[Eeglablist] EEG source localization in group level

[Iman Mohammad-Rezazadeh]

Hi John,
Using EEGLAB, the same head model is used  for all of the individual subjects . So, I would assume you may have the same smearing problem in source localization as well.  In addition,  it is suggested that the EEG head montage should be transformed to standard montage for all of the subjects ( after cleaning and channel rejection ) to provide the same montage for all of the subjects.

I like Scott explanation about my question. However, I am still thinking that it would be  probably better if we first limit our time of interest ( for example to an ERP component like ‘P1’ ) and then apply the source localization process on each individual subject. Using this method, we may get more relevant components which belong to same mental/cognitive/sensory/… process rather than clustering all of the components which are close to each other but not necessarily involved in the same brain process. Any thoughts?

Best
Iman


From: RICHARDS, JOHN [mailto:RICHARDS at mailbox.sc.edu]
Sent: Saturday, January 23, 2016 7:03 PM
To: Iman Mohammad-Rezazadeh <irezazadeh at UCDAVIS.EDU>; EEGLAB List <eeglablist at sccn.ucsd.edu>
Subject: Re: [Eeglablist] EEG source localization in group level

An additional concern….

The EEG on the scalp is a result of a cortical source current and the composition of the individual participant head (skull, gm, wm, …).  So when you average on the scalp, you are averaging data across participants that was generated at a location in the participant head, but filtered by that participant’s head characteristics.  The same “location” across individuals (e.g., an anatomical location) would come out with a different scalp distribution across individuals because of differences in the head composition across individuals.  I suspect this would lead to a smearing of the group average ERP and likely a different location when the source is analyzed on the grand average than when it is averaged on the individual participants.  Keeping the participant data intact (doing source analysis on the participant rather than the grand average) would remove this source of localization error.

Ditto clustering of individual participants electrodes; electrode placement differs across individuals due to head size, different net placements, etc; and the relation of a nominally designated electrode (#25; P3; P7) to the underlying cortical EEG generators differs across individuals given individual differences in head model composition across individuals.

But is is common to assume that a single electrode nominally located across individuals is capturing signal from a common brain area, so we do grand average ERPs across individuals on a single electrode, cluster ICA components across individuals based on electrode numbers, etc.

John

***********************************************
John E. Richards Carolina Distinguished Professor
Department of Psychology
University of South Carolina
Columbia, SC  29208
Dept Phone: 803 777 2079
Fax: 803 777 9558
Email: richards-john at sc.edu<mailto:richards-john at sc.edu>
HTTP: jerlab.psych.sc.edu
***********************************************


From: <eeglablist-bounces at sccn.ucsd.edu<mailto:eeglablist-bounces at sccn.ucsd.edu>> on behalf of Iman Mohammad-Rezazadeh <irezazadeh at ucdavis.edu<mailto:irezazadeh at ucdavis.edu>>
Date: Tuesday, January 19, 2016 at 7:48 PM
To: EEGLAB List <eeglablist at sccn.ucsd.edu<mailto:eeglablist at sccn.ucsd.edu>>
Cc: Scott Makeig <smakeig at ucsd.edu<mailto:smakeig at ucsd.edu>>
Subject: [Eeglablist] EEG source localization in group level

Hi EEGLABers,
I just wonder if anyone has done a research on the effect of averaging on source localization. In other words and in the group level analysis what is the difference between applying source localization on the grand average data AND applying clustering method ( like EEGLAB) on sources based on individual subjects.
Thanks
Iman

-------------------------------------------------------------
Iman M.Rezazadeh, Ph.D
Semel Intitute, UCLA , Los Angeles


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20160125/cb873c84/attachment.html>