Reference and EOG electrodes

Post Reply
Gianluca
Posts: 2
Joined: Thu Apr 07, 2016 7:49 pm

Reference and EOG electrodes

Post by Gianluca » Thu Apr 07, 2016 8:05 pm

Hi everyone,
I am sorry for the naive question. I tried to use search but can't find out a solution.

The problem: apparently I can't get EEGLAB to recognize that EXG1 and EXG2 are Eye electrodes. And I am also having some troubles with reference (exg3 exg4).

I have a 32 channels system. During recordings I had 2 vertical electrodes on the eyes (top/bottom EXG1 EXG2), and 2 mastoid (EXG3 EXG4).

This is what I am doing in EEGLAB:

File -> Import data -> Use eeglab functions and plugin -> From Biosemi BDF file(biosig toolbox).
Channels list (default all) -> I leave this empty to get all channels

Reference han(s) indices -> 35 36
Save file

Resample the data

Edit -> Channel location -> (here I tried both BESA file for 4-shell dipfit spherical model and "use spherical file with eye channels). Still, can't see where eye channels are anyway.

Once I do the Edit Channel location, I can see EXG1 to 4, but they have no coordinates. also, if I click ok accepting the coordinate system, I loose the reference.
It says reference -> Unknown
Even if I plot the channels (plot 2-D), I can't see any of the EXG

I tried: Set Reference 1->32
Reference (e.g. CZ) -> EXG5 EXG6. The first time I tried the system relabled these two as reference. Now it is not working anymore. Once I do the Channel location edit, I loose all the reference.

Most Importantly, it does not recognize the ocular electrodes. I tried to do a fast ICA, but once I remove the components by maps, it doesn't change anything. I think it is because the system doesn't know that EXG1 and EXG2 are EOG.

Again, sorry for the naive question. This is my very first time trying to analyze some EEG data.

Thank you in advance for any attempt of help.

Gianluca
Posts: 2
Joined: Thu Apr 07, 2016 7:49 pm

Re: Reference and EOG electrodes

Post by Gianluca » Thu Apr 21, 2016 5:55 pm

Just to let you know, I've better understood pretty much everything and solved the problem. In particular, there was a bug in the eeglab version I was using...

Post Reply