{"success":true,"database":"eegdash","data":{"_id":"696ffc370e08707cf2f2a02a","dataset_id":"ds007262","associated_paper_doi":null,"authors":["Matthew Barras","Liam Booth"],"bids_version":"1.9.0","contact_info":["Liam Booth"],"contributing_labs":null,"data_processed":false,"dataset_doi":"doi:10.18112/openneuro.ds007262.v1.0.6","datatypes":["eeg"],"demographics":{"subjects_count":18,"ages":[26,28,31,18,22,48,25,23,23,27,21,28,22,31,21,29,20,24,30,31],"age_min":18,"age_max":48,"age_mean":26.4,"species":null,"sex_distribution":{"f":6,"m":14},"handedness_distribution":{"r":17,"l":3}},"experimental_modalities":null,"external_links":{"source_url":"https://openneuro.org/datasets/ds007262","osf_url":null,"github_url":null,"paper_url":null},"funding":[],"ingestion_fingerprint":"a6e7b7d2fc730b1af7b20b92d09c80318e082391d19c387a324949008ea3caaa","license":"CC0","n_contributing_labs":null,"name":"Cognitive Workload 8-level arithmetic","readme":"This dataset was generated from LSL/XDF recordings. Converted to bids with instructions and code [presented here](https://github.com/LMBooth/QT-arithmetic_study/tree/main/conversion_package)\n- Original recordings are stored under sourcedata/xdf/ as .xdf files (non-BIDS).\n- EEG was converted to BrainVision format (.vhdr/.eeg/.vmrk) under each sub-*/eeg/.\n- *_events.tsv was generated from marker streams and then aligned so onset is relative to the EEG start time.\n- Marker streams include task markers (arithmetic-Markers) and acquisition dropout annotations (UoHDataOffsetStream); events include a marker_stream column and marker definitions are in task-arithmetic_events.json.\n- Pupil Labs gaze/pupil data was exported from the XDF pupil_capture stream into sub-*/eeg/ as eyetrack physio files (*_recording-eyetrack_physio.tsv.gz + *_recording-eyetrack_physio.json; PhysioType=eyetrack).\n- ECG is captured on the EEG system; the ECG channel is typed in *_channels.tsv and exported as *_recording-ecg_physio.tsv.gz + *_recording-ecg_physio.json.\n- ML analysis note: participants excluded from the ML analysis remain in participants.tsv with analysis_included=false; no epoch rejection was applied to this raw dataset.\n- Participant IDs match the original XDF filenames; missing IDs correspond to excluded participants.\nParticipants\n- N_recorded: 20\n- N_released: 18\n- Exclusions: 2 participants excluded due to multi-modal acquisition failures (sub-002, sub-017).\n- Demographics in participants.tsv: age (years), sex, handedness.\n- Excluded IDs remain in participants.tsv with analysis_included=false.\nHardware and data collection\n- Combined EEG+ECG mobile EEG system (Bateson and Asghar, 2021; Clewett et al., 2016) and Pupil Labs Pupil Core, synchronized via Lab Streaming Layer (LSL).\n- EEG: 19-channel 10-20 montage (Fp1, Fp2, F7, F3, Fz, F4, F8, T3, C3, Cz, C4, T4, T5, P3, Pz, P4, T6, O1, O2), Ag/AgCl electrodes with linked-ear reference, 250 Hz; impedances checked and Neurgel EEG gel applied.\n- ECG: 3-lead on the same system; positive lead right shoulder/clavicle, negative lead left shoulder/clavicle, feedback lead lower left torso.\n- Pupillometry: Pupil Labs Pupil Core eye tracking with infrared illuminators; LSL relay with asynchronous sampling (timestamps per sample).\nProtocol summary\n- Arithmetic task difficulty was defined using Q-value ranges and randomized order across trials.\n- Task events encode difficulty in `trial_type` and `difficulty_range` (e.g.,baseline, 0.6-1.5, 1.5-2.4, ..., 6.0-6.9).\n- Baseline for 60 seconds and then 70 questions, 10 at each difficulty level presented for 6 seconds each.\nTask: arithmetic\nRelease notes\n- Recorded 20 participants; released 18.\n- Reason: multi-modal acquisition QC failure.\n- Participant IDs match original XDF filenames; missing IDs indicate excluded participants.\nReferences\n----------\nAppelhoff, S., Sanderson, M., Brooks, T., Vliet, M., Quentin, R., Holdgraf, C., Chaumon, M., Mikulan, E., Tavabi, K., Höchenberger, R., Welke, D., Brunner, C., Rockhill, A., Larson, E., Gramfort, A. and Jas, M. (2019). MNE-BIDS: Organizing electrophysiological data into the BIDS format and facilitating their analysis. Journal of Open Source Software 4: (1896).https://doi.org/10.21105/joss.01896\nPernet, C. R., Appelhoff, S., Gorgolewski, K. J., Flandin, G., Phillips, C., Delorme, A., Oostenveld, R. (2019). EEG-BIDS, an extension to the brain imaging data structure for electroencephalography. Scientific Data, 6, 103.https://doi.org/10.1038/s41597-019-0104-8\nClewett CJ, Langley P, Bateson AD et al (2016) Non-invasive, home-based electroencephalography hypoglycaemia warning system for personal monitoring using skin surface electrodes: a single-case feasibility study. Healthc Technol Lett 3:2-5. https://doi.org/10.1049/htl.2015.0037\nBateson AD, Asghar AUR (2021) Development and evaluation of a smartphone-based electroencephalography (EEG) system. IEEE Access. https://doi.org/10.1109/ACCESS.2021.3079992","recording_modality":["eeg"],"senior_author":"Liam Booth","sessions":[],"size_bytes":397274375,"source":"openneuro","storage":{"backend":"s3","base":"s3://openneuro.org/ds007262","raw_key":"dataset_description.json","dep_keys":["CHANGES","README.md","datacite.yml","participants.json","participants.tsv","task-arithmetic_eeg.json","task-arithmetic_events.json"]},"study_design":null,"study_domain":null,"tasks":["arithmetic"],"timestamps":{"digested_at":"2026-04-22T12:30:10.312859+00:00","dataset_created_at":"2026-01-18T17:36:09.403Z","dataset_modified_at":"2026-02-24T12:21:29.000Z"},"total_files":18,"computed_title":"Cognitive Workload 8-level arithmetic","nchans_counts":[{"val":24,"count":18}],"sfreq_counts":[{"val":250.0,"count":18}],"stats_computed_at":"2026-04-22T23:16:00.312427+00:00","total_duration_s":16500.312,"tagger_meta":{"config_hash":"3557b68bca409f28","metadata_hash":"2dbe5b2b50bfe409","model":"openai/gpt-5.2","tagged_at":"2026-04-07T09:32:40.872789+00:00"},"tags":{"pathology":["Healthy"],"modality":["Unknown"],"type":["Attention"],"confidence":{"pathology":0.8,"modality":0.5,"type":0.7},"reasoning":{"few_shot_analysis":"Most similar few-shot example by research aim is the digit span dataset (Healthy / Auditory / Memory), which explicitly targets cognitive load/working memory via graded difficulty (\"sequences of 5, 9, or 13 digits\"). This guides the convention that graded task difficulty to elicit workload is typically labeled as a higher-level cognitive construct (Memory or Attention) rather than Perception/Motor. Unlike digit span, this dataset’s arithmetic stimuli modality is not explicitly stated, so we cannot directly inherit the Auditory/Visual mapping from the example.","metadata_analysis":"Key facts from provided metadata:\n1) Population appears normative (no diagnosis/clinical recruitment): \"Participants\\n- N_recorded: 20\\n- N_released: 18\" and demographics only: \"Demographics in participants.tsv: age (years), sex, handedness.\"\n2) Task purpose is cognitive workload manipulation via arithmetic difficulty: title \"Cognitive Workload 8-level arithmetic\" and protocol: \"Baseline for 60 seconds and then 70 questions, 10 at each difficulty level presented for 6 seconds each.\" Also: \"Arithmetic task difficulty was defined using Q-value ranges and randomized order across trials.\"","paper_abstract_analysis":"No useful paper information.","evidence_alignment_check":"Pathology:\n- Metadata says: no disorder mentioned; only demographics and exclusions for acquisition failures (e.g., \"Exclusions: 2 participants excluded due to multi-modal acquisition failures\").\n- Few-shot pattern suggests: datasets without clinical recruitment are labeled Healthy.\n- Alignment: ALIGN.\n\nModality:\n- Metadata says: arithmetic questions with time limits (\"70 questions... presented for 6 seconds each\") but does not specify whether questions were shown visually or presented auditorily.\n- Few-shot pattern suggests: use stimulus channel (e.g., digit span explicitly \"presented auditorily\" -> Auditory), but here the stimulus channel is not stated.\n- Alignment: PARTIAL / INSUFFICIENT; choose Unknown because explicit modality facts are missing.\n\nType:\n- Metadata says: explicit aim is workload manipulation (\"Cognitive Workload\"; \"difficulty ... randomized\"; multi-level difficulty ranges).\n- Few-shot pattern suggests: graded cognitive-load paradigms map to Memory or Attention depending on whether working-memory storage/recall is central (digit span -> Memory).\n- Alignment: ALIGN on being a cognitive-load construct; exact label (Memory vs Attention) requires inference. Arithmetic workload most strongly reflects sustained mental effort/attentional demand rather than explicit memorization/recall, so Attention is selected.","decision_summary":"Top-2 candidates and selection:\n\nPathology:\n- Candidate 1: Healthy — Evidence: no clinical terms anywhere; only demographic description (\"age (years), sex, handedness\") and non-clinical exclusions (\"excluded due to multi-modal acquisition failures\").\n- Candidate 2: Unknown — would apply if recruitment/health status were unstated and ambiguity remained.\nHead-to-head: Healthy wins because metadata provides no indication of any patient group and matches few-shot convention for normative cohorts.\n\nModality:\n- Candidate 1: Unknown — Evidence: task described as arithmetic questions with durations, but no statement like \"visual\"/\"auditory\" presentation.\n- Candidate 2: Visual — Weak inference from typical arithmetic-question presentation and the phrasing \"questions\" with fixed display time (\"presented for 6 seconds each\"), but not explicit.\nHead-to-head: Unknown wins because stimulus channel is not explicitly specified in the metadata, and the catalog rule is to base modality on the stimulus.\n\nType:\n- Candidate 1: Attention — Evidence: dataset framed as \"Cognitive Workload\" with multiple difficulty levels (\"8-level arithmetic\"; \"difficulty ... randomized\"; \"70 questions... 10 at each difficulty level\").\n- Candidate 2: Memory — Plausible because arithmetic workload often taxes working memory, and few-shot digit span maps cognitive load to Memory when explicit memorization/recall is central.\nHead-to-head: Attention wins because the metadata emphasizes workload/difficulty manipulation without explicit encoding/maintenance/recall structure (unlike digit span). Confidence reflects that Memory remains a close runner-up."}},"canonical_name":null,"name_confidence":0.87,"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":"Barras2026_Cognitive"}}