Skip to main content

About dataset

The Behaverse dataset Schema provides a comprehensive, standardized metadata format for describing cognitive science and neuroscience datasets. It enables rich documentation of experimental data including participant demographics, measurement techniques, cognitive tasks, data access information, and ethical considerations.

Motivation​

High-quality metadata is essential for dataset discovery, evaluation, and reuse in cognitive science research. Researchers need to understand what data was collected, from whom, using what methods, and under what conditions.

However, metadata practices vary widely across labs and repositories, making it difficult to find relevant datasets or assess their suitability for secondary analyses.

The dataset schema aims to facilitate discovery of relevant datasets by providing a structured, comprehensive format that captures all essential aspects of cognitive science datasets while maintaining compatibility with established metadata standards.

Features​

1. Comprehensive Coverage​

The schema includes 44 properties organized into logical field groups covering:

  • Core identification and licensing
  • Participant demographics and population characteristics
  • Measurement modalities (neuroimaging, behavioral, physiological)
  • Cognitive tasks and experimental paradigms
  • Study design and methodology
  • Data access and file formats
  • Ethics

This comprehensive coverage enables detailed dataset documentation without requiring multiple disparate metadata systems.

2. Standards Compatibility with Consistent Naming​

The schema provides consistent snake_case property naming conventions while maintaining compatibility with existing standards including Schema.org, DataCite, BIDS, DDI, and HuggingFace. Properties such as description, license, and creator map to their equivalents in these standards, while using developer-friendly naming that aligns with common programming conventions.

3. Flexible Nested Structures​

The schema supports both simple and complex metadata through nested object structures. Properties like measurement_technique, activity, and citation can contain rich structured information while simpler datasets can use minimal required fields.

Example nested structure:

{
  "measurement_technique": [
    {
      "type": "neuroimaging",
      "technique": "fMRI",
      "field_strength": 3.0,
      "tr": 2000,
      "manufacturer": "Siemens Prisma"
    }
  ]
}

4. Population and Sample Documentation​

Detailed fields for documenting participant characteristics including:

5. Multi-Modal Measurement Support​

Structured documentation of diverse measurement techniques through the measurement_technique property, with specialized fields for:

  • Neuroimaging: fMRI, EEG, MEG (field strength, timing parameters, manufacturer)
  • Behavioral: Response types, timing accuracy, apparatus details
  • Physiological: ECG, GSR, eye tracking (sampling rates, equipment)

6. Activity and Paradigm Description​

The activity property enables detailed documentation of:

  • Cognitive tasks with trial counts, conditions, and constructs measured
  • Questionnaires and assessments
  • Training or intervention protocols
  • Duration, measurements collected, and experimental conditions

7. Citation and Reference Tracking​

Multiple citation objects can document:

  • Published papers describing the dataset
  • Methodological references
  • Related datasets
  • Software and analysis tools used

Relationship to Standards​

The dataset schema builds on and maps to:

  • Schema.org Dataset: Core vocabulary for dataset descriptions
  • DataCite: Citation metadata and DOI registration
  • BIDS: Brain imaging data structure for neuroimaging datasets
  • DDI: Data Documentation Initiative for social science data
  • HuggingFace: Machine learning dataset cards
  • Dublin Core: Basic metadata elements (description, creator, date)

Namespace​

  • Base namespace: https://behaverse.org/schemas/dataset#
  • Context file: https://behaverse.org/schemas/dataset/context.jsonld

Learn more​

  • Overview — the full property reference (auto-generated from the schema), with a page per property.
  • Examples — a worked dataset example.
  • README on GitHub — full schema details.