{"success":true,"database":"eegdash","data":{"_id":"6953f4239276ef1ee07a32af","dataset_id":"ds002761","associated_paper_doi":null,"authors":["G. Elliott Wimmer","Yunzhe Liu","Neža Vehar","Timothy E.J. Behrens","Raymond J. Dolan"],"bids_version":"","contact_info":["Elliott Wimmer"],"contributing_labs":null,"data_processed":false,"dataset_doi":"doi:10.18112/openneuro.ds002761.v1.1.2","datatypes":["meg"],"demographics":{"subjects_count":25,"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/ds002761","osf_url":null,"github_url":null,"paper_url":null},"funding":["R.D. is supported by Wellcome Trust Investigator Award 098362/ Z/12/Z.","Y.L. is supported by a UCL Graduate Research Scholarship and an Overseas Research Scholarship.","The Wellcome Centre for Human Neuroimaging is supported by core funding from the Wellcome Trust (203147/Z/16/Z).","The Max Planck University College London Centre is a joint initiative supported by University College London and the Max Planck Society."],"ingestion_fingerprint":"79589f63aee29ace0446e728f3efc81d5d6b1f2a9a617ee83a43d81686f31d5d","license":"CC0","n_contributing_labs":null,"name":"memoryreplay","readme":"The MEG files contain a channel with triggers necessary for event marking and timing. Separate event files with onsets are provided in the participant directories for completeness only; the MEG triggers should be used for actual onsets in analysis. The delay between the trigger and the visual onset of an on-screen event sent by the projector is approximately 20 ms, as estimated using a photodiode.\nMemory phase triggers: At the onset of a trial, the first trigger represents the category (1-8) of the on-screen image. Categories 1-6 represent actual stimulus categories. Trigger values of 7 and 8 represent the 4 positive and 4 negative story-ending stimuli, respectively. The onset of the answer, approximately 5.5 sec later, is marked by a trigger value of 11.\nLocalizer phase triggers: As in the memory phase, at the onset of a trial, the first trigger represents the category (1-8) of the on-screen image. Categories 1-6 represent true categories. Trigger values of 7 and 8 represent the 4 positive and 4 negative story-ending stimuli, respectively. For a baseline, note that for the 2 s prior to picture onset, a word naming that picture was presented on the screen; thus, baseline values should be taken from data more than 2 s before the trigger onset.\nMethods note:  a sequenceness analysis step was omitted from the published 2020 Nature Neuroscience paper. The text should have read:\n\"We next asked whether the βi(Δt) was consistent with a specified 6 × 6 transition matrix by taking the Frobenius inner product between these two matrices (the sum of element-wise products of the two matrices). This resulted in a single number ZΔt, which pertained to lag Δt. For each trial, sequenceness results were then z-scored across lags. Finally, differential forward – backward sequenceness was defined as ZfΔt − ZbΔt.\"","recording_modality":["meg"],"senior_author":"Raymond J. Dolan","sessions":[],"size_bytes":1758242,"source":"openneuro","study_design":null,"study_domain":null,"tasks":["loc","mem"],"timestamps":{"digested_at":"2026-04-21T23:06:04.488102+00:00","dataset_created_at":"2020-04-29T12:47:53.734Z","dataset_modified_at":"2022-11-18T15:26:06.000Z"},"total_files":249,"storage":{"backend":"s3","base":"s3://openneuro.org/ds002761","raw_key":"dataset_description.json","dep_keys":["CHANGES","README","task-loc_events.json","task-mem_events.json"]},"tagger_meta":{"config_hash":"4a051be509a0e3d0","metadata_hash":"55350f3743b38166","model":"openai/gpt-5.2","tagged_at":"2026-01-20T10:11:14.719598+00:00"},"tags":{"pathology":["Healthy"],"modality":["Visual"],"type":["Memory"],"confidence":{"pathology":0.6,"modality":0.8,"type":0.7},"reasoning":{"few_shot_analysis":"Most similar few-shot convention is the digit span dataset (Healthy / Auditory / Memory), where the presence of an explicit memory task in the README (“working memory task – digit span task with serial recall”) maps the Type to \"Memory\" rather than to task mechanics. Here, the dataset explicitly describes a “Memory phase”, so by the same convention the construct should be labeled Memory. For Modality, the schizophrenia visual discrimination example shows that when the stimulus is on-screen visual material, Modality is \"Visual\".","metadata_analysis":"Key stimulus/task facts in the README indicate visual stimuli and a memory paradigm: (1) Visual presentation: \"the visual onset of an on-screen event\" and \"on the on-screen image\". (2) Memory paradigm: \"Memory phase triggers\" and \"Localizer phase triggers\". (3) Stimulus content categories: \"the first trigger represents the category (1-8) of the on-screen image\" and \"Trigger values of 7 and 8 represent the 4 positive and 4 negative story-ending stimuli\". No participant diagnosis/clinical recruitment details are provided in the supplied metadata, so pathology must be inferred rather than read directly.","paper_abstract_analysis":"No useful paper information.","evidence_alignment_check":"Pathology: Metadata SAYS nothing about diagnosis/recruitment (no quoted clinical group; only methods/triggers). Few-shot pattern SUGGESTS many cognitive MEG/EEG task datasets without clinical descriptors are typically treated as Healthy. ALIGN/CONFLICT: neither; metadata is silent, so we infer Healthy with low confidence.\nModality: Metadata SAYS visual stimuli (\"visual onset of an on-screen event\"; \"on-screen image\"; \"picture onset\"). Few-shot pattern SUGGESTS \"Visual\" for on-screen visual discrimination/stimuli. ALIGN: yes.\nType: Metadata SAYS explicit memory experiment structure (\"Memory phase triggers\"; plus separate \"Localizer phase\"). Few-shot pattern SUGGESTS labeling by cognitive construct (digit span -> Memory) when memory is the study focus. ALIGN: yes.","decision_summary":"Pathology top-2: (1) Healthy — evidence: metadata lacks any diagnosis/clinical recruitment information; typical convention for non-clinical cognitive task datasets. (2) Unknown — evidence: participants are not described at all, so we cannot verify healthy status. Winner: Healthy, but weakly supported. Confidence 0.6 because this is contextual inference only with no direct participant quote.\nModality top-2: (1) Visual — evidence quotes: \"visual onset of an on-screen event\", \"on-screen image\", \"picture onset\". (2) Other — only as a fallback if modality were unclear. Winner: Visual. Confidence 0.8 due to multiple explicit visual-stimulus quotes.\nType top-2: (1) Memory — evidence quotes: \"Memory phase triggers\" and the existence of a memory vs localizer structure; plus the sequenceness/replay-style analysis note suggests memory-related sequence processing. (2) Affect — evidence: \"positive and 4 negative story-ending stimuli\" could indicate emotional manipulation, but it is embedded within the memory task structure rather than described as the primary aim. Winner: Memory. Confidence 0.7 because we have explicit 'Memory phase' wording but limited additional study-purpose description."}},"nemar_citation_count":1,"computed_title":"memoryreplay","nchans_counts":[{"val":306,"count":249}],"sfreq_counts":[{"val":600.0,"count":249}],"stats_computed_at":"2026-04-21T23:17:03.699593+00:00","total_duration_s":null,"author_year":"Wimmer2020","canonical_name":null}}