{"success":true,"database":"eegdash","data":{"_id":"6953f4249276ef1ee07a342f","dataset_id":"ds006012","associated_paper_doi":null,"authors":["Mathias Sablé-Meyer","Lucas Benjamin","Cassandra Potier Watkins","Chenxi He","Maxence Pajot","Théo Morfoisse","Fosca Al Roumi","Stanislas Dehaene"],"bids_version":"1.6.0","contact_info":["Mathias Sablé-Meyer"],"contributing_labs":null,"data_processed":false,"dataset_doi":"doi:10.18112/openneuro.ds006012.v1.0.1","datatypes":["meg"],"demographics":{"subjects_count":21,"ages":[28,21,20,42,22,21,20,34,25,21,30,23,30,25,27,32,29,25,23,23],"age_min":20,"age_max":42,"age_mean":26.05,"species":null,"sex_distribution":null,"handedness_distribution":{"r":20}},"experimental_modalities":null,"external_links":{"source_url":"https://openneuro.org/datasets/ds006012","osf_url":null,"github_url":null,"paper_url":null},"funding":["ERC-ADG-2015 NeuroSyntax to Stanislas Dehaene","CEA to Stanislas Dehaene","INSERM to Stanislas Dehaene","Collège de France to Stanislas Dehaene","FYSSEN to Mathias Sablé-Meyer"],"ingestion_fingerprint":"9d1ccbf1c091034449b74ce052e2b8a09f108f9e4d3bd304b4f8a15abb6947ba","license":"CC0","n_contributing_labs":null,"name":"A geometric shape regularity effect in the human brain: MEG dataset","readme":"# A geometric shape regularity effect in the human brain: MEG dataset\nAuthors:\n* Mathias Sablé-Meyer*\n* Lucas Benjamin\n* Cassandra Potier Watkins\n* Chenxi He\n* Maxence Pajot\n* Théo Morfoisse\n* Fosca Al Roumi\n* Stanislas Dehaene\n*Corresponding author: [mathias.sable-meyer@ucl.ac.uk](mailto:mathias.sable-meyer@ucl.ac.uk)\n## Abstract\nThe perception and production of regular geometric shapes is a characteristic trait of human cultures since prehistory, whose neural mechanisms are unknown. Behavioral studies suggest that humans are attuned to discrete regularities such as symmetries and parallelism, and rely on their combinations to encode regular geometric shapes in a compressed form. To identify the relevant brain systems and their dynamics, we collected functional MRI and magnetoencephalography data in both adults and six-year-olds during the perception of simple shapes such as hexagons, triangles and quadrilaterals. The results revealed that geometric shapes, relative to other visual categories, induce a hypoactivation of ventral visual areas and an overactivation of the intraparietal and inferior temporal regions also involved in mathematical processing, whose activation is modulated by geometric regularity. While convolutional neural networks captured the early visual activity evoked by geometric shapes, they failed to account for subsequent dorsal parietal and prefrontal signals, which could only be captured by discrete geometric features or by more advanced transformer models of vision. We propose that the perception of abstract geometric regularities engages an additional symbolic mode of visual perception.\n## Notes about this dataset\nWe separately share the fMRI dataset at [https://openneuro.org/datasets/ds006010](https://openneuro.org/datasets/ds006010). Below are some notes about the\nMEG dataset of N=20 participants:\n* The code for the analyses associated to\n  [https://doi.org/10.1101/2024.03.13.584141](https://doi.org/10.1101/2024.03.13.584141)\n  are provided at\n  [https://github.com/mathias-sm/AGeometricShapeRegularityEffectHumanBrain](https://github.com/mathias-sm/AGeometricShapeRegularityEffectHumanBrain).\n  However, these analyses have been performed on pre-processed data _without_\n  this defacing steps. I am not publishing this raw data, but should there be\n  discrepancies or problems coming from the defacing, I have a copy of the following\n  information, which I may ask for permission to share in specific cases:\n    1. The original data\n    2. The seed used for the anonymization procedure\n    3. The shuffling information.\n* Anonymization (including defacing of the `anat` folder) has been performed\n  using the following command:\n  `python -c 'import mne_bids; mne_bids.anonymize_dataset(\"<input>\", \"<output>\", random_state=<number>, daysback=<number>)'`\n  This has shuffled the participant order, changed the dates, defaced the\n  anatomy, and stripped gender information from the dataset.\n* The data was pre-processed with the configuration file provided at\n  [https://github.com/mathias-sm/AGeometricShapeRegularityEffectHumanBrain/blob/main/MEG/POGS_MEG_config.py](https://github.com/mathias-sm/AGeometricShapeRegularityEffectHumanBrain/blob/main/MEG/POGS_MEG_config.py)\n  for `mne-bids-pipeline` with the development version at the time,\n  `bce60a79241731bdd03fccffa6cf315a35b33ab2` on\n  [https://github.com/mne-tools/mne-bids-pipeline/](https://github.com/mne-tools/mne-bids-pipeline/)","recording_modality":["meg"],"senior_author":"Stanislas Dehaene","sessions":["20211206","20211212","20211213","20211231","20220104","20220108","20220121","20220207","20220211","20220214","20220218","20220220","20220221","20220227","20220304"],"size_bytes":76309397809,"source":"openneuro","study_design":null,"study_domain":null,"tasks":["POGS","noise"],"timestamps":{"digested_at":"2026-04-22T12:29:01.573627+00:00","dataset_created_at":"2025-03-13T19:13:44.963Z","dataset_modified_at":"2025-03-14T17:05:14.000Z"},"total_files":193,"storage":{"backend":"s3","base":"s3://openneuro.org/ds006012","raw_key":"dataset_description.json","dep_keys":["CHANGES","README","participants.json","participants.tsv"]},"tagger_meta":{"config_hash":"4a051be509a0e3d0","metadata_hash":"eb6fe7ab993e4476","model":"openai/gpt-5.2","tagged_at":"2026-01-20T18:46:56.093962+00:00"},"tags":{"pathology":["Healthy"],"modality":["Visual"],"type":["Perception"],"confidence":{"pathology":0.75,"modality":0.85,"type":0.85},"reasoning":{"few_shot_analysis":"Closest convention match on task/stimulus is the few-shot example \"Meta-rdk: Preprocessed EEG data\", which is labeled (Modality=Visual, Type=Perception) for a visual discrimination paradigm. The current dataset similarly presents visual stimuli (geometric shapes) and targets perceptual processing/regularity encoding, which by the few-shot conventions maps to Visual + Perception. For pathology conventions, multiple few-shots show that when there is no disorder-based recruitment (e.g., \"EEG: Three armed bandit gambling task\"; \"EEG Motor Movement/Imagery Dataset\"), Pathology is labeled Healthy; this guides labeling here because no clinical diagnosis is described.","metadata_analysis":"Key dataset facts from the provided README:\n1) Visual perceptual paradigm: \"we collected functional MRI and magnetoencephalography data in both adults and six-year-olds during the perception of simple shapes such as hexagons, triangles and quadrilaterals.\" \n2) Focus on geometric regularity perception: \"Behavioral studies suggest that humans are attuned to discrete regularities such as symmetries and parallelism...\" and \"We propose that the perception of abstract geometric regularities engages an additional symbolic mode of visual perception.\"\n3) Population is non-clinical, developmental comparison: \"in both adults and six-year-olds\" with no mention of any diagnosis, patient group, or clinical recruitment.","paper_abstract_analysis":"No separate paper abstract field was provided beyond the abstract embedded in the README; the README abstract directly supports a visual-perception interpretation and contains no clinical-population information.","evidence_alignment_check":"Pathology:\n- Metadata says: \"in both adults and six-year-olds\" with no disorder/diagnosis mentioned.\n- Few-shot pattern suggests: non-clinical recruitment datasets are labeled \"Healthy\".\n- ALIGN: yes (both indicate no pathology-based recruitment).\n\nModality:\n- Metadata says: \"during the perception of simple shapes such as hexagons, triangles and quadrilaterals\" and contrasts \"geometric shapes\" with \"other visual categories\".\n- Few-shot pattern suggests: visual stimulus paradigms map to \"Visual\" modality.\n- ALIGN: yes.\n\nType:\n- Metadata says: \"during the perception of simple shapes\" and discusses \"perception of abstract geometric regularities\".\n- Few-shot pattern suggests: sensory/perceptual tasks (e.g., visual discrimination) map to \"Perception\".\n- ALIGN: yes.","decision_summary":"Pathology (top-2):\n1) Healthy (selected): supported by (a) \"in both adults and six-year-olds\" and (b) absence of any diagnosis/patient terminology in the README abstract; consistent with few-shot convention that normative cohorts are \"Healthy\".\n2) Development (runner-up): plausible because six-year-olds are included, but metadata does not indicate a developmental disorder/clinical recruitment; thus \"Healthy\" is stronger under the provided definition (clinical condition used to recruit).\n\nModality (top-2):\n1) Visual (selected): explicit visual-shape perception: \"perception of simple shapes such as hexagons, triangles and quadrilaterals\" and \"other visual categories\".\n2) Multisensory (runner-up): weak/no support; no auditory/tactile stimuli described.\n\nType (top-2):\n1) Perception (selected): explicitly framed as \"perception\" of shapes/regularities and comparison to other visual categories; aligns with few-shot mapping of visual discrimination/perceptual paradigms to Perception.\n2) Other (runner-up): could be argued as mathematical/symbolic processing, but the primary described construct is perceptual encoding of geometric regularity.\n\nConfidence justification:\n- Pathology: moderate-high because we have explicit non-clinical description but limited participant metadata beyond age group (no subjects.tsv quotes available).\n- Modality: high due to multiple explicit mentions of visual shape perception.\n- Type: high due to repeated explicit use of \"perception\" and clear few-shot analog (visual discrimination → Perception)."}},"computed_title":"A geometric shape regularity effect in the human brain: MEG dataset","nchans_counts":[{"val":336,"count":172},{"val":333,"count":1}],"sfreq_counts":[{"val":1000.0,"count":173}],"stats_computed_at":"2026-04-22T23:16:00.311212+00:00","total_duration_s":56111.827000000005,"author_year":"SableMeyer2025","canonical_name":null}}