Skip to content

Picks 'data' documentation and code different definitions #13657

Description

@behinger

Hi!
In the glossary
https://mne.tools/stable/documentation/glossary.html#term-data-channels
and
https://mne.tools/stable/documentation/glossary.html#term-non-data-channels

some channels are defined as non-data, e.g. eyegaze and pupil. but in the actual code, they are indeed data-channels:

e.g.

eyetrack=True,

I don't understand the documentation of MNE, in the glossary.rst there is some fancy action that links somewhere else that I couldnt find.

Anyway, with my eye-tracking combined EEG data, ICA.fit(data,picks='data') fails because it includes eyegaze and pupil, but it is my understanding it shouldnt.

Who is right, docs or code (I think docs but 🤷)
Cheers, Bene

Activity

  1. larsoner commented on Feb 13, 2026

    @larsoner
    Member

    Yeah I think they shouldn't be included in data based on our definition, this seems like a bug. We should open a PR to change it and see how much stuff breaks 🙈

  2. larsoner commented on Feb 13, 2026

    @larsoner
    Member

    @behinger let me know if you'd be up for trying!

  3. behinger commented on Feb 18, 2026

    @behinger
    ContributorAuthor

    running it locally now, will push a PR

  4. added this to the 1.12 milestone on Feb 24, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

      Milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions