Conflict between EEGLAB and LIMO on session in design matrix

[arnodelorme]

When EEGLAB declare that it wants to compare sessions in the STUDY.design, instead of being ignored, it is included in the LIMO design.

[CPernet]

It is not ignored - if you have two sessions or more, it creates sessions, runs the glm for each one, and computes between sessions contrasts - the reason to do this is because the variance between sessions is not the same as if LIMO was doing one design. ++- not a bug, a feature.

[arnodelorme]

No, it creates a categorical variable for the session (it also creates different files as it should).

[CPernet]

Do you mean, you have a session variable? Yes it does not deal with that -- and I do not want that (session variance needs to be accounted for)

[arnodelorme]

Let's change the code to remove the session variable from the STUDY design (temporarily) before computing LIMO files.