[Eeglablist] ICA for eye movement rejection

[Stephen Politzer-Ahles]

Hello Mary-Jane,

My understanding of this (and what I have been doing for my own datasets)
was that the effectiveness of the ICA can be improved by first removing
epochs or artifacts that are unrelated to the artifact you're trying to
clean up with ICA--for example, when I run ICA to clean up ocular
artifacts, I first go through the data and reject epochs with a lot of
muscle artifacts or electrode drifts (I guess a similar thing could be
accomplished using high- and low-pass filters rather than manual artifact
rejection). Then I run the ICA, hopefully get a slightly better
decomposition, and proceed to remove the eye-related components and go back
through the data to reject any epochs that might have eye movement
artifacts remaining after the ICA. But I would also love to hear some more
experienced users' take on this issue.

Best,
Steve Politzer-Ahles
Department of Linguistics
University of Kansas

On Thu, Mar 15, 2012 at 11:30 AM, Budd, Mary-Jane <mbudd at essex.ac.uk> wrote:

> Dear All,
> I know that this has been a much discussed topic but I am rather confused
> by
> some of the responses. I am looking at children's datasets that include
> many
> eye movements (both blinks and  horizontal movements). I have run ICA and
> identified eye blinks and distinct muscle artifacts. If I remove these
> components and then run an automatic rejection procedure (+-75microvolts on
> all electrodes) over half the epochs are rejected due to there still being
> eye movements on the eye and frontal electrode channels. I have read that I
> should not remove the components but instead scan the components and remove
> 'noisy' epochs (I assume this means removing epochs where eye blinks are
> present). RE-running the ICA will then result in 'cleaner' components which
> hopefully will remove the eye movements from my data. I have a couple of
> questions regarding this:
> 1. It would be good to avoid removing epochs as the children blink a  lot
> and so I am likely to lose much data.
> 2. I thought this was the benefit of using ICA for artifact removal as the
> components are removed form the data leaving you with all (or as many as
> possible) epochs to analyse.
> 3. What if after the second ICA I am still left with eye movements i.e.
> can't clearly identify which component is responsible for the eye
> movements?
>
> Have I misunderstood something here? Please help,
> Mary-Jane
>
> Dr Mary-Jane Budd
> Senior Research Officer
> Department of Psychology
> University of Essex
> Wivenhoe Park
> Colchester CO4 3SQ
> UK
>
> Room 4.726
>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



-- 
Stephen Politzer-Ahles
University of Kansas
Linguistics Department
http://www.linguistics.ku.edu/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20120315/a8be789a/attachment.html>