tinv / several questions

[Roemer van der Meij]

Hi Andrea,

I think I can answer the second question for you.

2) ft_clusterplot: Since cfg.highlight has changed, I keep getting a warning
although I have now switched the setting to cfg.highlight = 'on'. Is there a
possibility to implement something like the following example (taken from
the tutorial on cluster-based permutation tests on event-related fields)?


It looks as if you are referring to ft_topoplotER and not
ft_clusterplot. If that assumption is correct, than you can definitely
use something like:
cfg.highlight = find (pos_int==1|neg_int==-1);
However, cfg.highlight would then need to be changed into
cfg.highlightchannel. In the new highlighting configuration,
cfg.highlight is used to determine highlighting-method (like 'on',
'labels', 'numbers'), while cfg.highlightchannel is meant to indicate
which channels your actually want to highlight. This can be a numerical
index vector (like above), channel-labels, channel-labels with wildcards
and anything else ft_channelselection can take as input.

The tutorial on the wiki hasn't been updated yet to reflect the changes
to topoplotER (I'll do that soon), but the following should work:
cfg.highlight            = 'on';
cfg.highlightchannel = find (pos_int==1|neg_int==-1);
(using all the defaults this should mark every channel in
cfg.highlightchannel with a '*' and every other channel with a 'o')

Hope this helps!

Best,
Roemer


Andrea Ostendorf wrote:
> Dear Fieldtrip experts,
>
> I am currently working on a script for the evaluation of BESA TFC data with
> Fieldtrip and have found the tutorials and sample scripts a great help.
> Nevertheless, I have got a few, probably rather basic, questions and would
> be thankful for help.
>
>
> 1) cohspctrm: I would like to feed coherence data exported from BESA to
> ft_freqstatistics. The data have been brought to resemble the output of
> ft_freqanalysis so the data field has explicitly been renamed to "powspctrm"
> (and it works). Is this correct? (I have looked through the beginning of the
> ft_freqstatistics code and it looks as if the existence of the 'cohspctrm'
> field would lead to the input being brought to the format I put in.)
>
> 2) ft_clusterplot: Since cfg.highlight has changed, I keep getting a warning
> although I have now switched the setting to cfg.highlight = 'on'. Is there a
> possibility to implement something like the following example (taken from
> the tutorial on cluster-based permutation tests on event-related fields)?
>
>      pos_int = mean(pos(:,m(k):m(k+1))')';
>      neg_int = mean(neg(:,m(k):m(k+1))')';
>      cfg.highlight = find (pos_int==1|neg_int==-1);
>
> 3) Clustering over channels:
> Is it correct that I can prevent clustering over channels by setting
> cfg.neighbours = {} ? Is it correct to use this option also if I want to
> average over channels (avgoverchans = 'yes')?
>
> 4) Lately, some "nan" functions have been added to the statfun/private
> folder because they are not found otherwise in case somebody does not have
> access to the MATLAB statistics toolbox. May I suggest that a copy of tinv.m
> be added, too? (I also placed a copy of nanmean in public/private because I
> got an error message in avgoverdim.)
>
> I am really impressed with the Fieldtrip software and also with the exchange
> on the mailing list!
>
> Best regards to all of you
> Andrea
>
>
>
> ----------------------------------
> 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.
>

--
Roemer van der Meij MSc
Scientific Programmer & Data-Analyst
Donders Institute for Brain, Cognition and Behaviour
Centre for Cognition
P.O. Box 9104
6500 HE Nijmegen
The Netherlands
Tel: +31(0)24 3612631
E-mail: r.vandermeij at donders.ru.nl

----------------------------------
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.