matlab crash after sourceanalysis with correctm = 'cluster'

[Jan Mathijs Schoffelen]

Hi Ingrid,

Do you perhaps have very big functional volumes and many subjects?
Typically, I downsample my volumes back to about my original
grid-dimensions, after spatially smoothing the individual volumes.
Normally, memory problems could occur when applying a nonparametric cluster
threshold, but as far as I know, this is the default option. This therefore
should not cause the problem (however it's worth checking in the code, or
specifying cfg.clusterthreshold = 'parametric', of cours along with one of
my favourite configuration-fields cfg.critval).

Yours,

Jan-M

-----Original Message-----
From: FieldTrip discussion list [mailto:FIELDTRIP at NIC.SURFNET.NL] On Behalf
Of Ingrid Nieuwenhuis
Sent: Friday, June 01, 2007 5:48 PM
To: FIELDTRIP at NIC.SURFNET.NL
Subject: [FIELDTRIP] matlab crash after sourceanalysis with correctm =
'cluster'

Hi

I'm doing sourceanalysis with the following settings:

cfg = [];
cfg.parameter           = 'stat';
cfg.method              = 'montecarlo';
cfg.statistic           = 'pooledT' ;
cfg.numrandomization    = 1000;
cfg.correctm            = 'cluster';
cfg.design(1,1:2*Nsub)  = [ones(1,Nsub) 2*ones(1,Nsub)];
cfg.design(2,1:2*Nsub)  = [1:Nsub 1:Nsub];
cfg.ivar                = 1;
cfg.uvar                = 2;
cfg.clustercritval      = [-1.96 1.96]; %because Z-values

GA_source_norm_cl =
sourcestatistics(cfg,source_norm{1:Nsub},pseudosource_norm{1:Nsub});

Everything goes well, but after the 1000 randomizations matlab crashes, also
on the 16 GB WM mentats.
I think it's caused by the multiple comparison correction 'cluster', because
without that I doesn't crash (when cfg.correctm = 'no' or fi 'max' does
work).

Does anyone have any suggestions on how to solve this? I've attached the
crashdump.

Thanks,
Ingrid

----------------------------------
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/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. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip.