[Eeglablist] ICA components removal with pop_subcomp

[Chiara Terzo]

Dear all,

I am writing to ask you for help with the pop_subcomp function for
components removal in eeglab. Whenever I remove components (mainly blinks
or channel noise) and I check the outcome with "plot single trials", the
pre vs. post ICA eeg, the post ICA eeg has more amplitude than the pre ICA
eeg.
I thought it was a visualisation problem so I plotted the eeg pre vs. post
with matlab plot function but it is the same. I wonder whether anyone has
suggestions or had the same issue. I thought it was the removal of low
frequencies vs high frequencies with subtraction that created the problem,
but in that case this amplitude issue would appear only in those parts
where the component is removed (for example a blink). In my case the entire
eeg recording seems affected.

Thank you for your help,
Chiara