<table cellspacing="0" cellpadding="0" border="0" ><tr><td valign="top" style="font: inherit;"><DIV>Hi Katerina,</DIV>
<DIV> </DIV>
<DIV> </DIV>
<DIV> </DIV>
<DIV> </DIV>
<DIV> </DIV>
<DIV> </DIV>
<DIV> </DIV>
<DIV>I agree with explanation written by Dr.Scott Makeig. </DIV>
<DIV>I would like to add a few points.</DIV>
<DIV> </DIV>
<DIV> </DIV>
<DIV>Motor imagery tasks are occurring in mu-rhythm (8-12 Hz) and there is decrease during<BR>hand or finger movement or in preparation for movement (event-related desynchronization, ERD) and increase after movement and in relaxation (event-related synchronization, ERS). <BR>So both ERD and ERS are ERP events. </DIV>
<DIV>After Tools>Run ICA, you can do Plot>Channel ERPs,ERP map series,Sum/Compare ERPs and also Plot>Component ERPs.</DIV>
<DIV> </DIV>
<DIV> </DIV>
<DIV> </DIV>
<DIV> </DIV>
<DIV> </DIV>
<DIV> </DIV>
<DIV><BR>regards,</DIV>
<DIV>I. Jaganjac</DIV>
<DIV> </DIV>
<DIV> </DIV>
<DIV> </DIV>
<DIV> </DIV>
<DIV> </DIV>
<DIV> </DIV>
<DIV> </DIV>
<DIV> </DIV>
<DIV> </DIV>
<DIV> </DIV>
<DIV> </DIV>
<DIV> </DIV>
<DIV> </DIV>
<DIV><BR>Content-Transfer-Encoding: 7bit<BR>From: Katerina <<A href="mailto:jeux_d@yahoo.gr">jeux_d@yahoo.gr</A>><BR>Precedence: list<BR>MIME-Version: 1.0<BR>To: <A href="mailto:eeglablist@sccn.ucsd.edu">eeglablist@sccn.ucsd.edu</A><BR>Date: Tue, 24 Nov 2009 14:44:32 +0200<BR>Reply-To: <A href="mailto:jeux_d@yahoo.gr">jeux_d@yahoo.gr</A><BR>Message-ID: <<A href="mailto:4B0BD530.9080605@yahoo.gr">4B0BD530.9080605@yahoo.gr</A>><BR>Content-Type: text/plain; <SPAN>charset</SPAN>=ISO-8859-7; format=flowed<BR>Subject: [Eeglablist] ICA analysis for mu-rhythm detection<BR>Message: 1</DIV>
<DIV>Hi,<BR>i have performed ICA analysis through EEGLAB (Tools -> Run ICA) to data <BR>of three different experiments. All subjects had performed hand or <BR>finger movement. However, after performing ICA and ploting (Tools -> <BR>Component spectra and maps) at 10Hz, the topographic maps didn't show <BR>the correct activation, i.e. activation at the motor cortex of the <BR>contralateral hemisphere. As for the third experiment in particular, the <BR>subject had been repeatedly tapping the finger of the right hand, and <BR>after ICA analysis, the map only showed activation at the occipital lobe.</DIV>
<DIV>Is there any explanation for this behavior of the data? Is there any <BR>particular reason why ICA wouldn't work with mu-rhythm detection and <BR>hand movement?</DIV>
<DIV><BR><BR> </DIV></td></tr></table><br>