[Eeglablist] Plot components difference :: Help/clue Welcome

[Horácio Tomé Marques]

Dear all

some time ago (2013) we did an experiment on alpha asymmetry (EEG / EDFs).
As part of the procedures we made the plot of the components to proceed
with the necessary specific removal, and proceed with the rest of the
process.

Because this experiment proved to be very relevant and important (2015), we
wanted to make a confirmation of the findings by doing a complete new
analysis to the original EDFs.

The new plots of the components have the same features/characteristics - as
expected. But, incredibly, one of the trials (an important one) have three
of them with, what seems to us, reversed values.

We already have reviewed more than once the procedures and repeated the
steps to try to understand the possible origin of the problem, but without
success. I think we can swear that we are making the step-by-step exactly
as in the original analysis, but as this plot always shows that difference
something may be escaping us... is anyone able to give us a clue about what
might be happening?

NOTE: we even returned to the original EEGLab version (eeglab10_2_5_8b,
including the same plug-ins, etc), because the first 2015 new ones were
done on eeglab13_4_4b.

Please see the links bellow with the two plots (.jpgs — 2013 vs 2015) where
one can see that the comp1, comp3 and comp4 have the same spectral
distribution logic, but seem to have reversed values.

http://brainchronicles.horacioarts.com/files/Dani_2013PlotCOMP.jpg
http://brainchronicles.horacioarts.com/files/Dani_2015PlotCOMP.jpg


I look forward for a clue!
Thanks for your attention.
Best,


Horácio Tomé-Marques
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20150907/95c5e841/attachment.html>