{"success":true,"database":"eegdash","data":{"_id":"6953f4249276ef1ee07a33ea","dataset_id":"ds005473","associated_paper_doi":null,"authors":["Zhao Xiangyue","Zhou Jingyao","Zhang Libo","Duan Haoqing","Wei Shiyu","Bi Yanzhi","Hu Li"],"bids_version":"1.1.1","contact_info":["Xiangyue Zhao"],"contributing_labs":null,"data_processed":true,"dataset_doi":"doi:10.18112/openneuro.ds005473.v1.0.0","datatypes":["eeg"],"demographics":{"subjects_count":29,"ages":[18,23,23,20,22,23,22,19,22,20,22,19,25,22,22,23,20,23,23,19,23,25,21,21,20,22,20,21,21],"age_min":18,"age_max":25,"age_mean":21.517241379310345,"species":null,"sex_distribution":{"f":13,"m":16},"handedness_distribution":null},"experimental_modalities":null,"external_links":{"source_url":"https://openneuro.org/datasets/ds005473","osf_url":null,"github_url":null,"paper_url":null},"funding":[],"ingestion_fingerprint":"6ae275193cc28da33e66edc0e01aa7cf19bdace277a266f68c5f1fa43b8e1887","license":"CC0","n_contributing_labs":null,"name":"29 By BP","readme":null,"recording_modality":["eeg"],"senior_author":"Hu Li","sessions":["1","2"],"size_bytes":6637997774,"source":"openneuro","study_design":null,"study_domain":null,"tasks":["29ByBP"],"timestamps":{"digested_at":"2026-04-22T12:27:46.585275+00:00","dataset_created_at":"2024-09-11T04:15:49.006Z","dataset_modified_at":"2024-09-11T10:00:36.000Z"},"total_files":58,"storage":{"backend":"s3","base":"s3://openneuro.org/ds005473","raw_key":"dataset_description.json","dep_keys":["CHANGES","README","participants.json","participants.tsv","task-29ByBP_events.json"]},"tagger_meta":{"config_hash":"3557b68bca409f28","metadata_hash":"60f1c6beea8311ae","model":"openai/gpt-5.2","tagged_at":"2026-04-07T09:32:40.872789+00:00"},"tags":{"pathology":["Healthy"],"modality":["Unknown"],"type":["Unknown"],"confidence":{"pathology":0.6,"modality":0.4,"type":0.4},"reasoning":{"few_shot_analysis":"Few-shot examples illustrate a convention: when metadata explicitly states a clinical recruitment group (e.g., “Parkinsons disease”, “schizophrenia spectrum disorder”, “mTBI”, “Alzheimer’s disease”), Pathology is set to that label; when no disorder/clinical recruitment is stated (e.g., examples with only general participant descriptions), Pathology defaults to Healthy. They also show that Modality/Type are derived from described stimuli/paradigm (e.g., “auditory oddball”→Auditory; “resting state-closed eyes”→Resting State/Resting-state). For this dataset, there is no task/stimulus description, so the few-shot convention suggests using Unknown for Modality and Type when facts are missing.","metadata_analysis":"Available metadata is extremely sparse. Key snippets: (1) Title: \"29 By BP\". (2) Participants: \"Subjects: 29; Sex: {'f': 13, 'm': 16}; Age range: 18-25\". (3) Task label only: \"tasks\": [\"29ByBP\"]. There are no statements about diagnosis/clinical recruitment, no stimulus description, and no indication of resting-state vs task-based recording beyond an opaque task name.","paper_abstract_analysis":"No useful paper information.","evidence_alignment_check":"Pathology: Metadata says only general demographics (\"Age range: 18-25\") and does NOT mention any patient/diagnosis group. Few-shot pattern suggests that absence of a clinical recruitment statement implies a normative cohort → Healthy. ALIGN.\nModality: Metadata provides no stimulus channel information beyond an uninterpretable task name (\"29ByBP\"). Few-shot pattern requires explicit task/stimulus description to label modality (auditory/visual/resting/etc.). With no facts, use Unknown. ALIGN.\nType: Metadata provides no cognitive construct/task aim (only \"29ByBP\" task label). Few-shot pattern similarly depends on explicit paradigm description (oddball, digit span, motor imagery, resting, sleep). With no facts, use Unknown. ALIGN.","decision_summary":"Pathology top-2: (A) Healthy — supported by lack of any diagnosis/patient language and only general demographics: \"Subjects: 29... Age range: 18-25\"; consistent with few-shot convention that non-clinical unspecified cohorts are labeled Healthy. (B) Unknown — possible if recruitment/condition is simply undocumented. Head-to-head: Healthy is slightly stronger given typical OpenNeuro demographic-only entries representing healthy participants, but evidence is not explicit → moderate confidence.\nModality top-2: (A) Unknown — no stimulus/sensory channel described (only \"tasks\": [\"29ByBP\"]). (B) Resting State — could be plausible in many EEG datasets, but there is zero supporting text (no \"eyes open/closed\" etc.). Head-to-head: Unknown clearly wins → low confidence but appropriate.\nType top-2: (A) Unknown — no construct/task aim described. (B) Resting-state — plausible alternative but unsupported. Head-to-head: Unknown wins.\nConfidence justification: Pathology has only contextual inference from absence of clinical terms (no explicit “healthy”) → 0.6. Modality/Type have no supporting facts at all → 0.4 each and set to Unknown."}},"computed_title":"29 By BP","nchans_counts":[{"val":64,"count":58}],"sfreq_counts":[{"val":1000.0,"count":58}],"stats_computed_at":"2026-04-22T23:16:00.309632+00:00","total_duration_s":52673.0,"canonical_name":null,"name_confidence":0.42,"name_meta":{"suggested_at":"2026-04-14T10:18:35.343Z","model":"openai/gpt-5.2 + openai/gpt-5.4-mini + deterministic_fallback"},"name_source":"author_year","author_year":"Xiangyue2024_29_BP"}}