[Eeglablist] Frontal Alpha Asymmeytri Score Problem

gökhan alper figen gokhanalper at gmail.com
Mon Feb 22 06:51:34 PST 2021


Thanks  Cedric for the answer,
Participants watched commercial video and i collected their data by eeg. I
want to measure "all time average like/unlike(pleasantness/unpleasantness)
level". I found some papers for asymmetry study on my device, not necessary
re-reference . But i am not sure FAA score value is true or not, because i
did not see any "example value" in any paper or study.
You  sent:
https://urldefense.com/v3/__https://www.sciencedirect.com/science/article/abs/pii/S016787601630784X__;!!Mih3wA!WhvYKJeEll2nmz0Xba58yM0G3lwniDXFVo3FtoA_nedwr2eFprTeOH22AvzTo8LTg3EcQw$ 
this paper very helpful to calculate(window size, overlap etc.,sample rate)
FAA.

Many paper use F3/F4 but some use F7/F8, F5/F6 or  "several frontal dipol
score average"  to measure FAA. How am i decide? which one?
thanks again.

Cedric Cannard <ccannard at protonmail.com>, 22 Şub 2021 Pzt, 09:51 tarihinde
şunu yazdı:

> Hi,
>
> I believe you don’t need to remove baseline as high pass filter does it.
> Average-referencing is generally done before ICA yes, with different ways
> to avoid rank-deficiency:
> https://sccn.ucsd.edu/wiki/Makoto's_preprocessing_pipeline#Re-reference_the_data_to_average_.2808.2F02.2F2020_Updated.29
>
>
> <https://sccn.ucsd.edu/wiki/Makoto's_preprocessing_pipeline#Re-reference_the_data_to_average_.2808.2F02.2F2020_Updated.29>However,
> you need at least 64 channels for average reference. REST-referencing (
> https://urldefense.com/v3/__https://eeglab.org/tutorials/05_Preprocess/rereferencing.html__;!!Mih3wA!WhvYKJeEll2nmz0Xba58yM0G3lwniDXFVo3FtoA_nedwr2eFprTeOH22AvzTo8K5CpAmgg$ ; bottom) can
> be done with at least 16 channels, not sure what happens with 14.
> I am not aware of a paper testing it on this EEG setup. Comparing how
> asymmetry scores are affected by the referencing with another system with
> 16 channels could help.
>
> More info on EEG asymmetry and referencing for your other questions:
> https://urldefense.com/v3/__https://www.sciencedirect.com/science/article/abs/pii/S016787601630784X__;!!Mih3wA!WhvYKJeEll2nmz0Xba58yM0G3lwniDXFVo3FtoA_nedwr2eFprTeOH22AvzTo8LTg3EcQw$ 
>
>
> <https://urldefense.com/v3/__https://www.sciencedirect.com/science/article/abs/pii/S016787601630784X__;!!Mih3wA!WhvYKJeEll2nmz0Xba58yM0G3lwniDXFVo3FtoA_nedwr2eFprTeOH22AvzTo8LTg3EcQw$ >
> Cédric
>
>
> On Sat, Feb 20, 2021 at 05:28, gökhan alper figen <gokhanalper at gmail.com>
> wrote:
>
> Hi everyone,
> i have phd project. My eeg device is emotiv epoc + 14 channel.
> I applied on EEGLAB for raw egg preprocess:
> 1-remove baseline(for time)
> 2-select channel
> 3-load channel location
> 4-remove epoch baseline(DCOFF)
> 5-1-45 Hz Fır-filter
> 6-???? i dont apply re-reference. Because my device reference is
> Mastoid(P3-P4). But i am not sure for this step. I tried
> re-reference(average), but this time ICA calculate one a component is under
> calculated. HELP ME. Is it necessary to re-reference?
> 7- ICA
> 8-Remove for Artifact applied ADJUST
> 9- Last one on matlab, i applied frontal alpha asymmetry algorithm(8-13Hz).
> But i have got another problem in here. I got AF3-AF4, F7-F8, F3-F4,
> FC5-FC6 dipole electrode on frontal lobe. Which dipole is important frontal
> alpha asymmetry? Used one by one at different times in the literature.İn
> addition i have different score same trial(for example: AF3-AF4-->positive,
> F7-F8--> negative, F3-F4-->positive, FC5-FC6-->negative ). is it possible?
>
> FAA SCORE:
> AF4-AF3 F8-F7 F4-F3 FC6-FC5 MeanFAA
> 0,835219 1,803982 0,768355 1,929266 1,334206
> 1,552542 0,523332 1,283658 0,662677 1,005552
> 0,212559 0,767369 -0,09146 0,533678 0,355537
> 2,695686 0,651778 0,208955 1,1194 1,168955
> 0,021023 0,745326 0,28 1,529339 0,643922
> 0,400539 0,693247 0,435739 1,952571 0,870524
> 0,069624 0,294606 1,358425 0,82884 0,637874
> 0,646927 0,528689 0,728926 1,521723 0,856566
> 0,260962 -0,12185 1,093646 1,173347 0,601527
> 1,228531 1,233191 1,044209 1,793257 1,324797
> 0,020307 0,647727 1,570798 1,414552 0,913346
> -0,58441 0,484781 0,804384 1,398788 0,525887
> 1,199436 1,282988 0,766968 1,721898 1,242822
> -0,2001 -0,69111 0,976038 1,513998 0,399708
> 0,184563 1,183316 1,678075 2,736944 1,445725
> -0,10969 0,93127 0,343262 0,51734 0,420546
> -0,22403 0,445983 -0,2119 0,763368 0,193354
>
> thanks
> _______________________________________________
> 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
>
>
>
>


More information about the eeglablist mailing list