Questionnaire data

This is an outline of the requirements for questionnaire data that comes with your neuroimaging data.In the BIDS specification this type of data is referred to as phenotype data. The representation of phenotypic data within the ANC must conform to the BIDS specification.

Format summary

The current BIDS specification proposes phenotpypic data is stored in a top level directory phenotype. Data is stored in tabular files, along with accompanying .json files.

└── phenotype/
    ├── <measurement_tool-name>.tsv
    └── <measurement_tool-name>.json

To pass the validation the <measurement_tool_name>.tsv MUST have a column participant_id which contents MUST correspond to a sub- directory in the dataset and the participant.tsv.

Content of <measurement_tool-name.tsv

participant_id	item_1	item_2	item_3
sub-01	1	2	1
sub-02	3	1	2

Further requirements

The BIDS specification does not currently cover several common use cases:

• storing pre-screening data / storing varying questionnaires across subjects
• storing drop outs
• storing longitudinal data meaningfully

The following table describes how to represent these use cases in the ANC.

Issue	Error cause	Solution
Pre-screening data	The bids-validator errors out if a participant_idof the <measurement-tool>.tsvfile does not correspond to a sub- directory.	Store the phenotypic data for participants that do not have imaging data in a subdirectory (e.g. pheno-extra) in the /phenotype dir and add this to the .bidsignore.
Varying questionaires across subjects / Storing drop-outs	The bids-validator errors out if a participant_idof the <measurement-tool.tsvfile does not correspond to a sub- directory. Here, theoretically each participant_id corresponds to a sub- directory, but not every <measurement_tool>.tsv contains every participant_id	Create the participant_id in every <measurement_tool>.tsv and fill it with n/a.
Storing longitudinal data meaningfully	The current specification does not take repeated phenotypic measures into account	Add the ses- label to the measurement tool .tsv name (e.g., <measurement-tool-ses-1>.tsv, <measurement-tool-ses-2>.tsv).

Example

The following shows an example of the directory structure in case of participant drop-out.

├── sub-01/
│   ├── ses-01/
│   │   ├── anat/
│   │   └── func/
│   └── ses-02/
│       ├── anat/
│       └── func/
├── sub-02/
│   └── ses-01/
│       ├── anat/
│       └── func/
└── phenotype/
    ├── <measurement_tool-ses-01>.tsv
    ├── <measurement_tool-ses-01>.json
    ├── <measurement_tool-ses-02>.tsv
    └── <measurement_tool-ses-02>.json

The tabular file for the second session would look as follows:

participant_id	item_1	item_2	item_3
sub-01	1	2	1
sub-02	n/a	n/a	n/a