Update: Freqstatistics Now Yields Significant Clusters

Tue Jun 16 22:44:44 CEST 2009

Hi all,

I just wanted to post a follow-up to my previous posted problem of being
unable to obtain significant clusters using freqstats. I've included my code
for what I believe (and hope...) is our success at obtaining clusters in our
analysis, and acknowledge the help we received from the FieldTrip board
users, in particular Michael and Eric whose suggestions were immensely
helpful. Thanks again!

Here's our code below. Our study has both between (sex) and within group
components (task1 vs. task2). 

=====================================================================

% this is the list of BESA datafiles in the Female Feature condition
filename_femfeat = {
.
.
.
};

nsubj = length(filename_femfeat);

for i=1:nsubj
  femfeat{i} = besa2fieldtrip(filename_femfeat{i});
end

% this is the list of BESA datafiles in the Female Location condition
filename_femloc = {
.
.
.
};

for i=11
nsubj = length(filename_femloc);
end

for i=1:nsubj
  femloc{i} = besa2fieldtrip(filename_femloc{i});
end

% this is the list of BESA datafiles in the Male Feature condition
filename_malefeat = {
.
.
.
};
for i=11
nsubj = length(filename_malefeat);
end

% this is the list of BESA datafiles in the Male Location condition
filename_maleloc = {
.
.
.
};
for i=11
nsubj = length(filename_maleloc);
end
%}

% collect all single subject data in a convenient cell-array
for i=1:nsubj
  femfeat{i} = besa2fieldtrip(filename_femfeat{i});
  femloc{i} = besa2fieldtrip(filename_femloc{i});
  malefeat{i} = besa2fieldtrip(filename_malefeat{i});
  maleloc{i} = besa2fieldtrip(filename_maleloc{i});
end

%Read in the electrode locations for the Std81 montage
cfg = [];
elec = read_fcdc_elec('EGI-BESA_Standard_81_prime.sfp');
elec.pnt = 1000*elec.pnt;

% recompute the average, except do _not_ average but keepindividual
% this collects all identical time/frequency/channel samples over all
% subjects into a single data structure
cfg = [];
cfg.keepindividual = 'yes';
maleloc_all = freqgrandaverage(cfg, maleloc{:});
femloc_all = freqgrandaverage(cfg, femloc{:});
malefeat_all = freqgrandaverage(cfg, malefeat{:});
femfeat_all = freqgrandaverage(cfg, femfeat{:});

% perform the statistical test using randomization and a clustering approach
% using the NEW freqstatistics function
cfg = [];
cfg.neighbourdist    = 45;
cfg.elec             = elec;
cfg.statistic        = 'indepsamplesT';
cfg.minnbchan        = 0;
cfg.clusteralpha     = 0.05;
cfg.alpha            = 0.05;
cfg.clustertail      = 0;
cfg.numrandomization = 5000;

cfg.latency          = [0 1000];
cfg.frequency        = [4 7];
cfg.avgovertime      = 'no'; 
cfg.avgoverfreq      = 'yes';
cfg.avgoverchan      = 'no';

cfg.correctm         = 'cluster';
cfg.method           = 'montecarlo';
cfg.feedback         = 'gui';
cfg.design           = [1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
21 22;   % subject number
                        1 1 1 1 1 1 1 1 1 1  1  2  2  2  2  2  2  2  2  2  2
 2]; % condition number

%cfg.design           = [1 2 3 4 5 6 7 8 9 10 11 1  2  3  4  5  6  7  8  9 
10 11;   % subject number
                        %1 1 1 1 1 1 1 1 1 1  1  2  2  2  2  2  2  2  2  2 
2  2]; % condition number

cfg.uvar = 1;                                   % "subject" is unit of
observation
cfg.ivar = 2;                                   % "condition" is the
independent variable
stat = freqstatistics(cfg, maleloc_all, femloc_all);
%stat = freqstatistics(cfg, femfeat_all, malefeat_all);
%stat = freqstatistics(cfg, maleloc_all, malefeat_all);
%stat = freqstatistics(cfg, femfeat_all, femloc_all);

cfg = [];
cfg.elec  = elec;
cfg.rotate = 0;
%cfg.zlim = [-6 6];
cfg.alpha = 0.05;
cfg.label = stat.label;
cfg.electrodes = 'labels';
cfg.showxlim = 'yes';
cfg.showzlim = 'yes';
cfg.showylim = 'yes';
clusterplot (cfg, stat);
=====================================================================

Our results indicated the following:
.
.
.
reading time-frequency representation using BESA toolbox
reading power on 81 channels
not computing grand average, but keeping individual power for 11 subjects
not computing grand average, but keeping individual power for 11 subjects
not computing grand average, but keeping individual power for 11 subjects
not computing grand average, but keeping individual power for 11 subjects

selected 81 channels
selected 21 time bins
averaging over 4 frequency bins

Warning: PACK can only be used from the MATLAB command line.
> In fieldtrip\private\prepare_timefreq_data at 310
  In fieldtrip\private\statistics_wrapper at 206
  In freqstatistics at 132
  In CMCWM2_std81 at 139

Obtaining the electrode configuration from the configuration.
there are on average 4.7 neighbours per channel
using "statistics_montecarlo" for the statistical testing
using "statfun_indepsamplesT" for the single-sample statistics
constructing randomized design
total number of measurements     = 22
total number of variables        = 2
number of independent variables  = 1
number of unit variables         = 1
number of within-cell variables = 0
number of control variables      = 0
using a permutation resampling approach
repeated measurement in variable 1 over 22 levels
number of repeated measurements in each level is 1 1 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 
computing a parmetric threshold for clustering
estimated time per randomization is 0 seconds
found 4 positive clusters in observed data
found 7 negative clusters in observed data
using a cluster-based method for multiple comparison correction
the returned probabilities and the thresholded mask are corrected for
multiple comparisons

There are 2 clusters smaller than alpha (0.05)
Positive cluster: 1, pvalue: 0 (*), t = 100 to 450
Negative cluster: 1, pvalue: 0 (*), t = 0 to 100
creating layout from cfg.elec
creating layout from cfg.elec
creating layout from cfg.elec
creating layout from cfg.elec
creating layout from cfg.elec
creating layout from cfg.elec
creating layout from cfg.elec
creating layout from cfg.elec
creating layout from cfg.elec
creating layout from cfg.elec

=====================================================================

I've also attached a jpg of the above result and our .sfp file if anyone is
interested in further information. Thanks once again for the help!

Cheers,

Charles Cook

----------------------------------
The aim of this list is to facilitate the discussion between users of the FieldTrip  toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/neuroimaging/fieldtrip.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 4-7Hz.jpg
Type: image/jpeg
Size: 193169 bytes
Desc: not available
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20090616/e9534ceb/attachment-0001.jpg>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: EGI-BESA_Standard_81_prime.sfp
Type: application/octet-stream
Size: 3513 bytes
Desc: not available
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20090616/e9534ceb/attachment-0001.obj>