[Eeglablist] ICA Misinformation

[Robert Thatcher]

Iman,

     Thankyou for the information.  I could only find a power pointattachment of a simulation in your post.  I did notfind a scientific publication where you compared the phase differences changesbetween an original EEG recording and a ICA reconstruction after removing oneor more components.   Please re-send yourstudy.  Also please give the citation toany of your publications or other’s publications where phase differences werecompared between the original EEG recording and post ICA reconstruction.  It will be interesting to see if you foundsimilar changes like in the study by Montefusco-Siegmund et al or by GeorgesOtte or even in the example pre vs post data files that you can download fromthe internet.  I am assuming that youhave downloaded the EEG data and then used a JTFA like the Hilbert transform oreven the FFT cross-spectrum to prove to yourself that the phase differences betweenthe original and the ICA reconstruction have not been preserved.

 

As for the mathematics concerning reconstruction from alower dimensional matrix to a higher dimensional matrix where there are nosimple linear transforms I refer you to Taken’s theorem where “The reconstruction preserves theproperties of the dynamical system that do not change under smooth coordinatechanges, but it does not preserve the geometric shape of structures in phasespace.”  Also, in standarddifferential geometry math courses the issue of lower dimensional manifold mappingto higher dimensional manifolds shows a loss of information in all cases.  Also, commonsense operates here where one triesto reconstruct 19 channels of EEG using only 15 or 16 or 17 ICA componentshence a loss of information.

 

Finally, the brain isnot a total chaotic organ.   Asdemonstrated by many scientists (e.g., Nunez; Walter Freeman; Roberto-PascualMarqui; E. Roy John; Joel Lubar; etc) coherence and phase differences are well behavedand highly reproducible within and between subjects.  Coherence and phase are dependent on thenumber and strength of connections between groups of neurons.  Here is a URL to a study that tested PaulNunez’s two-compartmental model of Coherence and Phase Differences and foundthat these measures vary as a function of distance and packing density:

http://www.appliedneuroscience.com/TWO-COMPARTMENTAL_MODEL_EEG_COHERENCE.pdf

 

Here is a url to astudy that used EEG LORETA correlations to replicate Diffusion Tensor Imagingmeasures of connectivity in the brain:

http://www.appliedneuroscience.com/DTI-ThatcherHumanBrainmapping.pdf

 

Here is a url to astudy that measured phase lock and phase shift duration from birth to about 16years of age in 458 and where phase differences were stable and wellbehaved:  http://www.appliedneuroscience.com/PhaseresetDevelopment.pdf

 

If you do a search ofthe National Library of Medicine database (Pubmed) using the search terms “EEG coherence”you will find 2,874 citations.  There ishuge consistency in this vast literature which would be impossible if the brainwas totally chaotic.

 

Best regards,

 

Robert
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20170614/a248247b/attachment.html>