it seems that MATLAB changed its error message from "File may be
corrupt." to "File might be corrupt." (how sweet) so you didn't see the
warning that should have appeared:
> SnPM_ST file can not be loaded. Consider using 'set cluster-forming
> threshold now (fast)' option in SnPM 'Specify'.
I don't know what it means but it may (or might) to you.
On 11/07/16 08:12, Maria Hakonen wrote:
> Dear SPM experts,
> I am trying to analyze fMRI data with SnPM13 software.
> However, there seems to be something wrong with SnPM_ST.mat since I got the following error message:
> Correct Perm has max t 7.90171 & rank 5 out of 5000 completed permutations
> Run snpm_pp for full results
> Done 'Compute'
> Running 'Inference'
> SnPM: snpm_pp
> Working on spatial extent...
> Computing suprathreshold voxels...done
> Loading & conditioning SupraThreshold statistics...Failed 'Inference'
> Error using snpm_pp (line 831)
> Unable to read MAT-file SnPM_ST.mat
> File may be corrupt.
> In file "/spm12/toolbox/SnPM13/snpm_pp.m" (???), function "snpm_pp" at line 831.
> In file "/spm12/toolbox/SnPM13/config/snpm_run_pp.m" (v1716), function "snpm_run_pp" at line 20.
> The following modules did not run:
> Failed: Inference
> I wonder if someone could please let me know how to fix this problem.
> I have tried to run the analysis following the fMRI example on the SnPM webpage: http://www2.warwick.ac.uk/fac/sci/statistics/staff/academic-research/nichols/software/snpm/man/exnew/
> Many thanks already in advance!
Guillaume Flandin, PhD
Wellcome Trust Centre for Neuroimaging
University College London
12 Queen Square
London WC1N 3BG