[Eeglablist] Alpha contaminating EEG

[marco mcsweeney]

Dear all

I am fairly new to EEG and I am conducting my first EEG experiment so my
apologies if my query has an obvious solution that I cannot see. I searched
the mailing list and have looked online for a solution but am still unsure
about the best way to deal with my issue.

I have collected data from a number of participants who show very large
alpha activity (not just occipital electrodes but across most channels)
which in my case is problematic as I am interested in both slow negative
cortical potentials like the CNV and more posterior potentials like the
N170.

I have tried to reduce the likelihood of this occurring in the first place
by using jittered inter-trial intervals and I talk to participants between
blocks but for some participants this has not helped so much. For the
participants who show this alpha activity it is not possible to remove
these sections of data as I would have to remove too many trials.
Alternatively, running ICA in some participants clearly identifies alpha
activity as a component which I could remove.

I guess my issue is whether I should remove alpha activity at all or leave
it in and hope that this averages out across participants.

Any help on this would be greatly appreciated!

Best wishes

Marco
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20181112/9687e5f5/attachment.html>