<div dir="ltr">Dear all<div><br></div><div><br></div><div>My name is Toni and I am working on EEG signal processing at University of Deusto, Bilbao, Spain. More specifically, right now trying to apply cluster-based permutation testing to determine whether there exist differences among 3 groups of subjects.</div><div><br></div><div><br></div><div>So, for my testing, subject groups are E, EN and N, the 2 former being different types of patients and the latter being a control group. I am forming clusters based on proximity (in space) and each subject is characterized by a matrix of values that vary in 2D (per channel and another variable related but not equal to frequency).</div><div><br></div><div>By means of a T-statistic (indepsamplesT) I can find differences in between E and EN, but I find no clusters/differences in between E and N or EN and N, even though hypothetically (I think) they should exist. By means of an F-statistic (indepsamplesF) I can find differences among the 3 groups at the same time, then also between E and EN and no differences between E and N or EN and N...</div><div><br></div><div>Am I missing something? I do not understand why the might exist a significant cluster when comparing all 3 groups that does not exist in the 1vs1 comparisons, even though the cluster is not the same I obtain when I compare E vs EN...</div><div><br></div><div><br></div><div>For additional information, I attach how I run the test:</div><div><br></div><div><div><font face="monospace, monospace">cfg.method = 'distance';</font></div><div><font face="monospace, monospace">cfg.neighbourdist = 1;</font></div><div><font face="monospace, monospace">cfg.elec = ft_datatype_sens(subjEN{1}.elec);</font></div><div><font face="monospace, monospace">neighbours = ft_prepare_neighbours(cfg);</font></div><div><font face="monospace, monospace"><br></font></div><div><font face="monospace, monospace">cfg = [];</font></div><div><font face="monospace, monospace">cfg.channel = {'EEG'};</font></div><div><font face="monospace, monospace">cfg.latency = 'all';</font></div><div><font face="monospace, monospace">cfg.method = 'montecarlo';</font></div><div><font face="monospace, monospace">cfg.statistic = 'indepsamplesT';</font></div><div><font face="monospace, monospace">cfg.correctm = 'cluster';</font></div><div><font face="monospace, monospace">cfg.clusteralpha = 0.05;</font></div><div><font face="monospace, monospace">cfg.clusterstatistic = 'maxsum';</font></div><div><font face="monospace, monospace"><br></font></div><div><font face="monospace, monospace">cfg.minnbchan = 2;</font></div><div><font face="monospace, monospace"><br></font></div><div><font face="monospace, monospace">cfg.neighbours = neighbours; % same as defined for the between-trials experiment</font></div><div><font face="monospace, monospace">cfg.tail = 0;</font></div><div><font face="monospace, monospace">cfg.clustertail = 0;</font></div><div><font face="monospace, monospace">cfg.alpha = 0.025;</font></div><div><font face="monospace, monospace">cfg.numrandomization = 5000;</font></div><div><font face="monospace, monospace"><br></font></div><div><font face="monospace, monospace">nsubj = size(subjE,2)+size(subjEN,2);</font></div><div><font face="monospace, monospace">design = zeros(2,nsubj);</font></div><div><span style="font-family:monospace,monospace">design(1,:) = [1:nsubj/2 1:nsubj/2];</span><br></div><div><font face="monospace, monospace">design(2,1:size(subjE,2)) = 1;</font></div><div><font face="monospace, monospace">design(2,size(subjE,2)+1:nsubj) = 2;</font></div><div><font face="monospace, monospace"><br></font></div><div><font face="monospace, monospace">cfg.design = design;</font></div><div><font face="monospace, monospace">cfg.ivar = 2;</font></div><div><br></div><div><font face="monospace, monospace">[stat] = ft_timelockstatistics(cfg, subjEN{:}, subjE{:});</font></div></div><div><font face="arial, helvetica, sans-serif"><br></font></div><div><font face="arial, helvetica, sans-serif"><br></font></div><div><font face="arial, helvetica, sans-serif">When I compare the 3 groups</font></div><div><font face="arial, helvetica, sans-serif"><br></font></div><div><div><font face="monospace, monospace">cfg.method = 'distance';</font></div><div><font face="monospace, monospace">cfg.neighbourdist = 1;</font></div><div><font face="monospace, monospace">cfg.elec = ft_datatype_sens(subjEN{1}.elec);</font></div><div><font face="monospace, monospace">neighbours = ft_prepare_neighbours(cfg);</font></div><div><font face="monospace, monospace"><br></font></div><div><font face="monospace, monospace">cfg = [];</font></div><div><font face="monospace, monospace">cfg.channel = {'EEG'};</font></div><div><font face="monospace, monospace">cfg.latency = 'all';</font></div><div><font face="monospace, monospace">cfg.method = 'montecarlo';</font></div><div><font face="monospace, monospace">cfg.statistic = 'indepsamplesF'; %F-statistic</font></div><div><font face="monospace, monospace">cfg.correctm = 'cluster';</font></div><div><font face="monospace, monospace">cfg.clusteralpha = 0.05;</font></div><div><font face="monospace, monospace">cfg.clusterstatistic = 'maxsum';</font></div><div><font face="monospace, monospace"><br></font></div><div><font face="monospace, monospace">cfg.minnbchan = 2;</font></div><div><font face="monospace, monospace"><br></font></div><div><font face="monospace, monospace">cfg.neighbours = neighbours; % same as defined for the between-trials experiment</font></div><div><font face="monospace, monospace">cfg.tail = 1; %One sided</font></div><div><font face="monospace, monospace">cfg.clustertail = 1;</font></div><div><font face="monospace, monospace">cfg.alpha = 0.025;</font></div><div><font face="monospace, monospace">cfg.numrandomization = 5000;</font></div><div><font face="monospace, monospace"><br></font></div><div><font face="monospace, monospace">nsubj1 = size(subjE,2);</font></div><div><font face="monospace, monospace">nsubj2 = size(subjEN,2);</font></div><div><font face="monospace, monospace">nsubj3 = size(subjN,2);</font></div><div><font face="monospace, monospace">nsubj = nsubj1+nsubj2+nsubj3;</font></div><div><font face="monospace, monospace"><br></font></div><div><font face="monospace, monospace">design = zeros(2,nsubj);</font></div><div><font face="monospace, monospace">design(1,:) = [1:nsubj1 1:nsubj2 1:nsubj3];</font></div><div><font face="monospace, monospace">design(2,1:nsubj1) = 1;</font></div><div><font face="monospace, monospace">design(2,nsubj1+1:nsubj1+nsubj2) = 2;</font></div><div><font face="monospace, monospace">design(2,nsubj1+nsubj2+1:nsubj) = 3;</font></div><div><font face="monospace, monospace"><br></font></div><div><font face="monospace, monospace">cfg.design = design;</font></div><div><font face="monospace, monospace">cfg.ivar = 2; %2nd row is independent var</font></div><div><font face="monospace, monospace"><br></font></div><div><font face="monospace, monospace">[stat] = ft_timelockstatistics(cfg, subjE{:}, subjEN{:}, subjN{:});</font></div></div><div><br></div><div><div><br></div><div>Then, subjE, subjN and subjEN are like this:</div><div><br></div><div><div><font face="monospace, monospace">Eall = subjE{:}</font></div><div><font face="monospace, monospace"><br></font></div><div><font face="monospace, monospace">Eall = </font></div><div><font face="monospace, monospace"><br></font></div><div><font face="monospace, monospace"> struct with fields:</font></div><div><font face="monospace, monospace"><br></font></div><div><font face="monospace, monospace"> label: {1×18 cell}</font></div><div><font face="monospace, monospace"> fsample: 200</font></div><div><font face="monospace, monospace"> elec: [1×1 struct]</font></div><div><font face="monospace, monospace"> trial: {[18×115 double]}</font></div><div><font face="monospace, monospace"> time: {[1×115 double]}</font></div><div><font face="monospace, monospace"> cfg: [1×1 struct]</font></div></div><div><br></div><div><br></div><div>Thanks in advance</div><div><br></div>-- <br><div class="gmail_signature"><div dir="ltr"><div><div dir="ltr"><div style="color:rgb(136,136,136);font-size:12.8px"><b style="font-size:12.8px;color:rgb(34,34,34)"><font color="#000000">Jose Antonio Urigüen</font></b><br></div><div style="font-size:12.8px"><table border="0" cellpadding="0" cellspacing="0" style="color:rgb(0,0,0);font-size:medium;border-collapse:collapse"><tbody><tr><td style="font-size:13px"><span style="font-size:12.8px">PostDoc at Deustotech-LIFE (eVIDA Research Group)</span><font face="arial, helvetica, sans-serif"><br><a href="http://www.deusto.es/" target="_blank"><img src="http://www.deusto.es/img/deusto.jpg" alt="Deusto" border="0" style="padding: 15px 0px; width: 70px;"></a><br>Universidad de Deusto / Deustuko Unibertsitatea / </font><span style="font-size:12.8px">University of Deusto<br></span><span style="font-size:12.8px">Facultad de Ingeniería, 4ª Planta</span><span style="font-size:12.8px"><br></span></td></tr><tr><td style="padding-top:0px"><table border="0" cellpadding="0" cellspacing="0" style="border-collapse:collapse"><tbody><tr><td><font face="arial, helvetica, sans-serif" style="font-size:13px">Avda. Universidades 24. 48007 Bilbao<br>Tel. +34 94 413 90 00 / </font><span style="font-size:12.8px">Mov. </span><span style="font-size:12.8px">+34 656 711 643</span><font face="arial, helvetica, sans-serif" style="font-size:13px"><br>Ext. 2980</font><font><br><a href="mailto:sanchez.alain@deusto.es" style="font-family:arial,helvetica,sans-serif;font-size:13px" target="_blank">jose.uriguen@deusto.es</a><br></font><span style="font-size:12.8px">Web:</span><font style="font-size:12.8px"> <a href="http://evida.deusto.es/" target="_blank">evida.deusto.es</a></font><font face="arial, helvetica, sans-serif" style="font-size:13px"><br></font></td></tr></tbody></table></td></tr><tr><td style="padding-top:10px"><table border="0" cellpadding="0" cellspacing="0" style="border-collapse:collapse"><tbody><tr><td valign="center" style="vertical-align:middle"><font face="arial, helvetica, sans-serif"><b style="font-size:15px"><a href="http://www.deusto.es/" style="text-decoration:none" target="_blank">www.deusto.es</a></b></font></td></tr></tbody></table></td></tr></tbody></table></div></div></div></div></div>
</div></div>