Hi, Victor,
Sorry to response to you late. There is a bug in
snpm_MSC.m. We have fixed it and attached is the
updated version of snpm_MSC.m. Please replace the old
snpm_MSC.m file with this one. You can also check our
SnPM website for the "Bugs & Fixes" (right now, this
is the only bug).
http://www.sph.umich.edu/ni-stat/SnPM/
Thank you very much for letting us know the bug.
Best wishes,
Jun
--- Victor Vorobyev <[log in to unmask]> wrote:
> Dear SnPM users,
>
> When trying to use the option "Multisubject, 1
> covariate, 1 scan per
> subject" in SnPM for the second-level analysis with
> con-images, after
> entering covariate values and number of
> permutations Matlab gives me
> this error message:
>
> ************************************************
> ?Warning: Empty matrix multiplication with unequal
> inner dimensions.
> > In C:\MATLAB\snpm3b\snpm_MSC.m at line 168
> In C:\MATLAB\snpm3b\snpm_ui.m at line 261
> Warning: Future versions will return empty for empty
> == scalar
> comparisons.
> > In C:\MATLAB\snpm3b\snpm_MSC.m at line 168
> In C:\MATLAB\snpm3b\snpm_ui.m at line 261
> ??? Error using ==> snpm_msc
> Invalid PiCond computed!
>
> Error in ==> C:\MATLAB\snpm3b\snpm_ui.m
> On line 261 ==> eval(sDesFile);
> **************************************************
>
> Can anyone tell me what could be a source of the
> problem?
>
> I use SnPM3b on SPM2, Matlab 6.1.
>
> Thanks a lot!
>
> Victor
>
>
> *********************************
> Victor Vorobyev
> Centre for Cognitive Neuroscience
> Department of Psychology
> University of Turku
> Finland
> *********************************
>
|