[FieldTrip] TFR Cluster permutation on Single-sensor
Tahereh Toosi
tahere.toosi at gmail.com
Fri Jun 5 18:52:52 CEST 2015
Hi fieldtripers,
I am trying to do a cluster permutation on Time-frequency data of a single
channel (of 64 channel EEG) in a between subject design based on the
example provided here
<http://www.fieldtriptoolbox.org/example/apply_clusterrandanalysis_on_tfrs_of_power_that_were_computed_with_besa>.
Two weird things happen.
First, although I set cfg.frequency = 'all' for stats, but it always
returns results for only ~1-10Hz instead of 1:15 Hz I have in my freq data
(to be precise : stat.cfg.frequency =[0.9995 9.9951]).
Second, based on the explanation in this paper
<http://www.sciencedirect.com/science/article/pii/S0165027007001707>, I
assume cfg.neighbourdist in single-sensor test somehow shows the number of
adjacent points, but when I change this number the clusters don't change at
all.
Here is my code, I would appreciate if anyone could tell me what is
happening.
Thanks,
Tahereh
% Time frequency analysis
cfg = [];
cfg.output = 'pow';
cfg.method = 'mtmconvol';
cfg.taper = 'hanning';
f1 = 1;f2 = 15; fstep = 0.1;
cfg.foi =f1:fstep:f2;
t1 = -1; t2 = 1;
cfg.toi = t1:0.02:t2;
n_cycles = 5;
cfg.t_ftimwin = n_cycles ./cfg.foi;
cfg.keeptrials ='no';
[freqdata] = ft_freqanalysis(cfg, data);
%store freq data for each condition/subject
TFDesc(Condition,subject) = {ft_freqdescriptives(cfg,freqdata)}
%Comparing Condition1 and Condition2
cfg = [];
Condition1_avg= ft_freqgrandaverage(cfg, TFDesc{cond1,: });
Condition2_avg = ft_freqgrandaverage(cfg, TFDesc{cond2,: });
% make a dummy structure with the difference between Condition 1 & 2
avgDiff = Condition1_avg;
avgDiff.powspctrm = Condition1_avg.powspctrm-Condition2_avg.powspctrm;
cfg = [];
cfg.channel = 'FP2'
cfg.frequency = 'all'
cfg.neighbourdist = 4;
cfg.latency = 'all'
cfg.avgovertime = 'no';
cfg.avgoverfreq ='no';
cfg.avgoverchan = 'no';
cfg.clusteralpha = 0.05;
cfg.statistic = 'ft_statfun_depsamplesT';
cfg.numrandomization = 500;
cfg.correctm = 'cluster';
cfg.method = 'montecarlo';
cfg.design = [
1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 % subject number
1 1 1 1 1 1 1 1 1 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
cfg
stat = ft_freqstatistics(cfg, TFDesc{cond1,selectedSubj},
TFDesc{cond2,selectedSubj});
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20150605/b0500926/attachment.html>
More information about the fieldtrip
mailing list