Hi Guillaume,<br><br>Well, it makes sense to my intuitive understanding...<br>The
two components don't cancel each other, but then sum up, more or less.
Does that mean that I could somehow treat them as "one" component?
<br>
I get your point with this superclean dataset example. And it also makes
sense with the other answers I got, with the rank of my data being not
exactly as high as I expect it to be. Now I need to understand why/how
some of my channels turn out to be linearly related...<br>
<br>Thanks,<br>Mx<br><br><div class="gmail_quote">2011/8/24 Guillaume Rousselet <span dir="ltr"><<a href="mailto:Guillaume.Rousselet@psy.gla.ac.uk">Guillaume.Rousselet@psy.gla.ac.uk</a>></span><br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
<div style="word-wrap:break-word"><div class="im">Hey Max,<div><br></div><div>your components don't cancel one another. In your example, the topographies and the time courses have opposite signs, so if you multiply one by the other, your two components are essentially the same.</div>
<div>ICA returns maximally independent components, and therefore can still be correlated. The correlation can be quantified using the mutual information plugin, which I use to confirm my own judgement about component similarity when I look for eye blink ICs.</div>
<div>Imagine you have an absolutely clean signal, with only highly reliable evoked activity and no noise at all. In that case you would expect to get n times your number of electrodes the same component. In my experience, cleaner datasets tend to have more correlated - almost identical - ICs.</div>
<div>Makes sense?</div><div><br></div><div>Best,</div><div><br></div><div>Guillaume</div><div>
<span style="border-collapse:separate;color:rgb(0, 0, 0);font-family:Helvetica;font-style:normal;font-variant:normal;font-weight:normal;letter-spacing:normal;line-height:normal;text-align:auto;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;font-size:medium"><span style="border-collapse:separate;color:rgb(0, 0, 0);font-family:Helvetica;font-style:normal;font-variant:normal;font-weight:normal;letter-spacing:normal;line-height:normal;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;font-size:medium"><div style="word-wrap:break-word">
<span style="border-collapse:separate;color:rgb(0, 0, 0);font-family:Helvetica;font-style:normal;font-variant:normal;font-weight:normal;letter-spacing:normal;line-height:normal;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;font-size:medium"><div style="word-wrap:break-word">
<span style="border-collapse:separate;color:rgb(0, 0, 0);font-family:Helvetica;font-style:normal;font-variant:normal;font-weight:normal;letter-spacing:normal;line-height:normal;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;font-size:medium"><div style="word-wrap:break-word">
<div><div><br>************************************************************************************<br>Guillaume A. Rousselet, Ph.D., senior lecturer & deputy post-graduate convenor<br><br></div><div>Centre for Cognitive Neuroimaging (CCNi)<br>
Institute of Neuroscience and Psychology </div><div><div><div>College of Medical, Veterinary and Life Sciences</div><div>University of Glasgow<br>58 Hillhead Street<br>G12 8QB</div></div></div><div><br></div><div><div><div>
<a href="http://www.psy.gla.ac.uk/staff/index.php?id=GAR01" target="_blank">http://www.psy.gla.ac.uk/staff/index.php?id=GAR01</a></div></div><div><div><br></div></div></div><div>Email: <a href="mailto:Guillaume.Rousselet@glasgow.ac.uk" target="_blank">Guillaume.Rousselet@glasgow.ac.uk</a></div>
<div>Fax. <a href="tel:%2B44%20%280%29141%20330%204606" value="+441413304606" target="_blank">+44 (0)141 330 4606</a></div><div>Tel. <a href="tel:%2B44%20%280%29141%20330%C2%A06652" value="+441413306652" target="_blank">+44 (0)141 330 6652</a></div>
<div>Cell <a href="tel:%2B44%20%280%29791%20779%207833" value="+447917797833" target="_blank">+44 (0)791 779 7833</a></div><div><br></div><div></div><div><div>The University of Glasgow, charity number SC004401</div></div>
<div>************************************************************************************</div></div></div></span></div></span></div></span></span>
</div>
<br></div><div><div class="im"><div>On 23 Aug 2011, at 15:24, Maximilien Chaumon wrote:</div><br></div><blockquote type="cite"><div><div></div><div class="h5">Hi eeglabbers,<br><br>I sometimes get ICs with extremely similar topographies and time courses, like on <a href="http://oszilla.hgs.hu-berlin.de/public/Similar_ICs.PNG" target="_blank">this slide</a>.<br>
I know that ICA returns independent components.<br>
Does that not mean that they should not look the same?<br>I know the components are independent in a statistical sense, which is not
the same as uncorrelated, but still. I'm a bit surprised. What do these two components mean if they cancel one another? well, do they?<br><br>Sorry if my question is naive, but what is happening?<br><br>The data is <a href="http://oszilla.hgs.hu-berlin.de/public/Similar_ICs.mat" target="_blank">here</a>.<br>
<br>Best,<br>Max<br></div></div><div class="im">
_______________________________________________<br>Eeglablist page: <a href="http://sccn.ucsd.edu/eeglab/eeglabmail.html" target="_blank">http://sccn.ucsd.edu/eeglab/eeglabmail.html</a><br>To unsubscribe, send an empty email to <a href="mailto:eeglablist-unsubscribe@sccn.ucsd.edu" target="_blank">eeglablist-unsubscribe@sccn.ucsd.edu</a><br>
For digest mode, send an email with the subject "set digest mime" to <a href="mailto:eeglablist-request@sccn.ucsd.edu" target="_blank">eeglablist-request@sccn.ucsd.edu</a></div></blockquote></div><br></div></blockquote>
</div><br>