Hi all,<div><br></div><div>Thanks again for the continued help. I think the skews are for the reasons suggested here… I'm seeing them in a couple subjects in the positive direction as well as the negative, and I've made quite certain that my preprocessing is correct. I'll be mindful of the histos and, for now, continue on with analyses!</div>
<div><br></div><div>Best and thanks again,</div><div>Mike</div><div><br><div class="gmail_quote">On Wed, Jan 4, 2012 at 6:13 PM, Francisco Pereira <span dir="ltr"><<a href="mailto:francisco.pereira@gmail.com">francisco.pereira@gmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="im">On Wed, Jan 4, 2012 at 6:04 PM, J.A. Etzel <<a href="mailto:jetzel@artsci.wustl.edu">jetzel@artsci.wustl.edu</a>> wrote:<br>
><br>
> Non-independence is definitely a big issue (not the only one, unfortunately).<br>
<br>
</div>It's most definitely an issue. Michael Hanke helped me think about<br>
this and we tried to suggest one possibility for testing for the<br>
presence of that in page 7 of<br>
<a href="http://www.princeton.edu/~fpereira/Papers/searchmight.pdf" target="_blank">http://www.princeton.edu/~fpereira/Papers/searchmight.pdf</a><br>
(pardon the self reference)<br>
<br>
There will be code to help diagnose this in the next version of<br>
Searchmight. That said, I think that this is more of a problem in<br>
terms of making results that are marginal appear significant, rather<br>
than introducing bias one way or another.<br>
<div class="im"><br>
> doesn't seem unusual to have a few subjects with below-chance classification<br>
> in large numbers of voxels; I don't think the current methods deal with this (or even explain it) very well.<br>
<br>
</div>Right, with a large number of voxels where the underlying accuracy is<br>
at chance level a fraction of will them will appear much worse than<br>
chance, just as a fraction will appear much better than chance. With<br>
enough subjects...<br>
<span class="HOEnZb"><font color="#888888"><br>
Francisco<br>
</font></span><div class="HOEnZb"><div class="h5"><br>
_______________________________________________<br>
Pkg-ExpPsy-PyMVPA mailing list<br>
<a href="mailto:Pkg-ExpPsy-PyMVPA@lists.alioth.debian.org">Pkg-ExpPsy-PyMVPA@lists.alioth.debian.org</a><br>
<a href="http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/pkg-exppsy-pymvpa" target="_blank">http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/pkg-exppsy-pymvpa</a><br>
</div></div></blockquote></div><br></div>