[Eeglablist] Combining EEG datasets during preprocessing

[Stefan Đukić]

Dear EEGLAB Community,

I am currently working on the pipeline for preprocessing our 128-channel
EEG data, which comprises several paradigms recorded consecutively. One of
these paradigms is resting-state, including both eyes-open (6 minutes) and
eyes-closed (6 minutes) conditions. I am considering appending these two
conditions together and preprocessing/cleaning these data as a single
dataset.

My rationale for this approach is to incorporate a larger dataset, which is
useful for ICA, as well as to leverage the generally cleaner nature of
eyes-closed data, benefiting ASR. I am, however, concerned about the
potential issue of mixing activity from the two conditions when applying
methods like ASR. On the other hand, could this approach be considered
analogous to analysing any other paradigm with two conditions?

Ultimately, my question is: Is there a line that separates data that may be
preprocessed together from data that should be better preprocessed
separately? And if so, where is it? The only potential issue I can think of
is moving/changing electrode/cap positions or fixing electrode contacts
(e.g., adding gel) between the recording blocks.

I would greatly appreciate any insights or recommendations.
Thank you very much in advance.

Best,
Stefan Dukic
Postdoctoral Researcher at UMC Utrecht, The Netherlands