{"success":true,"database":"eegdash","data":{"_id":"6953f4239276ef1ee07a32cb","dataset_id":"ds003483","associated_paper_doi":null,"authors":["Cognitive and Computational Neuroscience Laboratory (UPM - UCM).","PI: Fernando Maestu.","PI: Carmen Requena","PI: Francisco Salto Alemany"],"bids_version":"BIDS 1.2.0","contact_info":["Luis Fernando Antón Toro"],"contributing_labs":null,"data_processed":true,"dataset_doi":"10.18112/openneuro.ds003483.v1.0.2","datatypes":["meg"],"demographics":{"subjects_count":21,"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/ds003483","osf_url":null,"github_url":null,"paper_url":null},"funding":[],"ingestion_fingerprint":"378c820d0835e2379c2c0bb3f239956c5c944290d6eac8e0723f061289e0fad4","license":"CC0","n_contributing_labs":null,"name":"Logical reasoning study","readme":null,"recording_modality":["meg"],"senior_author":"PI: Francisco Salto Alemany","sessions":["1"],"size_bytes":26285775768,"source":"openneuro","study_design":null,"study_domain":null,"tasks":["deduction","induction"],"timestamps":{"digested_at":"2026-04-22T12:25:32.652566+00:00","dataset_created_at":"2021-01-21T21:01:58.596Z","dataset_modified_at":"2021-01-24T10:47:46.000Z"},"total_files":41,"storage":{"backend":"s3","base":"s3://openneuro.org/ds003483","raw_key":"dataset_description.json","dep_keys":["CHANGES","participants.tsv"]},"tagger_meta":{"config_hash":"3557b68bca409f28","metadata_hash":"2523a8f6391f0f1d","model":"openai/gpt-5.2","tagged_at":"2026-04-07T09:32:40.872789+00:00"},"tags":{"pathology":["Healthy"],"modality":["Unknown"],"type":["Decision-making"],"confidence":{"pathology":0.6,"modality":0.45,"type":0.55},"reasoning":{"few_shot_analysis":"Closest few-shot convention match is the reinforcement-learning datasets labeled as Decision-making (e.g., “EEG: Reinforcement Learning in Parkinson's”), showing that when the primary construct is making/informing choices from information, the Type is mapped to “Decision-making” rather than task mechanics. However, none of the few-shots directly cover logical deduction/induction, and no few-shot provides a reliable cue for stimulus Modality when stimuli are not described.","metadata_analysis":"Key available metadata is very sparse. The dataset is titled \"Logical reasoning study\". The task labels are explicitly \"deduction\" and \"induction\". Participant count is given only as \"Subjects: 21\". No recruitment/diagnosis information, and no stimulus description (visual/auditory/etc.) is provided in the shown metadata.","paper_abstract_analysis":"No useful paper information.","evidence_alignment_check":"Pathology: Metadata says nothing about clinical recruitment (only \"Subjects: 21\"), so we infer a normative cohort from the non-clinical study framing (\"Logical reasoning study\"). Few-shot pattern suggests Healthy is typical for non-clinical cognitive tasks, and there is no conflict with any explicit diagnosis (ALIGN by absence of conflicting facts).\n\nModality: Metadata does not state whether stimuli were visual (text), auditory (spoken problems), etc. Few-shot conventions require using stimulus channel, but here no channel is specified; thus few-shot patterns cannot be safely applied (INSUFFICIENT EVIDENCE; no conflict, just missing info).\n\nType: Metadata explicitly indicates logical reasoning (\"deduction\", \"induction\"). Few-shot conventions map tasks centered on making judgments/choices from information to “Decision-making”; this is compatible with deduction/induction as a form of inferential decision. No metadata contradicts this (ALIGN, but weakly due to lack of detail about the exact cognitive construct emphasis).","decision_summary":"Pathology top-2: (1) Healthy — supported by non-clinical framing (\"Logical reasoning study\") and no stated disorder; (2) Unknown — because no explicit recruitment criteria are provided (only \"Subjects: 21\"). Winner: Healthy (inference consistent with EEGDash convention for generic cognitive studies). Confidence=0.6 because it is contextual inference with no explicit “healthy controls” wording.\n\nModality top-2: (1) Unknown — no stimulus channel described; (2) Visual — deduction/induction tasks are often visually presented as text, but that is not stated. Winner: Unknown. Confidence=0.45 because multiple plausible modalities exist with no direct metadata support.\n\nType top-2: (1) Decision-making — deduction/induction imply selecting/endorsing conclusions (inferential judgment), consistent with few-shot convention for choice/inference-focused paradigms; (2) Other — because the taxonomy lacks an explicit “Reasoning/Executive function” label and the dataset provides no detailed aims. Winner: Decision-making. Confidence=0.55 due to limited task detail beyond the labels \"deduction\" and \"induction\"."}},"nemar_citation_count":3,"computed_title":"Logical reasoning study","nchans_counts":[{"val":320,"count":41}],"sfreq_counts":[{"val":1000.0,"count":41}],"stats_computed_at":"2026-04-22T23:16:00.222088+00:00","total_duration_s":39668.0,"canonical_name":null,"name_confidence":0.38,"name_meta":{"suggested_at":"2026-04-14T10:18:35.342Z","model":"openai/gpt-5.2 + openai/gpt-5.4-mini + deterministic_fallback"},"name_source":"author_year","author_year":"Cognitive2021"}}