{"success":true,"database":"eegdash","data":{"_id":"6953f4249276ef1ee07a33b2","dataset_id":"ds005065","associated_paper_doi":null,"authors":["Evan M. Russek","Rani Moran","Yunzhe Liu","Ray Dolan","Quentin Huys"],"bids_version":"v1.5.0","contact_info":["Evan Russek"],"contributing_labs":null,"data_processed":true,"dataset_doi":"doi:10.18112/openneuro.ds005065.v1.0.0","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/ds005065","osf_url":null,"github_url":null,"paper_url":null},"funding":[],"ingestion_fingerprint":"765549ff825e057bcb35e92470d348826234776071123fb80a873f3d96b74753","license":"CC0","n_contributing_labs":null,"name":"Heuristics in risky decision-making relate to preferential representation of information MEG data","readme":"The task consisted of 13 scanner runs (except for subject 1 who completed 5 rather than 3 localizer runs). Runs 1-3 (1-5 for subject 1) are the localizer task. Runs 4-5 are non-analyzed data from the 'probability learning' task. Runs 6-13 (8-15 for subject 1) are the risky decision-making task.\nEvent times were recorded with a photodiode, which is accessible as a MEG channel. This has been processed so that event times are listed in derivatives/Event_Info_Tables. Raw times of events in the scan are in column \"onset_time\". The corresponding index into the unprocessed MEG data is in column \"scanner_onset_idx\". The onset into the downsampled data is in \"onset_idx_ds\". In the table, each row corresponds to an event. Block number denotes which scanner run that event belongs to. For the localizer task (denoted in phase column), events are image onsets. \"image_type\" specifies the role of that image in the task (\"CHOICE\" or \"OUTCOME\") and \"image_number\" denotes which choice or outcome it is (see paper Fig. 1). Finally, \"image_name\" denotes which image category was shown (e.g. \"Hand\"). For the task, events correspond to gamble information onset (Info), Probability stimulus presentation (\"Choice\"), response (\"Gamble Response\") and outcome onset (\"Outcome\"). Columns denote which image was shown and what the response was (accept).\nderivatives/Epoched_Data contains epoched preprocessed data for each subject for the localizer task and then around each choice in the main choice task. Both are epoched from from 0-500 ms following the event.\nCode to analyze the data along with additional behavioral data is available at https://github.com/evanrussek/MEG_Heuristics_Risk_Preferential_Information","recording_modality":["meg"],"senior_author":"Quentin Huys","sessions":[],"size_bytes":457153490815,"source":"openneuro","study_design":null,"study_domain":null,"tasks":["RiskyDecision"],"timestamps":{"digested_at":"2026-04-22T12:27:16.368840+00:00","dataset_created_at":"2024-04-07T02:26:32.589Z","dataset_modified_at":"2024-04-09T01:21:46.000Z"},"total_files":275,"storage":{"backend":"s3","base":"s3://openneuro.org/ds005065","raw_key":"dataset_description.json","dep_keys":["CHANGES","README","participants.tsv"]},"tagger_meta":{"config_hash":"4a051be509a0e3d0","metadata_hash":"ca97cfb865b264a2","model":"openai/gpt-5.2","tagged_at":"2026-01-20T17:40:11.800942+00:00"},"tags":{"pathology":["Healthy"],"modality":["Visual"],"type":["Decision-making"],"confidence":{"pathology":0.6,"modality":0.7,"type":0.8},"reasoning":{"few_shot_analysis":"Most similar few-shot convention is the \"EEG: Three armed bandit gambling task\" example (Healthy + Visual + Affect). That example shows that gambling/reward/outcome paradigms are labeled with a decision-related Type rather than a perceptual Type. Another close convention match is \"EEG: Reinforcement Learning in Parkinson's\" (Visual + Decision-making), where choice and feedback/outcome structure drives labeling as Decision-making. Here, metadata explicitly describes a \"risky decision-making task\" with gamble info/choice/response/outcome events, which by few-shot convention maps best to Type=Decision-making and Modality=Visual (images/stimuli shown).","metadata_analysis":"Key task/purpose facts from the dataset README:\n1) Task purpose: \"Runs 6-13 ... are the risky decision-making task.\" This explicitly indicates a decision-making paradigm.\n2) Stimulus format: \"For the localizer task... events are image onsets\" and \"Finally, \\\"image_name\\\" denotes which image category was shown (e.g. \\\"Hand\\\").\" This indicates visual stimuli.\n3) Decision structure: \"For the task, events correspond to gamble information onset (Info), Probability stimulus presentation (\\\"Choice\\\"), response (\\\"Gamble Response\\\") and outcome onset (\\\"Outcome\\\").\" This aligns with gambling/risk choice and outcome feedback.","paper_abstract_analysis":"No useful paper information.","evidence_alignment_check":"Pathology:\n- Metadata says: no diagnosis/clinical recruitment is mentioned (no pathology keywords in provided README).\n- Few-shot pattern suggests: when no disorder is described, label as Healthy.\n- Alignment: ALIGN (absence of clinical population in metadata fits Healthy by convention).\n\nModality:\n- Metadata says: \"events are image onsets\" and \"image_name denotes which image category was shown\".\n- Few-shot pattern suggests: image-based tasks are Visual.\n- Alignment: ALIGN.\n\nType:\n- Metadata says: \"risky decision-making task\" and events include \"Choice\", \"Gamble Response\", and \"Outcome\".\n- Few-shot pattern suggests: gambling/choice-with-outcome paradigms map to Decision-making (see bandit / reinforcement learning examples).\n- Alignment: ALIGN.","decision_summary":"Top-2 candidates per category with head-to-head selection:\n\nPathology:\n- Candidate 1: Healthy\n  Evidence: no clinical recruitment described in README; task-focused description only.\n- Candidate 2: Unknown\n  Evidence: participants are not described at all in provided metadata.\nDecision: Healthy (stronger default when no disorder is mentioned; aligns with few-shot convention).\nConfidence justification: inference-only from absence of pathology statements → moderate (0.6).\n\nModality:\n- Candidate 1: Visual\n  Evidence: \"events are image onsets\"; \"image_name denotes which image category was shown\".\n- Candidate 2: Other\n  Evidence: decision task includes \"Probability stimulus presentation\" without explicitly stating modality, but localizer clearly uses images.\nDecision: Visual (explicitly image-based).\nConfidence justification: 2 direct stimulus-related quotes → 0.7.\n\nType:\n- Candidate 1: Decision-making\n  Evidence: explicit phrase \"risky decision-making task\"; gamble pipeline with \"Choice\"/\"Gamble Response\"/\"Outcome\".\n- Candidate 2: Learning\n  Evidence: mention of \"probability learning\" runs, but they are \"non-analyzed\" and not the main described task focus.\nDecision: Decision-making (risky gambling choices are primary analyzed paradigm; matches few-shot gambling/RL conventions).\nConfidence justification: multiple explicit task-purpose and event-structure quotes + strong few-shot analog → 0.8."}},"nemar_citation_count":1,"computed_title":"Heuristics in risky decision-making relate to preferential representation of information MEG data","nchans_counts":[{"val":415,"count":210},{"val":341,"count":65}],"sfreq_counts":[{"val":1200.0,"count":272}],"stats_computed_at":"2026-04-22T23:16:00.308905+00:00","total_duration_s":244800.0,"author_year":"Russek2024","canonical_name":null}}