GitHub - OpenNeuroDatasets/ds008092: OpenNeuro dataset - Food-image categorization with consumer EEG (vegetarian vs. meat)

Branches Tags

Name		Name	Last commit message	Last commit date
Latest commit History 13 Commits
.datalad		.datalad
code		code
derivatives/eeglab		derivatives/eeglab
sub-01/eeg		sub-01/eeg
sub-02/eeg		sub-02/eeg
sub-03/eeg		sub-03/eeg
sub-04/eeg		sub-04/eeg
sub-05/eeg		sub-05/eeg
sub-06/eeg		sub-06/eeg
sub-07/eeg		sub-07/eeg
sub-08/eeg		sub-08/eeg
sub-09/eeg		sub-09/eeg
sub-10/eeg		sub-10/eeg
sub-11/eeg		sub-11/eeg
sub-12/eeg		sub-12/eeg
sub-13/eeg		sub-13/eeg
sub-14/eeg		sub-14/eeg
sub-15/eeg		sub-15/eeg
sub-16/eeg		sub-16/eeg
sub-17/eeg		sub-17/eeg
sub-18/eeg		sub-18/eeg
sub-19/eeg		sub-19/eeg
sub-20/eeg		sub-20/eeg
sub-21/eeg		sub-21/eeg
sub-22/eeg		sub-22/eeg
sub-23/eeg		sub-23/eeg
sub-24/eeg		sub-24/eeg
sub-25/eeg		sub-25/eeg
sub-26/eeg		sub-26/eeg
sub-27/eeg		sub-27/eeg
sub-28/eeg		sub-28/eeg
sub-29/eeg		sub-29/eeg
sub-30/eeg		sub-30/eeg
sub-31/eeg		sub-31/eeg
sub-32/eeg		sub-32/eeg
sub-33/eeg		sub-33/eeg
sub-34/eeg		sub-34/eeg
sub-35/eeg		sub-35/eeg
sub-36/eeg		sub-36/eeg
sub-37/eeg		sub-37/eeg
sub-38/eeg		sub-38/eeg
sub-39/eeg		sub-39/eeg
sub-40/eeg		sub-40/eeg
sub-41/eeg		sub-41/eeg
sub-42/eeg		sub-42/eeg
sub-43/eeg		sub-43/eeg
sub-44/eeg		sub-44/eeg
sub-45/eeg		sub-45/eeg
sub-46/eeg		sub-46/eeg
sub-47/eeg		sub-47/eeg
sub-48/eeg		sub-48/eeg
sub-49/eeg		sub-49/eeg
sub-50/eeg		sub-50/eeg
sub-51/eeg		sub-51/eeg
sub-52/eeg		sub-52/eeg
sub-53/eeg		sub-53/eeg
sub-54/eeg		sub-54/eeg
sub-55/eeg		sub-55/eeg
sub-56/eeg		sub-56/eeg
sub-57/eeg		sub-57/eeg
sub-58/eeg		sub-58/eeg
sub-59/eeg		sub-59/eeg
sub-60/eeg		sub-60/eeg
sub-61/eeg		sub-61/eeg
sub-62/eeg		sub-62/eeg
sub-63/eeg		sub-63/eeg
sub-64/eeg		sub-64/eeg
sub-65/eeg		sub-65/eeg
sub-66/eeg		sub-66/eeg
sub-67/eeg		sub-67/eeg
sub-68/eeg		sub-68/eeg
sub-69/eeg		sub-69/eeg
sub-70/eeg		sub-70/eeg
sub-71/eeg		sub-71/eeg
sub-72/eeg		sub-72/eeg
sub-73/eeg		sub-73/eeg
sub-74/eeg		sub-74/eeg
sub-75/eeg		sub-75/eeg
sub-76/eeg		sub-76/eeg
sub-77/eeg		sub-77/eeg
.gitattributes		.gitattributes
CHANGES		CHANGES
README		README
datacite.yml		datacite.yml
dataset_description.json		dataset_description.json
participants.json		participants.json
participants.tsv		participants.tsv
task-foodcat_channels.json		task-foodcat_channels.json
task-foodcat_events.json		task-foodcat_events.json

Repository files navigation

# Food-image categorization with consumer EEG (vegetarian vs. meat)

## Overview
14-channel EEG (EMOTIV EPOC X / EPOC+, scalp positions AF3 F7 F3 FC5 T7 P7 O1
O2 P8 T8 FC6 F4 F8 AF4) recorded while participants categorized prepared-food
images as meat or vegetarian. Each recording also includes a short eyes-open /
eyes-closed resting baseline that precedes the task.

These data were collected within longer sessions that also recorded other
measures (reported elsewhere). Only the categorization paradigm and its resting
baseline are released here.

## Task
The same set of 37 prepared-food images was presented twice (up to ~74 trials).
On each trial a participant pressed M (meat) or X (vegetarian). Response times
were recorded.

In the events files:
- `stim_category` is the displayed image's true category: m (meat) or v (vegetarian).
- `response_key` is the physical key pressed: M or X.
- `response` is that key mapped back to category (m/v) so it lines up with `stim_category`.
- `accuracy` is 1 when `response` matches `stim_category`.

## Files
- `participants.tsv` / `participants.json`: one row per participant, with
  demographics, baseline questionnaire, derived dietary indicators, and per-subject
  task summaries. `.tsv` is tab-separated and opens like a CSV.
- `sub-NN/eeg/`: the trimmed recording (`_eeg.edf`), `_events.tsv`,
  `_channels.tsv` (with per-channel contact-quality), and `_eeg.json` sidecar.

## Data quality
Data are released minimally processed: no filtering, no artifact rejection, no
ICA. Per-channel EMOTIV contact-quality values are provided in each
`channels.tsv` so re-users can apply their own quality criteria. The resting
baseline supports a per-participant eyes-closed alpha-reactivity check.

## Notes for re-use
- EEG is unfiltered (sidecars report `SoftwareFilters: n/a`).
- All recordings are sampled at 128 Hz.
- Event labels are embedded in each EDF as EDF+ annotations (rest_eyesopen, rest_eyesclosed,
  categorize/meat, categorize/veg), so any reader displays them directly. The events.tsv holds
  the full trial-level record (response, response time, accuracy).
- Resting segments were advanced by participant button-press; a fixed 30 s window
  inside each instructed eyes-open and eyes-closed segment is released.
- Recording timestamps have been removed; recordings carry a neutral fixed date.