Non parametric test on coherence

[jan-mathijs schoffelen]

Hi Matteo,

As far as I know, the indepsamplesZcoh only works properly if you
compute your TFR using ft_freqanalysis with cfg.output = 'fourier'.
Note that the output of the statistics-function probably gives
differential coherence spectra between all pairs of channels. The
clustering will in this case not work, because the spatial channel
neighbourhood structure supported to my knowledge only works with
univariate test-statistics. This means that you shouldn't specify
cfg.correctm = 'cluster'.

Cheers,

Jan-Mathijs

On Apr 28, 2010, at 1:57 PM, Matteo Demuru wrote:

> Dear Eric,
>
> I have tried the between-trials experiment too, but the two problems
> still remain (the statfun_indepsamplesZcoh looks for label field.
> Furtheremore if I add it the reshape function crashes). Any other
> suggestions?
>
> I have also another question relative to your reply: the baseline
> and activation trials were already divided in the within-trials
> experiment, the only difference with the between-trials experiment
> are relative to the configuration parameters (i.e.  in between-
> trials only cfg.ivar is set while in within-trials cfg.ivar and
> cfg.uvar are set) am I wrong?
>
> Regarding the 'label field' problem, it seems a required field for
> the configuration struct because it is used in
> statfun_indepsamplesZcoh to calculate the channel combinations for
> the coherence.
>
> Thanks a lot
>
> Matteo
>
>
>
> On Wed, Apr 28, 2010 at 9:17 AM, Eric Maris <e.maris at donders.ru.nl>
> wrote:
> Dear Matteo,
>
>
>
> The statfun indepsamplesZcoh can only be used in a between-trials
> experiment. You could cut out your baseline and activation segments
> separately and then compare them in a non-paired fashion.
>
>
>
> Best,
>
>
> Eric
>
>
>
> From: FieldTrip discussion list [mailto:FIELDTRIP at NIC.SURFNET.NL] On
> Behalf Of Matteo Demuru
> Sent: dinsdag 27 april 2010 16:21
> To: FIELDTRIP at NIC.SURFNET.NL
> Subject: [FIELDTRIP] Non parametric test on coherence
>
>
> Dear all,
>
>
> I have a problem using freqstatistics to calculate significance for
> coherence.
>
>
> Specifically I have a subject and I am performing a within-trials
> experiment using 'indepsamplesZcoh' as statistic.
>
>
> I calculate the TFR of my data with the cfg.output='powandcsd' and
> cfg.keeptrials='yes' parameters. Then I call freqstatistics to
> compare my different experimental conditions (baseline vs activation).
>
>
> The function crashes with this output:
>
>
> ??? Reference to non-existent field 'label'.
>
>
> Error in ==> statfun_indepsamplesZcoh at 76
>
>   nchan = length(cfg.label);
>
>
> I have tried to add this field to the cfg struct assigning the cell
> that contains the interested channels. However this time I have
> another error:
>
>
> ??? Error using ==> reshape
>
> To RESHAPE the number of elements must not change.
>
>
> Error in ==> clusterstat at 178
>
>       posclusobs = findcluster(reshape(postailobs, [cfg.dim,
> 1]),channeighbstructmat,cfg.minnbchan);
>
>
>
> Any suggestions?
>
>
> Thanks in advance
>
>
> Matteo Demuru
>
>
> ----------------------------------
>
> 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.
>
> http://listserv.surfnet.nl/archives/fieldtrip.html
>
> http://www.ru.nl/fcdonders/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.
>
> http://listserv.surfnet.nl/archives/fieldtrip.html
>
> http://www.ru.nl/fcdonders/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.
>
> http://listserv.surfnet.nl/archives/fieldtrip.html
>
> http://www.ru.nl/fcdonders/fieldtrip/
>

Dr. J.M. (Jan-Mathijs) Schoffelen
Donders Institute for Brain, Cognition and Behaviour,
Centre for Cognitive Neuroimaging,
Radboud University Nijmegen, The Netherlands
J.Schoffelen at donders.ru.nl
Telephone: 0031-24-3668063


----------------------------------
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 --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20100428/2d911a6b/attachment-0002.html>