<div dir="ltr"><div>Dear List,</div><div><br></div><div>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!</div><div><br></div><div>As far as I know, there are three known limitations in ICA model.</div><div><ol><li>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.</li><li>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).<br></li><li>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)</li></ol><div>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.</div><div><br></div><div>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.</div><div><br></div><div>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.)</div><div><br></div><div>All models are wrong, but some are useful... but I want to go beyond this statement to reach the ground truth of EEG!</div></div><div><br></div><div>-- <br><div class="gmail_signature"><div dir="ltr">Makoto Miyakoshi<br>Swartz Center for Computational Neuroscience<br>Institute for Neural Computation, University of California San Diego<br></div></div>
</div></div>