory/visual/tactile/motor/rest). TMS is not an allowed modality label; the closest allowed label is \"Other\" (perturbation/neuromodulation), with \"Resting State\" as a runner-up because one task is \"rest\". PARTIAL CONFLICT/AMBIGUITY: metadata supports both rest and TMS; selecting \"Other\" prioritizes the dominant distinctive manipulation (TMS).\n\nType: Metadata indicates a perturbation/stimulation protocol (\"TMS-EEG\", \"tmseeg1\", \"tmseeg2\") rather than a cognitive task; few-shot conventions map purely passive recordings to Resting-state, and clinical biomarker/intervention-style datasets to Clinical/Intervention. ALIGNMENT: although no clinical population is specified, TMS is an intervention/perturbation method, so Clinical/Intervention best matches the study purpose category given the available labels.","decision_summary":"Top-2 candidates per category:\n\nPathology:\n1) Unknown — Evidence: only \"Subjects: 20\"; no diagnosis/control/healthy wording.\n2) Healthy — Weak inference only (many small-sample physiology datasets are healthy), but not stated.\nDecision: Unknown (metadata lacks explicit recruitment/pathology). Confidence justification: no supporting quotes for any specific pathology.\n\nModality:\n1) Other — Evidence: \"Single-pulse open-loop TMS-EEG dataset\" implies non-sensory perturbation; tasks include \"tmseeg1\", \"tmseeg2\".\n2) Resting State — Evidence: tasks include \"rest\".\nDecision: Other (TMS perturbation is the distinctive input/manipulation; rest appears as an additional condition). Confidence justification: inferred mapping because TMS is not a standard sensory modality label; supported by title + task names.\n\nType:\n1) Clinical/Intervention — Evidence: TMS is an intervention/perturbation method (\"TMS-EEG\", \"tmseeg1\", \"tmseeg2\").\n2) Resting-state — Evidence: inclusion of a \"rest\" task.\nDecision: Clinical/Intervention (dataset appears primarily organized around TMS-EEG perturbation recordings rather than only resting-state). Confidence justification: contextual inference supported by two metadata snippets (title and tasks), but no abstract/protocol details."}},"total_duration_s":70564.7194,"tagger_meta":{"config_hash":"3557b68bca409f28","metadata_hash":"49107fea12e79dc8","model":"openai/gpt-5.2","tagged_at":"2026-04-07T09:32:40.872789+00:00"},"author_year":"DS2094_Single_pulse","canonical_name":null}}