[Eeglablist] about a blinking period in an EEG protocol

[Tarik S Bel-Bahar]

Greetings Jose, hoping you find the thoughts below useful. Cheers!

*if you haven't had a chance to, check out your topics of interest on
the eeglab list archives which are searchable via Google.

*you might enjoy checking out some articles on eyeblink artifact
reduction and/or cleaning techniques from the last few years. These
include toolboxes for selecting artifactual ICs. See for example the
ADJUST, or SASICA from Chaumon, or PREP from Bigdely-Shamlo, or TAPEEG
eeglab-processing toolboxes, amongst many others.

*one can find articles that show that blink suppression does engage,
for example, ACC regions - and this may be an important and
uncontrolled confound in many studies, especially studies with a
strong "blink-reduction" or "no-blinking" instructions.

*you could have a "blink a lot now" period in between trials or every
couple of trials, or you could have a "response period" where they can
"blink much" and where you don't care about the EEG.

*you could have a "try to blink very little" period that is essentially
a second before the trial begins and up to two or three seconds into
the trial, and then make it clear they can blink outside of that
period.

*you could have a few minutes of just eye blinking and eye moving as a
"localizer" task, where they just blink and move eyes a lot. That
would help you subtract out their characteristic eye artifacts.
Several eeg tools can use such eye-EEG information. It would also be
useful for ICA

*many studies (methods and cognitive research papers) show that
ICA-based removal of eyeblink and eye movement artifacts is easy and
effective

*a good many researchers just allow natural blinking and then simply
subtract out the blinks. See for example the recent Onton and Makeig
chapter in Luck's second Handbook on Event-Related Potentials.

*the answer to the questions of how much "cognitive activity" is
removed when "subtracting" or "regressing" out blinks is not fully
understood at this point. One can compare the results of various
cleaning techniques to determine what degree of "cognitive" EEG
activity is "altered".









On Wed, Jun 24, 2015 at 6:24 PM, José Luis
<joseluisulloafulgeri at gmail.com> wrote:
> Hello,
>
> I’m working at the Clinical Research Unit in the Ville-Evrard Hospital in
> France. We are planning to perform EEG studies in psychiatric subjects. In
> this context, we would like to do classical experiments such as oddball,
> stroop and lexical decision.
>
> Looking the literature on ERP studies, I realise that often these paradigms
> do not include a "blinking" period, i.e., a period where one explicitly ask
> the participant to perform eye movements (blinking or saccades). I'm aware
> of the several methods to correct these artefacts. Also, I'm aware that the
> mental effort involved in suppressing eye blinks may impair task
> performance, and this can be even more pertinent when we deal with
> psychiatric subjects (a simple task can become a double task). However,
> could be this be an issue in a purely visual task? because when a subject
> blinks during the stimulus display she/he is simply not seeing the stimulus
> (or at least not seeing the way it could see it when is not blinking).
> Couldn't be better to have a blinking period? Also, even if we remove
> artefactual activity from the data, I wonder if blinking could have an
> effect on the cognitive process we want to measure,
>
> Any comment will be appreciated,
> José Luis
>
> --
> José Luis ULLOA FULGERI, PhD
> https://sites.google.com/site/joseluisulloafulgeri/
> 33 (0)6.29.50.64.93
>
> _______________________________________________
> 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