[Eeglablist] Alternative reference methods?

[Schomaker, J.]

Dear all,

I recently received data from an experiment recorded with a 10-20 system from a clinical population. The experiment was included in a clinical procedure, and the usual EEG recording set-up was used. Unfortunately no specific reference electrode was used in this procedure; instead electrode Fz was used as a reference.
This distorts the EEG signal in an unwanted fashion. It's especially problematic since I am interested in Fz and it's neighbouring electrodes. The dataset contains 23 other electrodes. Plus signal from SO1 and SO2 (supraorbital electrodes), ECG and EMG. Plus 4 channels that were not plugged (containing noise).
Could anyone please recommend me an alternative re-referencing method. I have considered referencing to the average of all electrodes, but fear that the amount of electrodes is too small. Another possible (not optimal of course) option would be to use the SO1 and SO2 after extensive cleaning of eyeblink/eye movement components using ICA and epoch removal. But I am not sure whether this unconventional referencing technique would lead to other problems.

Any tips or suggestions are welcome, because I dearly like to use this dataset. Thanks a lot!

Kind Regards,
Judith Schomaker

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20120511/e94e2f52/attachment.html>