{"success":true,"database":"eegdash","data":{"_id":"6953f4249276ef1ee07a340c","dataset_id":"ds005662","associated_paper_doi":null,"authors":["Sophie Smit","Almudena Ramírez-Haro","Manuel Varlet","Denise Moerel","Genevieve L. Quek","Tijl Grootswagers"],"bids_version":"1.0.0","contact_info":["Sophie Smit","Tijl Grootswagers"],"contributing_labs":null,"data_processed":true,"dataset_doi":"doi:10.18112/openneuro.ds005662.v2.0.1","datatypes":["eeg"],"demographics":{"subjects_count":80,"ages":[34,24,35,30,36,22,33,23,32,22,30,27,22,44,26,18,28,35,21,24,20,51,21,37,18,45,29,42,18,20,19,18,19,23,40,26,21,20,39,40,24,37,38,23,35,53,76,25,32,38,19,23,21,30,21,25,53,52,18,19,48,44,37,24,28,33,22,22,34,42,21,40,31,23,26,32,21,20,33,31],"age_min":18,"age_max":76,"age_mean":30.075,"species":null,"sex_distribution":{"f":54,"m":24,"o":2},"handedness_distribution":{"a":3}},"experimental_modalities":null,"external_links":{"source_url":"https://openneuro.org/datasets/ds005662","osf_url":null,"github_url":null,"paper_url":null},"funding":["This research was supported by Australian Research Council grants DP220103047 (MV) and DE230100380 (TG)."],"ingestion_fingerprint":"8d879632bb50d7fbf9eb9e7ab3a183789a690482174906b03b2eca3b057d2654","license":"CC0","n_contributing_labs":null,"name":"A comprehensive EEG dataset for investigating\nvisual touch perception","readme":"Data collection took place at The MARCS Institute for Brain, Behaviour and Development in Sydney, Australia. The study was approved by the Western Sydney University Ethics Committee.\nWe recorded EEG data while participants viewed rapid streams of videos adapted from the Validated Touch-Video Database (Smit & Rich, 2025) depicting touch to a hand. Both the adapted videos used in this project, and original videos and validation data, are available on OSF (https://osf.io/jvkqa/).\nThere were 32 sequences in total with a total of 2880 non-target trials (90 unique videos, each presented 8 times) alongside a variable number of target trials (showing touch to an object). Between trials there was an inter-trial-interval of 200ms. The experimental task lasted approximately 55 minutes including breaks. We also recorded questionnaire responses.\nWhole brain 64-channel EEG data were recorded using an Active Two Biosemi system (Biosemi, Inc.) at 2048Hz and 10-20 standard caps. Stimuli were presented using Python and PsychoPy software version 2023.3.1.","recording_modality":["eeg"],"senior_author":"Tijl Grootswagers","sessions":[],"size_bytes":115754787329,"source":"openneuro","study_design":null,"study_domain":null,"tasks":["video"],"timestamps":{"digested_at":"2026-04-22T12:28:40.256134+00:00","dataset_created_at":"2024-11-27T04:56:45.541Z","dataset_modified_at":"2026-02-06T01:59:03.000Z"},"total_files":80,"storage":{"backend":"s3","base":"s3://openneuro.org/ds005662","raw_key":"dataset_description.json","dep_keys":["CHANGES","README.txt","participants.json","participants.tsv","task-video_eeg.json","task-video_events.json"]},"tagger_meta":{"config_hash":"4a051be509a0e3d0","metadata_hash":"db9dfa170ebb3541","model":"openai/gpt-5.2","tagged_at":"2026-01-20T18:34:05.536148+00:00"},"tags":{"pathology":["Healthy"],"modality":["Visual"],"type":["Perception"],"confidence":{"pathology":0.6,"modality":0.8,"type":0.7},"reasoning":{"few_shot_analysis":"Closest few-shot by stimulus/task structure is the Cross-modal Oddball Task example: it has frequent standard trials and rarer oddball conditions/targets and is labeled with an attention/cognitive-control purpose in that clinical context. This guides considering an Attention type due to the explicit 'target trials' among many non-targets. Closest few-shot by stimulus modality and perceptual judgment is the schizophrenia visual discrimination example (moving dots) labeled Visual + Perception, which guides mapping visually presented stimuli about sensory content to the Visual modality and a Perception type when the task centers on processing stimulus properties.","metadata_analysis":"Population: no clinical recruitment criteria are mentioned; only the study site and ethics are described (suggesting a non-clinical cohort). Key lines: (1) \"We recorded EEG data while participants viewed rapid streams of videos ... depicting touch to a hand.\" (2) \"There were 32 sequences in total with a total of 2880 non-target trials ... alongside a variable number of target trials (showing touch to an object).\" Stimulus/input channel is clearly visual video: \"Stimuli were presented using Python and PsychoPy\" and \"participants viewed rapid streams of videos\". The paradigm includes target detection embedded in rapid visual streams: \"non-target trials\" vs \"target trials\".","paper_abstract_analysis":"No useful paper information.","evidence_alignment_check":"Pathology — Metadata says: no disorder/diagnosis is stated (e.g., only \"Data collection took place ...\" and task/stimulus details). Few-shot pattern suggests: when no clinical group is described, label as Healthy. ALIGN.\nModality — Metadata says: \"participants viewed rapid streams of videos\" (visual). Few-shot pattern suggests: video/dots/screen-based tasks map to Visual modality (as in the visual discrimination example). ALIGN.\nType — Metadata says: stimulus content is observation of touch videos (\"videos ... depicting touch to a hand\"), but also includes a target/non-target structure (\"2880 non-target trials ... alongside ... target trials\"). Few-shot pattern suggests: oddball/target detection structures often map to Attention (as in the oddball example), while stimulus-focused discrimination/viewing maps to Perception (as in the visual discrimination example). PARTIAL ALIGNMENT/AMBIGUITY; no explicit statement of primary cognitive construct, so choose the label most directly supported by described stimulus goal (perceptual processing of viewed touch videos), with Attention as runner-up due to targets.","decision_summary":"Top-2 Pathology candidates: (1) Healthy — supported by lack of any stated diagnosis/recruitment of a clinical population (no disorder terms anywhere in README); matches few-shot convention that unspecified cohorts are labeled Healthy. (2) Unknown — possible because participants are not explicitly called \"healthy\". Final: Healthy.\nTop-2 Modality candidates: (1) Visual — supported by \"participants viewed rapid streams of videos\" and PsychoPy stimulus presentation. (2) Multisensory — could be hypothesized because videos depict touch, but there is no tactile stimulation described. Final: Visual.\nTop-2 Type candidates: (1) Perception — supported by the core activity \"viewed ... videos ... depicting touch to a hand\" (sensory/perceptual processing of visual stimuli). (2) Attention — supported by target detection structure: \"2880 non-target trials ... alongside ... target trials\" and rapid streams. Final: Perception, because the only explicitly described content focus is processing visual touch videos, while targets appear as task mechanics without an explicit attention/cognitive-control aim.\nConfidence notes: Pathology confidence is limited because \"healthy\" is not explicitly stated; Type confidence is moderate due to Perception vs Attention ambiguity."}},"computed_title":"A comprehensive EEG dataset for investigating\nvisual touch perception","nchans_counts":[{"val":65,"count":80}],"sfreq_counts":[{"val":2048.0,"count":80}],"stats_computed_at":"2026-04-22T23:16:00.310753+00:00","source_url":"https://openneuro.org/datasets/ds005662","total_duration_s":null,"author_year":"Smit2024","canonical_name":null}}