[Eeglablist] re-referencing when online reference is a scalp channel

[Jason Palmer]

Hi Denise,

If your dataset has the 5Z channel in it when loaded (and you have 65 channels) then it should be all zeroes, since the reference channel is zero relative to itself. In that case you cam reref to average and keep the 5Z channel. Usually though, the reference channel won't be included, and you will have 64 channels after loading.

If you have 64 channels, you can add the 5Z channel to the chanlocs assuming you have the location coordinates, and add a corresponding channel of all zeroes, and increase EEG.nbchan to 65, then do average reference.

Often people will do average rereference with 64 channels without adding the zero reference. You lose a degree of freedom doing average reference.

If you are going to do ICA, then in principle it is not necessary to do average reference (and then pca reduce by 1d) since the ICA unmixing matrix can easily contain a row of all 1s equivalent to average reference, or a more refined source, corresponding to common mode line noise for example.

Best,
Jason
________________________________
From: Jason Palmer <japalmer29 at gmail.com>
Sent: Wednesday, March 26, 2025 12:58:00 AM
To: Denise Dal Lago <denise.dallago at unibo.it>; Denise Dal Lago via eeglablist <eeglablist at sccn.ucsd.edu>
Subject: Re: [Eeglablist] re-referencing when online reference is a scalp channel

Hi Denise,

If your dataset has the 5Z channel in it when loaded (and you have 65 channels) then it should be all zeroes, since the reference channel is zero relative to itself. In that case you cam reref to average and keep the 5Z channel. Usually though, the reference channel won't be included, and you will have 64 channels after loading.

If you have 64 channels, you can add the 5Z channel to the chanlocs assuming you have the location coordinates, and add a corresponding channel of all zeroes, and increase EEG.nbchan to 65, then do average reference.

Often people will do average rereference
________________________________
From: eeglablist <eeglablist-bounces at sccn.ucsd.edu> on behalf of Denise Dal Lago via eeglablist <eeglablist at sccn.ucsd.edu>
Sent: Tuesday, March 25, 2025 11:06:45 AM
To: Denise Dal Lago via eeglablist <eeglablist at sccn.ucsd.edu>
Subject: [Eeglablist] re-referencing when online reference is a scalp channel

Hi all,
I recorded EEG data using a 65-channel ANT Neuro cap with an online reference at channel 5Z (ANT Neuro labeling). I am now pre-processing the data in MATLAB with EEGLAB and need to re-reference to the common average. If I simply apply the re-reference, I will still have 65 channels with 5Z as an active channel. Should I remove the 5Z channel after re-referencing, or is it better to retain it? What would be the best practice in this case?
I am more familiar with the Biosemi recording system, where I didn’t encounter this issue.
Many thanks,
Denise



Denise Dal Lago, PhD
Research Fellow
Department of Psychology, University of Bologna
Viale Berti Pichat, 5 -40127 Bologna, Italy
_______________________________________________
To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu or visit https://sccn.ucsd.edu/mailman/listinfo/eeglablist  .