<div dir="ltr"><div class="gmail_default" style="font-family:tahoma,sans-serif"><div class="gmail_default" style="font-size:13px"><span style="background-color:rgb(255,255,255)"><font color="#000000" style="color:rgb(34,34,34)">Dear </font><font>FieldTrip list,</font></span></div><div class="gmail_default" style="color:rgb(34,34,34);font-size:13px"><font color="#000000" style="background-color:rgb(255,255,255)"><br></font></div><div class="gmail_default" style="color:rgb(34,34,34);font-size:13px"><span style="background-color:rgb(255,255,255)"><font color="#000000">I'm planning to apply Granger Causality (GC) models to my </font><span style="color:rgb(0,0,0)">continuous</span><span style="color:rgb(0,0,0)">  </span><span style="color:rgb(0,0,0)">EEG data, but  I'm not  sure about  the preprocessing steps. </span></span></div><div class="gmail_default" style="color:rgb(34,34,34);font-size:13px"><font color="#000000" style="background-color:rgb(255,255,255)">I have read an  <a href="http://www.sciencedirect.com/science/article/pii/S0165027011004687" target="_blank">article</a> about filtering EEG data before <span class="" style="color:rgb(34,34,34);background-image:initial;background-repeat:initial">GC</span> analysis, but still it is not clear for me how to deal with removing some data segments with strong <span style="background-image:initial;background-repeat:initial"><span class="" style="color:rgb(34,34,34);background-image:initial;background-repeat:initial">artifacts</span></span>.</font></div><div class="gmail_default" style="color:rgb(34,34,34);font-size:13px"><font color="#000000" style="background-color:rgb(255,255,255)">Right now I'm planning to clean the data (2-minutes trials) by removing the <span style="background-image:initial;background-repeat:initial"><span class="" style="color:rgb(34,34,34);background-image:initial;background-repeat:initial">artifacts</span></span> using visual detection (+filtering notch+detrending)  followed by <span class="" style="color:rgb(34,34,34);background-image:initial;background-repeat:initial">GC</span> analysis using  2 sec. windows (actually the final length of the windows should be identified by VAR model) .</font></div><div class="gmail_default" style="color:rgb(34,34,34);font-size:13px"><font color="#000000" style="background-color:rgb(255,255,255)">Using such preprocessing approach I would receive continuous data with "gaps" therefore some 2-sec segments may consist of the mixed data before and after the removed part. Such situations seem to me not appropriate  for <span class="" style="color:rgb(34,34,34);background-image:initial;background-repeat:initial">GC</span> analysis.</font></div><div class="gmail_default" style="color:rgb(34,34,34);font-size:13px"><font color="#000000" style="background-color:rgb(255,255,255)">As a possible solutions I thought about 1) removing these mixed segments or</font></div><div class="gmail_default" style="color:rgb(34,34,34);font-size:13px"><font color="#000000" style="background-color:rgb(255,255,255)">2) Dividing raw data (before <span style="background-image:initial;background-repeat:initial"><span class="" style="color:rgb(34,34,34);background-image:initial;background-repeat:initial">artifacts</span></span> rejection) on 2-sec segments and to remove completely such segments  that will contain an ar<span style="background-image:initial;background-repeat:initial">tifact</span>.  Do I miss something?</font></div><div class="gmail_default" style="color:rgb(34,34,34);font-size:13px"><font color="#000000" style="background-color:rgb(255,255,255)"><br></font></div><div class="gmail_default" style="color:rgb(34,34,34);font-size:13px"><font color="#000000" style="background-color:rgb(255,255,255)">I would appreciate your advice on this matter.</font></div><div class="gmail_default" style="color:rgb(34,34,34);font-size:13px"><font color="#000000" style="background-color:rgb(255,255,255)"><br></font></div><div class="gmail_default" style="color:rgb(34,34,34);font-size:13px"><font color="#000000" style="background-color:rgb(255,255,255)">Thanks,</font></div><div class="gmail_default" style="color:rgb(34,34,34);font-size:13px"><font color="#000000" style="background-color:rgb(255,255,255)"><br></font></div><div class="gmail_default" style="color:rgb(34,34,34);font-size:13px"><font color="#000000" style="background-color:rgb(255,255,255)">Pavel</font></div></div></div>