<html><head></head><body><div style="font-family:Helvetica Neue, Helvetica, Arial, sans-serif;font-size:small;"><div><div><font face="times new roman, new york, times, serif">Makoto,</font></div></div><div><font face="times new roman, new york, times, serif"> The math that you show is not in dispute - it is simple linear algebra for separating two mixed signals. The problem is that it does not address invalid assumptions about ICA decomposition and sources.</font></div><div><font face="times new roman, new york, times, serif"><br></font></div><div><font face="times new roman, new york, times, serif">1- ICA invalid assumption #1 - EEG sources are independent.</font></div><div><font face="times new roman, new york, times, serif"><br></font></div><div><font face="times new roman, new york, times, serif">2- ICA invalid assumption #2 - The Central Limit Theorem regarding Gaussian distributions is invalid.</font></div><div><font face="times new roman, new york, times, serif"><br></font></div><div><font face="times new roman, new york, times, serif">3- ICA invalid assumption #3 - The temporal complexity of any signal mixture is greater than that of its simplest constituent source signal.</font></div><div><font face="times new roman, new york, times, serif"><br></font></div><div>
<!--StartFragment-->
<p style="margin: 6pt 0in; background-image: initial; background-position: initial; background-size: initial; background-repeat: initial; background-attachment: initial; background-origin: initial; background-clip: initial;"><font face="times new roman, new york, times, serif"><span style="color: rgb(34, 34, 34);">Those principles contribute to the basic establishment of ICA.
If the signals we happen to extract from a set of mixtures are independent like
sources signals, or have non-Gaussian histograms like source signals, or have
low complexity like source signals, then they must be source signals.</span><span style="color: rgb(34, 34, 34);"></span></font></p>
<p style="margin: 6pt 0in; background-image: initial; background-position: initial; background-size: initial; background-repeat: initial; background-attachment: initial; background-origin: initial; background-clip: initial;"><span style="color: rgb(34, 34, 34);"><font face="times new roman, new york, times, serif"> </font></span></p>
<p style="margin: 6pt 0in; background-image: initial; background-position: initial; background-size: initial; background-repeat: initial; background-attachment: initial; background-origin: initial; background-clip: initial;"><span style="color: rgb(34, 34, 34);"><font face="times new roman, new york, times, serif">Now here is the problem in order for the ICA to identify
anything: THE SOURCES HAVE TO BE INDEPENDENT. Moreover when the people who create
ICA talk about sources they do not seem to address the inverse
solution (von Helmholtz, Maxwell, Poisson, etc) and that is a mistake in our opinion. Let's begin, with the inverse
solution and find brain areas that can be:</font></span><span style="font-family: Arial, sans-serif; color: rgb(34, 34, 34);"></span></p>
<p style="margin: 6pt 0in; background-image: initial; background-position: initial; background-size: initial; background-repeat: initial; background-attachment: initial; background-origin: initial; background-clip: initial;"><span style="font-family: Arial, sans-serif; color: rgb(34, 34, 34);"> </span></p>
<p style="margin: 6pt 0in; background-image: initial; background-position: initial; background-size: initial; background-repeat: initial; background-attachment: initial; background-origin: initial; background-clip: initial;"><span style="font-family: Arial, sans-serif; color: rgb(34, 34, 34);">1. Non-independent and what is more correlated.</span><span style="font-family: Arial, sans-serif; color: rgb(34, 34, 34);"></span></p>
<p style="margin: 6pt 0in; background-image: initial; background-position: initial; background-size: initial; background-repeat: initial; background-attachment: initial; background-origin: initial; background-clip: initial;"><span style="font-family: Arial, sans-serif; color: rgb(34, 34, 34);"> </span></p>
<p style="margin: 6pt 0in; background-image: initial; background-position: initial; background-size: initial; background-repeat: initial; background-attachment: initial; background-origin: initial; background-clip: initial;"><span style="font-family: Arial, sans-serif; color: rgb(34, 34, 34);">2. EEG sources are highly temporal correlated because information is flowing from one area to another and this means temporal correlations. </span><span style="font-family: Arial, sans-serif; color: rgb(34, 34, 34);"></span></p>
<p style="margin: 6pt 0in; background-image: initial; background-position: initial; background-size: initial; background-repeat: initial; background-attachment: initial; background-origin: initial; background-clip: initial;"><span style="font-family: Arial, sans-serif; color: rgb(34, 34, 34);"> </span></p>
<p style="margin: 6pt 0in; background-image: initial; background-position: initial; background-size: initial; background-repeat: initial; background-attachment: initial; background-origin: initial; background-clip: initial;"><span style="font-family: Arial, sans-serif; color: rgb(34, 34, 34);">3. The number of independent components ICA are correlated with the number of sensors.</span><span style="font-family: Arial, sans-serif; color: rgb(34, 34, 34);"></span></p>
<p style="margin: 6pt 0in; background-image: initial; background-position: initial; background-size: initial; background-repeat: initial; background-attachment: initial; background-origin: initial; background-clip: initial;"><span style="font-family: Arial, sans-serif; color: rgb(34, 34, 34);"> </span></p>
<p style="margin: 6pt 0in; background-image: initial; background-position: initial; background-size: initial; background-repeat: initial; background-attachment: initial; background-origin: initial; background-clip: initial;"><span style="font-family: Arial, sans-serif; color: rgb(34, 34, 34);">So the problem is not a mathematical problem itself, your math is correct but it does not solve the problem. The main problem is
that ICA cannot find sources that are correlated and for that reason it puts them under the same component and when one removes that component then one removes all of that information. Of course by removing all the information you also
will remove the phase information that goes with it and that is the phase
distortion that every one can demonstrate. This problem is worse because if we take into account
that we can't have more IC's components than sensors when you have 19 channels
then the ICA method is forced to put things together and there is no way around this problem because one needs to explain the data, and by doing that if you remove a single component
you will remove a lot of information and that is what you may be missing because you are only focusing on the math and forgot 2 things:</span><span style="font-family: Arial, sans-serif; color: rgb(34, 34, 34);"></span></p>
<p style="margin: 6pt 0in; background-image: initial; background-position: initial; background-size: initial; background-repeat: initial; background-attachment: initial; background-origin: initial; background-clip: initial;"><span style="font-family: Arial, sans-serif; color: rgb(34, 34, 34);">1. The ICA assumptions </span><span style="font-family: Arial, sans-serif; color: rgb(34, 34, 34);"></span></p>
<div style="margin: 6pt 0in; background-image: initial; background-position: initial; background-size: initial; background-repeat: initial; background-attachment: initial; background-origin: initial; background-clip: initial;"><span style="font-family: Arial, sans-serif; color: rgb(34, 34, 34);">2. The reality of the physiology and physics of the signals that one is looking at.</span></div><div style="margin: 6pt 0in; background-image: initial; background-position: initial; background-size: initial; background-repeat: initial; background-attachment: initial; background-origin: initial; background-clip: initial;"><span style="font-family: Arial, sans-serif; color: rgb(34, 34, 34);"><br></span></div><div style="margin: 6pt 0in; background-image: initial; background-position: initial; background-size: initial; background-repeat: initial; background-attachment: initial; background-origin: initial; background-clip: initial;"><span style="font-family: Arial, sans-serif; color: rgb(34, 34, 34);">Kind regards,</span></div><div style="margin: 6pt 0in; background-image: initial; background-position: initial; background-size: initial; background-repeat: initial; background-attachment: initial; background-origin: initial; background-clip: initial;"><span style="font-family: Arial, sans-serif; color: rgb(34, 34, 34);"><br></span></div><div style="margin: 6pt 0in; background-image: initial; background-position: initial; background-size: initial; background-repeat: initial; background-attachment: initial; background-origin: initial; background-clip: initial;"><span style="font-family: Arial, sans-serif; color: rgb(34, 34, 34);">Robert</span></div>
<!--EndFragment--></div><div><font size="3"><br></font></div><div><br></div><div id="yahoo_quoted_2253875939" class="yahoo_quoted"><div style="font-family:'Helvetica Neue', Helvetica, Arial, sans-serif;font-size:13px;color:#26282a;"><div>On Tuesday, August 1, 2017, 5:19:10 PM EDT, Makoto Miyakoshi <mmiyakoshi@ucsd.edu> wrote:</div><div><br></div><div><br></div><div><div id="yiv1645420803"><html><div><div dir="ltr">Dear Robert,<div><br clear="none"></div><div>I paste a link below to show<b> mathematical process of how phase CHANGES after rejecting a component</b> obtained by a linear method.</div><div><span style="font-size:12.8px;">You are calling the difference between 18 Bitcoins and 19 Bitcoins 'distortion'. It's a due change.</span></div><div><div style="font-size:12.8px;"><a rel="nofollow" shape="rect" target="_blank" href="https://sccn.ucsd.edu/wiki/How_phase_is_calculated_in_linear_decomposition">https://sccn.ucsd.edu/wiki/ How_phase_is_calculated_in_ linear_decomposition</a></div></div><div style="font-size:12.8px;"><br clear="none"></div><div style="font-size:12.8px;">If your understanding is different, <b>please show it in math</b>.</div><div style="font-size:12.8px;"><br clear="none"></div><div style="font-size:12.8px;">Makoto</div></div><div class="yiv1645420803gmail_extra"><br clear="none"><div class="yiv1645420803gmail_quote">On Thu, Jul 27, 2017 at 12:30 PM, Robert Thatcher <span dir="ltr"><<a rel="nofollow" shape="rect" ymailto="mailto:rwthatcher2@yahoo.com" target="_blank" href="mailto:rwthatcher2@yahoo.com">rwthatcher2@yahoo.com</a>></span> wrote:<br clear="none"><blockquote class="yiv1645420803gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;"><div class="yiv1645420803yqt7620872945" id="yiv1645420803yqt31613"><div><div style="font-family:Helvetica Neue, Helvetica, Arial, sans-serif;font-size:small;"><div><div><font size="3">Dear Makoto,</font></div><div><font size="3"><br clear="none"></font></div><div><font size="3">You stated: "So far I know Montefusco-Siegmunt et al. (2013) is the only paper that makes this invalid claim. If you know other papers, please let me know."<br clear="none"></font></div></div><div><font size="3"><br clear="none"></font></div><div><font size="3">The phase distortion by ICA reconstruction was only discovered in 2014 so there are not a lot of publications on this topic. However, you are the author of one publication yourself on this Eeglablist.</font></div><div><font size="3"><br clear="none"></font></div><div><font size="3">For example: “If you remove IC and
reconstruct channel EEG by back projecting the remaining ICs, of course it
changes channel EEG phase!” (Makoto Miyakoshi, Eeglablist ICA and signal phase
content, Sept. 16, 2014) </font></div><div><font size="3"><br clear="none"></font></div><div><font size="3">The proof of the distortion was discovered and validated by comparing the phase differences of the original EEG to the ICA reconstruction time series thereby invalidating the cross-spectrum which is essential for network analyses and also inverse source solutions. The proof is by observation and mathematics for example by yourself and the following other Eeglablist publishers:</font></div><span class="yiv1645420803"></span><div><font size="3"><br clear="none"></font></div><div><p class="yiv1645420803m_-4508916694874837763ydp3a98696dMsoNormal"><font size="3">“The EEG
reconstruction after removing bad components/sources MAY change the phase value
of the signal at any electrode.” (M. Rezazadeh Eeglablist ICA and signal phase
content, Sept. 18, 2014).</font></p>
<p class="yiv1645420803m_-4508916694874837763ydp3a98696dMsoNormal"><font size="3"> </font></p>
<p class="yiv1645420803m_-4508916694874837763ydp3a98696dMsoNormal"><font size="3"><span style="">“The
reconstructed data after removing spurious ICA components differs from the
original time series, and because of that there are phase differences.” (Arnaud
Delorme, </span>Eeglablist
ICA misinformation, June 10, 2017).</font></p>
<p class="yiv1645420803m_-4508916694874837763ydp3a98696dMsoNormal"><font size="3"> <span style="">“</span><span style="">I first noticed the problem with phase distortion more than a
decade ago” (Robert Lawson, </span>Eeglablist ICA misinformation, June 14, 2017).</font></p>
<p class="yiv1645420803m_-4508916694874837763ydp3a98696dMsoNormal"><font size="3"><span style="">“I
think Bob is right that the relative phase will be changed by deleting 1 or 2
artifact components.” (Ramesh Srinivasan, </span>Eeglablist ICA misinformation, June 14, 2017).</font></p>
<p class="yiv1645420803m_-4508916694874837763ydp3a98696dMsoNormal"><font size="3"><span style="">“We
found phase distortions in the 8-10 Hz alfa band (greatest near the source of
artefact) but also on more remote electrodes such as occipital and also in
artefact free strokes of EEG.” (Georges Otte, </span>Eeglablist ICA misinformation, June 15,
2017).</font></p></div><div><font size="3"><br clear="none"></font></div><div><font size="3">Additional proof is by direct comparisons like Arno did showing about 98% of the phase differences are statistically significantly altered at P < 0.0001. Here is a url to some of the statistics and tutorial demonstrations that allow one to verify for themselves:</font></div><div><font size="3"><a rel="nofollow" shape="rect" target="_blank" href="http://www.appliedneuroscience.com/Tutorial_Adulteration_Phase_Relations_when_using_ICA.pdf">http://www. appliedneuroscience.com/ Tutorial_Adulteration_Phase_ Relations_when_using_ICA.pdf</a><br clear="none"></font></div><div><font size="3"><br clear="none"></font></div><div><font size="3">Myself and colleaques will be publishing more statistical comparisons and also show how ICA reconstruction distorts other network measures such as the Phase Slope Index and phase shift and phase lock duration and phase-amplitude coupling and cross-frequency coupling, etc.</font></div><div><font size="3"><br clear="none"></font></div><div><font size="3">Other publications are: </font></div><div><font size="3"> Bridwell et al (2016) <b>Spatiospectral Decomposition of
Multi-subject EEG: Evaluating Blind Source Separation Algorithms on Real and
Realistic Simulated Data</b>. Brain Topogr DOI 10.1007/s10548-016-0479-1 Feb
2016 - see page 13 where they state: “The current group spatiospectral BSS approach
discards phase information …” (Pg 13).</font></div><div><font size="3"><br clear="none"></font></div><div><font size="3">R.W. Thatcher (2012) "Handbook of QEEG and EEG Biofeedback" , Anipublishing Co., St. Petersburg, Fl</font></div><div><font size="3"><br clear="none"></font></div><div><font size="3">Otte, G. "ICA Reconstruction" Presented at the ANT workshop, Beaune, France</font></div><div><font size="3"><br clear="none"></font></div><div><font size="3">I hope that this is helpful.</font></div><div><font size="3"><br clear="none"></font></div><div><font size="3">Best wishes,</font></div><div><font size="3"><br clear="none"></font></div><div><font size="3">Robert</font></div><div><br clear="none"></div><div class="yiv1645420803m_-4508916694874837763yahoo_quoted" id="yiv1645420803m_-4508916694874837763yahoo_quoted_2098300648"><div style="font-family:'Helvetica Neue', Helvetica, Arial, sans-serif;font-size:13px;color:#26282a;"><div>On Thursday, July 27, 2017, 1:58:05 AM EDT, otte georges <<a rel="nofollow" shape="rect" ymailto="mailto:georges.otte@pandora.be" target="_blank" href="mailto:georges.otte@pandora.be">georges.otte@pandora.be</a>> wrote:</div><div><br clear="none"></div><div><br clear="none"></div><div><div id="yiv1645420803m_-4508916694874837763yiv0984032019"><div><div class="yiv1645420803m_-4508916694874837763yiv0984032019WordSection1"><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US" style="color:#993366;">Dear Bob</span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US" style="color:#993366;"> </span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US" style="color:#993366;">I reposted thismessage below to the EEGLablist and asked Makoko what caused his opinion switch since 2014 ….</span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US" style="color:#993366;"> </span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US" style="color:#993366;">Maybe another mail that will get “lost”…. ? We’ll wait and see….</span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US" style="color:#993366;"> </span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US" style="color:#993366;">Sincerely</span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US" style="color:#993366;"> </span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US" style="color:#993366;"> </span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US" style="color:#993366;">Georges</span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US" style="color:#993366;"> </span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US" style="color:#993366;"> </span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US" style="color:#993366;"> </span></p><div><div style="border:none;border-top:solid #e1e1e1 1.0pt;padding:3.0pt 0cm 0cm 0cm;"><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><b>From:</b> otte georges [mailto:<a rel="nofollow" shape="rect" ymailto="mailto:georges.otte@pandora.be" target="_blank" href="mailto:georges.otte@pandora.be">georges.otte@pandora. be</a>] <br clear="none"><b>Sent:</b> Thursday, July 27, 2017 7:55 AM<br clear="none"><b>To:</b> '<a rel="nofollow" shape="rect" ymailto="mailto:mmiyakoshi@ucsd.edu" target="_blank" href="mailto:mmiyakoshi@ucsd.edu">mmiyakoshi@ucsd.edu</a>' <<a rel="nofollow" shape="rect" ymailto="mailto:mmiyakoshi@ucsd.edu" target="_blank" href="mailto:mmiyakoshi@ucsd.edu">mmiyakoshi@ucsd.edu</a>><br clear="none"><b>Subject:</b> RE: Beyond good and evil of ICA</p></div></div><span class="yiv1645420803"></span><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"> </p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US" style="font-size:14.0pt;color:#1f497d;">Dear Makoto</span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US" style="font-size:14.0pt;color:#1f497d;"> </span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US" style="font-size:14.0pt;color:#1f497d;">Below is the mail I have send to the EEGLab list and that could maybe also be relevant as a reply to Mr. Andrew Smart. </span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US" style="font-size:14.0pt;color:#1f497d;"> </span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US" style="font-size:14.0pt;color:#1f497d;">I can imagine that managing a busy list in extra time is quite a hard task so therefore I can understand that messages get lost in transit. No offense taken.</span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US" style="font-size:14.0pt;color:#1f497d;"> </span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US" style="font-size:14.0pt;color:#1f497d;">PS the bitcoin image (a string of chars) is mine. It reflects to the fact that if one has 19 strings or components and omits 4 or 5 the reconstructed ones will not be accepted as true bitcoins. In case I am wrong I will send you my sincere apologies and some char strings (just joking)</span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US" style="font-size:14.0pt;color:#1f497d;"> </span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US" style="font-size:14.0pt;color:#1f497d;">Sincerely</span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US" style="font-size:14.0pt;color:#1f497d;"> </span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US" style="font-size:14.0pt;color:#1f497d;">PS: (no joke) in a mail You send me in 2014 when we had this discussion again you did state that ICA reconstruction does indeed change phase relations between channels. What causes Your switch of opinion ?</span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US" style="font-size:14.0pt;color:#1f497d;"> </span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US" style="font-size:14.0pt;color:#1f497d;">Sincerely</span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US" style="font-size:14.0pt;color:#1f497d;"> </span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US" style="font-size:14.0pt;color:#1f497d;">Georges</span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US" style="font-size:14.0pt;color:#1f497d;"> </span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US" style="font-size:14.0pt;color:#1f497d;">Mail of july</span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US" style="font-size:14.0pt;color:#1f497d;"> </span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US">There is a major evolution in modern neuropsychiatry that aims at linking clinical symptoms to brain network dysfunctions. While this approach was successful in grounding neurological symptoms to structural pathologic alternations in brain networks, in psychiatry the main momentum was not structural but functional network dysfunctions. While fMRI was the pioneer, the much better time resolution of MEG and EEG made them the preferred tools. Their output ( time series) is but a means to further construct a functional image of the networks involved where phase dynamics teach us the directionality of the information flow in the network nodes and allows us by comparison with a database of normal values what functional abnormalities can be detected. For me phase integrity in the data is thus very important to be able to construct valuable graph theory models of those networks be them dysfunctional or compensatory. Much work has been devoted on this topic since many decades by DrThatcher but also by many other authors such as Vinod Menon ( Stanford) linking psychiatric symptoms to specific network dysfunctions. For us, clinicians this introduces a new approach to neuroscientific psychiatry that links psychiatry back to it's neurobiological roots and can hopefully one day send the DSM categorization to the museum of the history of psychiatry.</span></p><span class="yiv1645420803"></span><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US">As phase is IMHO a most important parameter in order to establish the network internode information flow, it should not come as a surprise to hear that some find phase unimportant as contaminated by continuous artefact or hear about ICA’ s a signal reconstruction method that presents the danger of changing the phase dynamics in the original time series especially in low channel (19ch) recordings with perhaps more prominent effect due to overcompleteness. </span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US">If in a 19 ch. EEG a clinician rejects (nulls out the rows of the mixing matrix ) ICAas components for blinks EMG, pletysmo and ecg ( 4 ) and then does a "reconstruction" ( creating 19 channels out of 15 ??) what we then get might look nice but is IMHO not a valid base for a graph theoretical model of the underlying brain network.</span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US">I think this is the reason this discussion is important and certainly not a trivial pro or contra ICA pugilism. </span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US"> </span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="NL-BE">Sincerely</span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="NL-BE"> </span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="NL-BE">Georges</span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US" style="font-size:14.0pt;color:#1f497d;"> </span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US" style="color:#1f497d;"> </span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><b><span lang="EN-US">From:</span></b><span lang="EN-US"> Makoto Miyakoshi [</span><a rel="nofollow" shape="rect" ymailto="mailto:mmiyakoshi@ucsd.edu" target="_blank" href="mailto:mmiyakoshi@ucsd.edu"><span lang="EN-US">mailto:mmiyakoshi@ucsd.edu</span></a><span lang="EN-US">] <br clear="none"><span class="yiv1645420803"><b>Sent:</b> Wednesday, July 26, 2017 11:40 PM<br clear="none"><b>To:</b> Robert Thatcher <</span></span><a rel="nofollow" shape="rect" ymailto="mailto:rwthatcher2@yahoo.com" target="_blank" href="mailto:rwthatcher2@yahoo.com"><span lang="EN-US">rwthatcher2@yahoo.com</span></a><span lang="EN-US">><br clear="none"><b>Cc:</b> EEGLAB List <</span><a rel="nofollow" shape="rect" ymailto="mailto:eeglablist@sccn.ucsd.edu" target="_blank" href="mailto:eeglablist@sccn.ucsd.edu"><span lang="EN-US">eeglablist@sccn.ucsd.edu</span></a><span lang="EN-US">>; Georges Otte <</span><a rel="nofollow" shape="rect" ymailto="mailto:georges.otte@telenet.be" target="_blank" href="mailto:georges.otte@telenet.be"><span lang="EN-US">georges.otte@telenet.be</span></a><span lang="EN-US">><br clear="none"><b>Subject:</b> Re: Beyond good and evil of ICA</span></p><div><div class="yiv1645420803h5"><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span lang="EN-US"> </span></p><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal">Dear Robert,</p><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"> </p></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal">I want to know all publications that makes a clear claim that 'ICA distorts phase'. I will include all of them for our clarification paper. So far I know Montefusco-Siegmunt et al. (2013) is the only paper that makes this invalid claim. If you know other papers, please let me know.</p></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"> </p></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal">Again, you are calling the difference between 18 Bitcoins and 19 Bitcoins 'distortion'. It's a due change. See the pages below.</p></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><a rel="nofollow" shape="rect" target="_blank" href="https://sccn.ucsd.edu/wiki/How_phase_is_calculated_in_linear_decomposition">https://sccn.ucsd.edu/wiki/ How_phase_is_calculated_in_ linear_decomposition</a></p></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><a rel="nofollow" shape="rect" target="_blank" href="https://sccn.ucsd.edu/wiki/ICA_phase_distortion">https://sccn.ucsd.edu/wiki/ ICA_phase_distortion</a></p></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"> </p></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal">Georges, Ramon told me that all the posts were published on the list. If otherwise, please let us know. Sorry for the trouble.</p></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"> </p></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal">Makoto</p></div></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"> </p><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal">On Wed, Jul 26, 2017 at 12:06 PM, Robert Thatcher <<a rel="nofollow" shape="rect" ymailto="mailto:rwthatcher2@yahoo.com" target="_blank" href="mailto:rwthatcher2@yahoo.com">rwthatcher2@yahoo.com</a>> wrote:</p><blockquote style="border:none;border-left:solid #cccccc 1.0pt;padding:0cm 0cm 0cm 6.0pt;margin-left:4.8pt;margin-top:5.0pt;margin-right:0cm;margin-bottom:5.0pt;"><div><div><div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span style="font-size:12.0pt;">Dear Makoto,</span></p></div></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span style="font-size:12.0pt;"> I think your criticisms are important and note that there are traveling waves in the EEG and also there is nonlinearity in the form of wave dispersion as noted by Nunez, 1981 and demonstrated in the paper that can be downloaded at this url (see Table IV):</span></p></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span style="font-size:12.0pt;"> </span></p></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span style="font-size:12.0pt;"><a rel="nofollow" shape="rect" target="_blank" href="http://www.appliedneuroscience.com/TWO-COMPARTMENTAL_MODEL_EEG_COHERENCE.pdf">http://www. appliedneuroscience.com/TWO- COMPARTMENTAL_MODEL_EEG_ COHERENCE.pdf</a></span></p></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span style="font-size:12.0pt;"> </span></p></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span style="font-size:12.0pt;">It seems that your 3rd criticism does not recognize that ICA reconstruction of a new time series violates the "Reciprocity" theorem of Helmoltz and the "Lead Field" necessary for a valid inverse solution.</span></p></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span style="font-size:12.0pt;"> </span></p></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span style="font-size:12.0pt;">You mentioned a recent criticism on ICA that you stated is "technically invalid". I doubt that you are referring to the criticism about ICA reconstruction adulterating phase differences between EEG channels? The issue of ICA reconstruction and phase alteration is a settled issue based on math (not the separation of mixtures of phase or frequencies but rather the cross-spectrum at the same frequency at different locations) as well as multiple empirical demonstrations and tutorial demonstrations that anyone can verify for themselves. Also, I am copying from the Eelablist statements by yourself and five others agreeing that ICA reconstruction alters phase differences. </span></p></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019m2860405150461759669ydpd4758496msonormal"><span style="font-size:14.0pt;">“If you remove IC and reconstruct channel EEG by back projecting the remaining ICs, of course it changes channel EEG phase!” (Makoto Miyakoshi, Eeglablist ICA and signal phase content, Sept. 16, 2014) </span><span style="font-size:12.0pt;"></span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019m2860405150461759669ydpd4758496msonormal"><span style="font-size:14.0pt;"> </span><span style="font-size:12.0pt;"></span></p><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span style="font-size:14.0pt;">“The EEG reconstruction after removing bad components/sources MAY change the phase value of the signal at any electrode.” (M. Rezazadeh Eeglablist ICA and signal phase content, Sept. 18, 2014).</span><span style="font-size:12.0pt;"></span></p></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span style="font-size:12.0pt;"> </span></p></div><p class="yiv1645420803m_-4508916694874837763yiv0984032019m2860405150461759669ydpd4758496msonormal"><span style="font-size:14.0pt;">“The reconstructed data after removing spurious ICA components differs from the original time series, and because of that there are phase differences.” (Arnaud Delorme, </span><span style="font-size:14.0pt;">Eeglablist ICA misinformation, June 10, 2017).</span><span style="font-size:12.0pt;"></span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019m2860405150461759669ydpd4758496msonormal"><span style="font-size:14.0pt;"> </span><span style="font-size:12.0pt;"></span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019m2860405150461759669ydpd4758496msonormal"><span style="font-size:14.0pt;"> </span><span style="font-size:14.0pt;">“I first noticed the problem with phase distortion more than a decade ago” (Robert Lawson, </span><span style="font-size:14.0pt;">Eeglablist ICA misinformation, June 14, 2017).</span><span style="font-size:12.0pt;"></span></p><p class="yiv1645420803m_-4508916694874837763yiv0984032019m2860405150461759669ydpd4758496msonormal"><span style="font-size:14.0pt;"> </span><span style="font-size:12.0pt;"></span></p><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span style="font-size:14.0pt;">“I think Bob is right that the relative phase will be changed by deleting 1 or 2 artifact components.” (Ramesh Srinivasan, </span><span style="font-size:14.0pt;">Eeglablist ICA misinformation, June 14, 2017).</span><span style="font-size:12.0pt;"></span></p></div></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span style="font-size:12.0pt;"> </span></p></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019m2860405150461759669ydpceab13d3msonormal"><span style="font-size:14.0pt;">“We found phase distortions in the 8-10 Hz alfa band (greatest near the source of artefact) but also on more remote electrodes such as occipital and also in artefact free strokes of EEG.” (Georges Otte, </span><span style="font-size:14.0pt;">Eeglablist ICA misinformation, June 15, 2017).</span><span style="font-size:12.0pt;"></span></p></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span style="font-size:12.0pt;">Best regards,</span></p></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span style="font-size:12.0pt;"> </span></p></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span style="font-size:12.0pt;">Robert</span></p></div><div><div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span style="font-size:12.0pt;"> </span></p></div><div id="yiv1645420803m_-4508916694874837763yiv0984032019m_2860405150461759669ydpb3d79c09yahoo_quoted_1295088611"><div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span style="font-size:10.0pt;">On Wednesday, July 26, 2017, 2:01:59 PM EDT, Makoto Miyakoshi <<a rel="nofollow" shape="rect" ymailto="mailto:mmiyakoshi@ucsd.edu" target="_blank" href="mailto:mmiyakoshi@ucsd.edu">mmiyakoshi@ucsd.edu</a>> wrote:</span></p></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span style="font-size:10.0pt;"> </span></p></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span style="font-size:10.0pt;"> </span></p></div><div><div id="yiv1645420803m_-4508916694874837763yiv0984032019m_2860405150461759669ydpb3d79c09yiv4477574556"><div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span style="font-size:10.0pt;">Dear List,</span></p></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span style="font-size:10.0pt;"> </span></p></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span style="font-size:10.0pt;">Recently there was a criticism against ICA on the list. Unfortunately it is technically invalid so I remained unsatisfied. Let me share real problems of the ICA model (Onton and Makeig 2006) to re-do it. This is a continued discussion from the one titled 'How phase is calculated in linear decomposition' and now this is my turn to criticize ICA!</span></p></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span style="font-size:10.0pt;"> </span></p></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span style="font-size:10.0pt;">As far as I know, there are three known limitations in ICA model.</span></p></div><div><ol start="1" type="1"><li class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal" style="color:#26282a;margin-left:0cm;"><span style="font-size:10.0pt;">Spatial stationarity. I have seen a nice traveling waves in ECoG grid data during Joaquin Repela's presentation at SCCN. This clearly violates the assumptions of spatial stationarity in ICA.</span></li><li class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal" style="color:#26282a;margin-left:0cm;"><span style="font-size:10.0pt;">Temporal stationarity. Shawn Hsu at SCCN presented time-series data of ICA model likelihood during drowsy driving task. Also, Jason Palmer's AMICA also demonstrated temporal changes in model likelihood. So one model per data does not fit the truth (unless the task has a strong control over a subject's cognitive and behavioral states).</span></li><li class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal" style="color:#26282a;margin-left:0cm;"><span style="font-size:10.0pt;">Dipolar source model. Although most of ICA results are fit with dipole models, it seems ICA also returns (probably) non-point sources. When one fits a dipole model to such a non-point source, the location tend to end up with physiologically invalid depth (this is the most annoying thing about ICA today)</span></li></ol><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span style="font-size:10.0pt;">I'd like to hear detailed criticism about these points. Note I saw these critical counterevidence in SCCN; we are not a boring ICA cult who have blind belief in it.</span></p></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span style="font-size:10.0pt;"> </span></p></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span style="font-size:10.0pt;">Nonetheless, ICA model has a critical merit. I named it <i>Independence-Dipolarity Identity (I-D Identity, or IDId)</i>. I-D Identity means that when ICA solves temporally problem, it also solves spatial problem at the same time <i>without using ANY spatial constraint</i>. Dipolarity can be thought of, in short, <i>biophysical origin-ness</i>. Hence I believe that this is evidence that ICA hits <i>some </i>physiological truth of EEG generation.</span></p></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span style="font-size:10.0pt;"> </span></p></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span style="font-size:10.0pt;">There could be multiple criticisms against the limitations of ICA model, but at the same time any criticism, at least so far, was NOT strong enough to deny <i>I-D Identity </i>of the ICA model<i>. </i>After all, because of this <i>I-D Identity</i>, I still advocate ICA (but similar dipolarity can be achieved by using very different approach, such as SOBI... so independence is not the only requirement to reach the biophysical validity. It's still a mystery to me.)</span></p></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span style="font-size:10.0pt;"> </span></p></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span style="font-size:10.0pt;">All models are wrong, but some are useful... but I want to go beyond this statement to reach the ground truth of EEG!</span></p></div></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span style="font-size:10.0pt;"> </span></p></div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span style="font-size:10.0pt;">-- </span></p><div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><span style="font-size:10.0pt;">Makoto Miyakoshi<br clear="none">Swartz Center for Computational Neuroscience<br clear="none">Institute for Neural Computation, University of California San Diego</span></p></div></div></div></div></div></div></div></div></div></div></div></div></blockquote></div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"><br clear="none"><br clear="all"></p><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal"> </p></div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal">-- </p><div><div><p class="yiv1645420803m_-4508916694874837763yiv0984032019MsoNormal">Makoto Miyakoshi<br clear="none">Swartz Center for Computational Neuroscience<br clear="none">Institute for Neural Computation, University of California San Diego</p></div></div></div></div></div></div></div></div></div></div></div></div></div></div><br clear="none">______________________________ _________________<br clear="none">
Eeglablist page: <a rel="nofollow" shape="rect" target="_blank" href="http://sccn.ucsd.edu/eeglab/eeglabmail.html">http://sccn.ucsd.edu/eeglab/ eeglabmail.html</a><br clear="none">
To unsubscribe, send an empty email to <a rel="nofollow" shape="rect" ymailto="mailto:eeglablist-unsubscribe@sccn.ucsd.edu" target="_blank" href="mailto:eeglablist-unsubscribe@sccn.ucsd.edu">eeglablist-unsubscribe@sccn. ucsd.edu</a><br clear="none">
For digest mode, send an email with the subject "set digest mime" to <a rel="nofollow" shape="rect" ymailto="mailto:eeglablist-request@sccn.ucsd.edu" target="_blank" href="mailto:eeglablist-request@sccn.ucsd.edu">eeglablist-request@sccn.ucsd. edu</a><br clear="none"></blockquote></div><br clear="none"><br clear="all"><div><br clear="none"></div>-- <br clear="none"><div class="yiv1645420803gmail_signature"><div dir="ltr">Makoto Miyakoshi<br clear="none">Swartz Center for Computational Neuroscience<br clear="none">Institute for Neural Computation, University of California San Diego<br clear="none"></div></div>
</div></div></html></div></div></div></div></div></body></html>