Freqstatistics Yields Zero Significant Clusters?
Michael Wibral
wibral at BIC.UNI-FRANKFURT.DE
Fri Jun 5 10:58:47 CEST 2009
Hi Charles,
from your output:
...
computing statistic 100 from 100
performing Bonferoni correction for multiple comparisons
...
it seems that you're only computing 100 randomizations. It follows that the best p-value you could EVER get is 0.01. You then do bonferroni correction (not the cluster based correction you intended!). So if you set an alpha of 0.9 and divide this by - say - 2000 for your bonferoni corrcetion you alpha is 0.9/2000=0.00045. You see that you'll never reach this limit given that you do only 100 randomizations and by defibition cannot get below p=0.01. In addition you won't get clusters when you use Bonferroni.
I suggest using:
cfg.numrandomization = 5000;
cfg.corectm='cluster'; % or 'fdr'
Good Luck!
Michael
> -----Ursprüngliche Nachricht-----
> Von: "Charles Cook" <charles.cook at ULETH.CA>
> Gesendet: 04.06.09 18:07:02
> An: FIELDTRIP at NIC.SURFNET.NL
> Betreff: [FIELDTRIP] Freqstatistics Yields Zero Significant Clusters?
> I've been having trouble still trying to perform cluster-based permutation
> tests with Fieldtrip. What I'm attempting to do is compare male and female
> participant's time-frequency data (between group or independent samples)
> generated from BESA on a spatial memory task. We've been trying to
> increasing the alpha levels to determine if we have any significant
> clusters, and even moving it up to 0.9 still does not provide any
> significance.
>
> Any suggestions would be much appreciated.
>
> Cheers,
>
> Charles
> --------------------------------
>
> % this is the list of BESA datafiles in the Female Location condition
> filename_femloc = {
> .
> .
> };
> for i=1:11
> femloc{i} = besa2fieldtrip(filename_femloc{i});
> end
>
> % this is the list of BESA datafiles in the Male Location condition
> filename_maleloc = {
> .
> .
> };
> for i=11
> maleloc{i} = besa2fieldtrip(filename_maleloc{i});
> end
> %}
>
> % collect all single subject data in a convenient cell-array
> for i=1:11
> femloc{i} = besa2fieldtrip(filename_femloc{i});
> maleloc{i} = besa2fieldtrip(filename_maleloc{i});
> end
>
>
> %Reading in the electrode locations for the Std.81 montage
> elec = read_fcdc_elec('EGI-BESA_Standard_81.sfp');
>
> % 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{:});
>
>
> % perform the statistical test using randomization and a clustering approach
> % using the NEW freqstatistics function
> cfg = [];
> cfg.elec = elec;
> cfg.neighbourdist = 4;
> cfg.statistic = 'indepsamplesT';
> cfg.minnbchan = 0;
> cfg.clusteralpha = 0.05;
> cfg.clustertail = 0;
> crg.makeclusters = 'yes';
> cfg.numrandomization = 100;
> cfg.latency = 'all';
> cfg.frequency = 'all';
> cfg.avgovertime = 'no';
> cfg.avgoverfreq = 'no';
> cfg.avgoverchan = 'no';
> cfg.correctm = 'bonferoni';
> cfg.method = 'montecarlo';
> 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 is 1-11 males and 1-11 females
> 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2]; %
> condition number with 1 being males and 2 being females
>
> cfg.uvar = 1; % "subject" is unit of
> observation
> cfg.ivar = 2; % "condition" is the
> independent variable
> [stat] = freqstatistics(cfg, maleloc_all, femloc_all);
>
> cfg = [];
> [freq_maleloc] = freqdescriptives(cfg, maleloc_all);
> [freq_femloc] = freqdescriptives(cfg, femloc_all);
>
> cfg = [];
> cfg.zlim = [-6 6];
> cfg.alpha = 0.025;
> clusterplot(cfg, stat);
>
> --------------------------------
> 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
> selected 81 channels
> selected 31 time bins
> selected 79 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 194
> 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 11 levels
> number of repeated measurements in each level is 2 2 2 2 2 2 2 2 2 2 2
> computing statistic
> estimated time per randomization is 1 seconds
> computing statistic 1 from 100
> .
> .
> computing statistic 100 from 100
> performing Bonferoni correction for multiple comparisons
> the returned probabilities are uncorrected, the thresholded mask is corrected
> the input is freq data with 81 channels, 79 frequencybins and 31 timebins
>
> computing the leave-one-out averages [---| ]
> computing the leave-one-out averages [-------/ ]
> computing the leave-one-out averages [----------- ]
> computing the leave-one-out averages [-------------\ ]
> computing the leave-one-out averages [----------------| ]
> computing the leave-one-out averages [--------------------/ ]
> computing the leave-one-out averages [------------------------ ]
> computing the leave-one-out averages [--------------------------\ ]
> computing the leave-one-out averages [-----------------------------| ]
> computing the leave-one-out averages [---------------------------------/ ]
> computing the leave-one-out averages [-------------------------------------]
> the input is freq data with 81 channels, 79 frequencybins and 31 timebins
>
> computing the leave-one-out averages [---| ]
> computing the leave-one-out averages [-------/ ]
> computing the leave-one-out averages [----------- ]
> computing the leave-one-out averages [-------------\ ]
> computing the leave-one-out averages [----------------| ]
> computing the leave-one-out averages [--------------------/ ]
> computing the leave-one-out averages [------------------------ ]
> computing the leave-one-out averages [--------------------------\ ]
> computing the leave-one-out averages [-----------------------------| ]
> computing the leave-one-out averages [---------------------------------/ ]
> computing the leave-one-out averages [-------------------------------------]
> no significant clusters in data; nothing to plot
> >>
>
> ----------------------------------
> 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.
>
----------------------------------
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: Michael Wibral.vcf
Type: text/x-vcard
Size: 344 bytes
Desc: not available
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20090605/6d431982/attachment.vcf>
More information about the fieldtrip
mailing list