Differences

This shows you the differences between two versions of the page.

Link to this comparison view

tutorial:cluster_permutation_timelock [2012/07/21 16:03]
89.156.125.19 [Plotting the results]
tutorial:cluster_permutation_timelock [2013/03/29 11:33] (current)
jan-mathijs
Line 111: Line 111:
 We now briefly discuss the configuration fields that are not specific for **[[reference:ft_timelockstatistics|ft_timelockstatistics]]**:  We now briefly discuss the configuration fields that are not specific for **[[reference:ft_timelockstatistics|ft_timelockstatistics]]**: 
  
-  cfg.channel = {'MEG'};          % cell-array with selected channel labels +  cfg_neighb        = [];
-  cfg.latency = [0 1];            % time interval over which the experimental  +
-                                  % conditions must be compared (in seconds)+
   cfg_neighb.method = 'distance';            cfg_neighb.method = 'distance';         
-  neighbours = ...                specifies with which sensors other sensors +  neighbours        ft_prepare_neighbours(cfg_neighb, dataFC_LP); 
-  ft_prepare_neighbours(...       % can form clusters +   
-  cfg.neighb, dataFC_LP)+  cfg.neighbours    = neighbours;  the neighbours specify for each sensor with  
-  cfg.neighbours neighbours+                                   % which other sensors it can form clusters 
 +  cfg.channel       = {'MEG'}    % cell-array with selected channel labels 
 +  cfg.latency       [0 1]      % time interval over which the experimental  
 +                                   % conditions must be compared (in seconds) 
 +  
  
 With these two options, we select the spatio-temporal dataset involving all MEG channels and the time interval between 0 and 1 second. The two experimental conditions will only be compared on this selection of the complete spatio-temporal dataset. Also, feel free to consult  With these two options, we select the spatio-temporal dataset involving all MEG channels and the time interval between 0 and 1 second. The two experimental conditions will only be compared on this selection of the complete spatio-temporal dataset. Also, feel free to consult 
Line 163: Line 164:
     clusterstat: -1.0251e+04     clusterstat: -1.0251e+04
  
-It is possible that the p-values in your output are a little bit different from 0. This is because **[[reference:ft_timelockstatistics|ft_timelockstatistics]]** calculated as a Monte Carlo approximation of the permutation p-values: the p-value for the k-th positive cluster is calculated as the proportion of random draws from the permutation distribution in which the maximum of the cluster-level statistics is more larger than stat.posclusters(k).clusterstat. +It is possible that the p-values in your output are a little bit different from 0. This is because **[[reference:ft_timelockstatistics|ft_timelockstatistics]]** calculated as a Monte Carlo approximation of the permutation p-values: the p-value for the k-th positive cluster is calculated as the proportion of random draws from the permutation distribution in which the maximum of the cluster-level statistics is larger than stat.posclusters(k).clusterstat. 
  
 ==== Plotting the results ==== ==== Plotting the results ====
Line 176: Line 177:
   raweffectFICvsFC     = avgFIC;   raweffectFICvsFC     = avgFIC;
   % Then take the difference of the averages.   % Then take the difference of the averages.
-  raweffectFICvsFC.avg = avg_FIC.avg - avg_FC.avg;  +  raweffectFICvsFC.avg = avgFIC.avg - avgFC.avg;  
  
 We then construct a boolean matrix indicating membership in the significant clusters.  This matrix has size [Number_of_MEG_channels × Number_of_temporal_samples], like stat.posclusterslabelmat.  We'll make two such matrices: one for positive clusters (named pos), and one for negative (neg).  All (channel,time)-pairs belonging to the significant clusters will be coded in the new boolean matrix as 1, and all those that don't will be coded as 0. We then construct a boolean matrix indicating membership in the significant clusters.  This matrix has size [Number_of_MEG_channels × Number_of_temporal_samples], like stat.posclusterslabelmat.  We'll make two such matrices: one for positive clusters (named pos), and one for negative (neg).  All (channel,time)-pairs belonging to the significant clusters will be coded in the new boolean matrix as 1, and all those that don't will be coded as 0.
Line 469: Line 470:
  
 ----- -----
-This tutorial was last tested by Jörn, with version 20110507 of FieldTrip, using Matlab 2009b on a Windows 7 platform.+This tutorial was last tested by Jan-Mathijs, with version 20130301 of FieldTrip, using Matlab 2011a on a Windows 7 platform.
tutorial/cluster_permutation_timelock.1342879395.txt.gz · Last modified: 2012/07/21 16:03 by 89.156.125.19

You are here: starttutorialcluster_permutation_timelock
This DokuWiki features an Anymorphic Webdesign theme, customised by Eelke Spaak and Stephen Whitmarsh.
www.chimeric.de Valid CSS Driven by DokuWiki do yourself a favour and use a real browser - get firefox!! Recent changes RSS feed Valid XHTML 1.0