<html dir="ltr">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<style type="text/css" id="owaParaStyle"></style><style type="text/css"></style><style type="text/css"></style>
</head>
<body fpstyle="1" ocsi="0">
<div style="direction: ltr;font-family: Tahoma;color: #000000;font-size: 10pt;">
<div><span style="font-family: 'Times New Roman';">Hi everyone,</span></div>
<div><span style="font-family: 'Times New Roman';"><br>
</span></div>
<div><span style="font-family: 'Times New Roman';">I have a question regarding a rank inconsistency problem that arises when we run ICA on a dataset that has been re-referenced to averaged mastoids. The problem appears related to the fact that we have re-referenced
offline to the average of two channels, and ICA "expects" to find 65 channels (the total number we begin with) but only finds 64. </span><span style="font-family: 'Times New Roman'; font-size: 10pt;">As a consequence it defaults to using PCA to reduce the
number of dimensions - something we'd prefer to avoid with these particular data. </span></div>
<span style="font-family: 'Times New Roman';">
<div><span style="font-family: 'Times New Roman';"><br>
</span></div>
<div><span style="font-family: 'Times New Roman';">The problem arises regardless of the ICA algorithm we use and also apparently regardless of the re-referencing parameters we use (as long as two reference channels are involved - one channel works fine). </span><span style="font-size: 10pt;">I
should also note that the data were recorded in ANT as .cnt files, but I'm guessing that's not critical to this problem. We're using EEGLAB 11_0_5_4b. </span></div>
<div><span style="font-size: 10pt;"><br>
</span></div>
<div><span style="font-size: 10pt;">Has anyone else run into this and found a workaround? Or is there something else we should be doing to ensure that ICA runs, rather than PCA? (I'm also open to the suggestion that I'm completely misdiagnosing the problem!)</span></div>
<div><span style="font-size: 10pt;"><br>
</span></div>
<div><span style="font-size: 10pt;">I've pasted a portion of our ICA/PCA output below. (We're using binica in this example, but the same issue arises with the regular extended infomax algorithm.)</span></div>
<div><span style="font-size: 10pt;"><br>
</span></div>
<div><span style="font-size: 10pt;">Thanks!</span></div>
<div><span style="font-size: 10pt;">Pete</span></div>
<div><span style="font-family: 'Times New Roman';">---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------</span></div>
<div><span style="font-size: 10pt;">Warning: If the binary ICA function does not work, check that you have added the</span></div>
</span><span style="font-family: 'Times New Roman';">binary file location (in the EEGLAB directory) to your Unix /bin directory (.cshrc file)</span><br style="font-family: 'Times New Roman';">
<span style="font-family: 'Times New Roman';">Warning: fixing rank computation inconsistency (63 vs 64) most likely because running under Linux 64-bit MatlabData rank (64) is smaller than the number of channels (65).</span><br style="font-family: 'Times New Roman';">
<span style="font-family: 'Times New Roman';">binica: using source file '.../matlab/eeglab11_0_5_4b/functions/sigprocfunc/binica.sc'</span><br style="font-family: 'Times New Roman';">
<span style="font-family: 'Times New Roman';">binica(): using binary ica file '?.../matlab/eeglab11_0_5_4b/functions/resources/ica_linux'</span><br style="font-family: 'Times New Roman';">
<span style="font-family: 'Times New Roman';">binica(): processing 4 (flag, arg) pairs.</span><br style="font-family: 'Times New Roman';">
<span style="font-family: 'Times New Roman';"> setting lrate, 0.001</span><br style="font-family: 'Times New Roman';">
<span style="font-family: 'Times New Roman';"> setting pca, 64</span><br style="font-family: 'Times New Roman';">
<span style="font-family: 'Times New Roman';"> setting extended, 1</span><br style="font-family: 'Times New Roman';">
<span style="font-family: 'Times New Roman';">scriptfile = binica8147.sc</span><br style="font-family: 'Times New Roman';">
<br style="font-family: 'Times New Roman';">
<span style="font-family: 'Times New Roman';">Running ica from script file binica8147.sc</span><br style="font-family: 'Times New Roman';">
<span style="font-family: 'Times New Roman';"> Finding 64 components.</span><br style="font-family: 'Times New Roman';">
<span style="font-family: 'Times New Roman';">alias erplab '.../data/erplab': Command not found.</span><br style="font-family: 'Times New Roman';">
<br style="font-family: 'Times New Roman';">
<span style="font-family: 'Times New Roman';">ICA Version 1.4 (Feb. 14, 2002)</span><br style="font-family: 'Times New Roman';">
<br style="font-family: 'Times New Roman';">
<span style="font-family: 'Times New Roman';">Input data size [65,754696] = 65 channels, 754696 frames.</span><br style="font-family: 'Times New Roman';">
<span style="font-family: 'Times New Roman';">After PCA dimension reduction,</span><br style="font-family: 'Times New Roman';">
<span style="font-family: 'Times New Roman';"> finding 64 ICA components using extended ICA.</span><br style="font-family: 'Times New Roman';">
<span style="font-family: 'Times New Roman';">PDF will be calculated initially every 1 blocks using 6000 data points.</span><br style="font-family: 'Times New Roman';">
<span style="font-family: 'Times New Roman';">Initial learning rate will be 0.001, block size 501.</span><br style="font-family: 'Times New Roman';">
<span style="font-family: 'Times New Roman';">Learning rate will be multiplied by 0.98 whenever angledelta >= 60 deg.</span><br style="font-family: 'Times New Roman';">
<span style="font-family: 'Times New Roman';">Training will end when wchange < 1e-07 or after 512 steps.</span><br style="font-family: 'Times New Roman';">
<span style="font-family: 'Times New Roman';">Online bias adjustment will be used.</span><br style="font-family: 'Times New Roman';">
<div>
<div style="font-family: 'Times New Roman';">---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------</div>
</div>
<div><br>
</div>
<div><br>
<div style="font-family:Tahoma; font-size:13px">
<div><font size="2" face="Tahoma">
<div dir="ltr"><font size="1" face="Arial">
<div><font size="2" face="Tahoma">
<div dir="ltr"><font size="1" face="Arial"></font> </div>
<div dir="ltr"><font size="1" face="Arial">Peter Bachman, PhD</font></div>
<div dir="ltr"><font size="1" face="Arial"><span id="misspelled" tabindex="-1"><font size="1" face="Arial">Staglin</font></span>
<span id="misspelled" tabindex="-1">IMHRO</span> Center for Cognitive Neuroscience,
</font></div>
<div dir="ltr"><font size="1" face="Arial">Center for the Assessment and Prevention of
<span id="misspelled" tabindex="-1">Prodromal</span> States (CAPPS) </font></div>
<div dir="ltr"><font size="1" face="Arial">& Adolescent Brain and Behavior Research Clinic (<span id="misspelled" tabindex="-1">ABBRC</span>)</font></div>
<div dir="ltr"><font size="1" face="Arial"><span id="misspelled" tabindex="-1"><font size="1" face="Arial">Semel</font></span> Institute for Neuroscience and Human Behavior, UCLA</font></div>
<div dir="ltr"><font size="1" face="Arial">Office: (310) 206-4245</font></div>
<div dir="ltr"><font color="#000000" size="1" face="Arial"><span id="misspelled" tabindex="-1"><a href="mailto:bachman@psych.ucla.edu"><font color="#000000" size="1" face="Arial">bachman@psych.ucla.edu</font></a></span></font></div>
<div dir="ltr"><font size="1" face="arial"><span tabindex="-1"></span></font> </div>
<div dir="ltr"><font face="arial"><span tabindex="-1"><font size="1">
<div>
<hr>
</div>
</font><font color="#333333" size="1" face="Arial">IMPORTANT WARNING: This email (and any attachments) is only intended for the use of the person or entity to which it is addressed, and may contain information that is privileged and confidential. You, the recipient,
are obligated to maintain it in a safe, secure and confidential manner. Unauthorized redisclosure or failure to maintain confidentiality may subject you to federal and state penalties. If you are not the intended recipient, please immediately notify us by
return email, and delete this message from your computer.</font></span></font></div>
</font></div>
</font></div>
</font></div>
</div>
</div>
</div>
</body>
</html>