I have a raw file for review. I'm using projectors to correct artifacts (e.g. blinks). When I'm satisfied at first (say for condition 1), I choose a bunch of SSPs to remove and import my data time locked to events of condition 1. The channel file (containing SSP projectors) gets copied to a new folder. (correct?).
After a while, I go back to my raw data to process condition 2 and choose a different SSP (or even a different method, say ICA or even no projector at all) to correct my artifacts. Import these new trials again. What happens to the channel file? does it include previous and/or new projectors?
My understanding is that only the first set of projectors is stored in the channel.mat. Is that not a problem?
I admit this is a borderline use case, and I suppose the correct thing to do would be to rename the formerly imported folder before importing again, so as to create a new channel file in a different folder (I just tested that, and I get the expected result). But is this not a potential source of error (a difficult one to catch)?
Or perhaps I don't understand something in the way projectors are stored and/or applied at import...
Otherwise, I suppose it would be important for provenance tracking and consistency in the channel.mat file to at least issue a warning or error message, and at best import under a different study folder.
This is obviously not just a thought experiment (though I simplified the situation here for clarity), but an actual use case I've been facing recently.
Thanks a lot for any comment.