[FieldTrip] Only 1 and Inf as output of ft_connectivityanalysis

[jan-mathijs schoffelen]

Hi Lucia,

The warnings indicate that you are dividing by zero. Are you sure that your cfg is correct?

Please try

cfg = [];
cfg.method = 'coh';
coh = ft_connectivityanalysis(cfg, freq);

This should work.

Best,
Jan-Mathijs

On May 2, 2013, at 6:56 PM, Lucia Melloni wrote:

> Hi jan-mathijs
> 
> Unfortunately I get the same result whether or not I remove the mean. 
> Best,
> -- Lucía
> 
> On May 2, 2013, at 9:57 AM, jan-mathijs schoffelen <jan.schoffelen at donders.ru.nl> wrote:
> 
>> Hi Lucia,
>> 
>> Could you remove cfg.removemean = 'yes' and try again?
>> 
>> Best,
>> 
>> Jan-Mathijs
>> 
>> On Apr 29, 2013, at 2:38 AM, Lucia Melloni wrote:
>> 
>>> Hi!
>>> I am trying to calculate coherence using ft_connectivityanalysis, but
>>> I am getting only ones and infinity as output in cohspctrm with all of
>>> my data sets.
>>> 
>>> I first run ft_freqanalysis with
>>> cfg.method = 'mtmfft';
>>> cfg.taper = 'dpss';
>>> cfg.output = 'fourier';
>>> cfg.foilim = [1 45];
>>> cfg.tapsmofrq = 1;
>>> cfg.keeptrials = 'yes';
>>> cfg.keeptapers = 'yes';
>>> freq = ft_freqanalysis(cfg, data);
>>> 
>>> Whether I do or I do not specify cfg.channel and cfg.channelcmb here
>>> does not seem to make a difference for my problem later on. The
>>> spectra look absolutely fine.
>>> 
>>> I then run ft_connectivityanalysis with
>>> cfg.method    = 'coh';
>>> cfg.removemean= 'yes';
>>> cfg.trials    = 'all';
>>> cfg.feedback  = 'yes';
>>> cfg.complex   = 'abs';
>>> coh           = ft_connectivityanalysis(cfg, freq);
>>> 
>>> The cohspctrm looks like this:
>>> val(:,:,1,1) =
>>> 
>>>     1   Inf   Inf   Inf   Inf   Inf   Inf   Inf   Inf   Inf   Inf
>>> Inf   Inf   Inf (etc.)
>>> 
>>> I get the following warning during ft_connectivityanalysis
>>> 
>>> Warning: Divide by zero.
>>>> In ft_connectivity_corr at 103
>>>  In ft_connectivityanalysis at 524
>>> Warning: Divide by zero.
>>>> In ft_connectivity_corr at 104
>>>  In ft_connectivityanalysis at 524
>>> 
>>> The same happens if I try to calculate PLV through ft_connectivityanalysis.
>>> This is with Fieldtrip version 20130422 and 20130427, Matlab R2009a,
>>> Windows 7 (64bit).
>>> 
>>> Any advice?
>>> Thanks!
>>> Lucia
>>> _______________________________________________
>>> fieldtrip mailing list
>>> fieldtrip at donders.ru.nl
>>> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip
>> 
>> Jan-Mathijs Schoffelen, MD PhD 
>> 
>> Donders Institute for Brain, Cognition and Behaviour, 
>> Centre for Cognitive Neuroimaging,
>> Radboud University Nijmegen, The Netherlands
>> 
>> Max Planck Institute for Psycholinguistics,
>> Nijmegen, The Netherlands
>> 
>> J.Schoffelen at donders.ru.nl
>> Telephone: +31-24-3614793
>> 
>> http://www.hettaligebrein.nl
>> 
>> _______________________________________________
>> fieldtrip mailing list
>> fieldtrip at donders.ru.nl
>> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip
> _______________________________________________
> fieldtrip mailing list
> fieldtrip at donders.ru.nl
> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip

Jan-Mathijs Schoffelen, MD PhD 

Donders Institute for Brain, Cognition and Behaviour, 
Centre for Cognitive Neuroimaging,
Radboud University Nijmegen, The Netherlands

Max Planck Institute for Psycholinguistics,
Nijmegen, The Netherlands

J.Schoffelen at donders.ru.nl
Telephone: +31-24-3614793

http://www.hettaligebrein.nl

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20130502/92a4183e/attachment-0002.html>