[Eeglablist] Oddly "Wavy" Data

[Stephen Politzer-Ahles]

Hi Matthew,

To me this looks a bit like alpha waves (common when a subject is getting
tired, shutting their eyes, about to doze off, etc.) but I can't really be
sure. I can't read the channel labels in your images, but alpha waves tend
to show up the most robustly on posterior channels, so that would be one
way to tell if that's what they are. If it is alpha waves, there are
different opinions out there about what to do with them. Some experimenters
usually reject stretches with alpha waves, whereas others believe they can
often be left in if they're not too big (I believe Steve Luck's (2005) book
has more information about this in the chapter on artifact rejection; I
don't have a copy handy right now so I don't remember exactly what it
says). In our lab we never try to remove them by filtering, because they
tend to be in more or less the same frequency range as the
(language-related) cognitive activity we're interested in; I'm not sure
what frequency ranges you're interested in so I'm not sure if filtering
would be appropriate or not. In my experience, alpha waves of about this
size have seemed to get more or less averaged out of the ERPs if there are
enough trials, but others on the list may be able to give more feedback.

Good luck,
Steve Politzer-Ahles

On Mon, Feb 27, 2012 at 7:56 AM, Matthew Stief <ms2272 at cornell.edu> wrote:

> Hi Everyone,
>
> I was wondering if anyone has had any experience with the "wavy" activity
> that seems to go across almost all of the channels in this subjects
> data.   I am sure it is familiar to you but I have not seen it in any of
> the text books or guides I can find online and don't have an experienced
> person supervising me.
>
> Here's what the strange activity looks like in filtered, cleaned, epoched,
> and baselined data...
>
>
> http://s1153.photobucket.com/albums/p512/mstief/?action=view&current=wavychannels7.jpg
>
>
> Going back to the same dataset before any processing has been done, you
> can see the same waves.  Obviously this is not particularly clean data, but
> is it salvageable?
>
>
> http://s1153.photobucket.com/albums/p512/mstief/?action=view&current=wavychannels8.jpg
>
>
> Here's the exact same stretch of data after it has gone through 1hz high
> pass and 55hz low pass filters.
>
>
> http://s1153.photobucket.com/albums/p512/mstief/?action=view&current=wavychannels10.jpg
>
>
> It does seem to be affecting the ERPs at least to my untutored eye, here's
> one channel.
>
>
> http://s1153.photobucket.com/albums/p512/mstief/?action=view&current=wavychannels4.jpg
>
>
> So what is it?  Normal?  Horrible?  Fatal or merely deleterious?  Given
> that it seems to contaminate most or all of the data for this person should
> I just throw it out or can I get something out of it?  In particular how
> will ICA handle it?  Do I still have a chance of getting a stable
> decomposition out of it with a reliably identifiable P1?  This person was
> particularly bad but there are less drastic examples in several of my other
> participants.  Keep in mind some of my participants are rather rare and
> would be difficult to replace (I study sexual orientation and some
> varieties are less common than others).
>
> --
> _________________________________________________________________
> Matthew Stief
> Human Development | Sex & Gender Lab | Cornell University
> http://www.human.cornell.edu/HD/sexgender
>
>
> Heterosexuality isn't normal, it's just common.
> -Dorothy Parker
>
> _______________________________________________
> 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/20120227/713beee3/attachment.html>