Behavioral And Phenotype Definitions

Calibench fear-conditioning datasets extend BIDS participant and phenotype metadata rules. Behavioral and participant-level metadata are stored at the dataset level using participants.tsv / participants.json (see the BIDS participants file specification) and, when available, files in phenotype/ (see the BIDS phenotypic and assessment data specification).

Task events and physiological recordings are stored in physio/ as documented in the event and physiology pages.

Participant metadata

BIDS requires participant_id and allows participant descriptors such as age, sex, and handedness. For Calibench datasets, the top-level participants.tsv file MUST contain one row per participant and the following columns:

Table 1: Required participant columns

Column

Description

Example

participant_id

Participant label used in the dataset.

sub-<participant_label>

age

Participant age in years.

23

sex

Self-reported biological sex.

F

handedness

Participant handedness.

left

The matching participants.json sidecar MUST document every column present in participants.tsv. Column names in the TSV and JSON must match exactly.

Recommended levels:

Table 2: Recommended participant levels

Column

Allowed values

sex

M, F, O, n/a

handedness

left, right, ambidextrous, n/a

Phenotype metadata

BIDS stores participant-level questionnaire and assessment data in phenotype/. For Calibench datasets, each phenotype TSV file MUST have a matching JSON sidecar with the same basename.

phenotype/
├── participant_info.tsv
├── participant_info.json
├── PreAcquisitionRatings.tsv
├── PreAcquisitionRatings.json
├── PostAcquisitionRatings.tsv
├── PostAcquisitionRatings.json
├── PostExtinctionRatings.tsv
└── PostExtinctionRatings.json

Every phenotype TSV file MUST use participant_id as the first column. Every remaining TSV column MUST be documented by an exactly matching key in the JSON sidecar. The sidecar SHOULD include MeasurementToolMetadata with a short description and source URL when available.

Calibench phenotype files use these naming patterns:

Table 3: Phenotype files and column patterns

File

Required columns

participant_info.tsv

participant_id, recorded_at, room_temperature, humidity, age, sex, handedness

PreAcquisitionRatings.tsv

participant_id, pre_acq_1 through pre_acq_6

PostAcquisitionRatings.tsv

participant_id, post_acq_1 through post_acq_6

PostExtinctionRatings.tsv

participant_id, post_ext_1 through post_ext_8

bfi60_german.tsv

participant_id, bfi60_1 through bfi60_60

gad7_german.tsv

participant_id, gad7_1 through gad7_7

ius18_german.tsv

participant_id, ius18_1 through ius18_18

phq9_german.tsv

participant_id, phq9_1 through phq9_9

soc12_german.tsv

participant_id, soc12_1 through soc12_12

stai20_german.tsv

participant_id, stai20_1 through stai20_20

Stimulus presentation metadata

The BIDS Events specification defines StimulusPresentation metadata for visual stimuli and eye-tracking. For visual-stimulus or eye-tracking Calibench datasets, stimulus presentation metadata MUST be available in the task sidecar. These fields are stored in the corresponding *_events.json file under StimulusPresentation.

Note

All physical distance and size measurements in StimulusPresentation MUST be reported in millimeters. This applies to fields such as ScreenDistance and ScreenSize.

Table 4: Required stimulus presentation fields

Key

Description

Example

StimulusPresentation.ScreenDistance

Distance between participant and screen in millimeters.

500

StimulusPresentation.ScreenOrigin

Origin of the screen coordinate system.

top, left

StimulusPresentation.ScreenRefreshRate

Refresh rate of the screen in Hertz.

56

StimulusPresentation.ScreenResolution

Screen resolution in pixels.

[1920, 1080]

StimulusPresentation.ScreenSize

Physical screen size in millimeters.

[476.0, 268.0]

Example StimulusPresentation object:

{
    "StimulusPresentation": {
        "ScreenDistance": 500,
        "ScreenOrigin": [
            "top",
            "left"
        ],
        "ScreenRefreshRate": 56,
        "ScreenResolution": [
            1920,
            1080
        ],
        "ScreenSize": [
            476.0,
            268.0
        ]
    }
}