Dear list,
I am finding some low-variance problems with my second-level stats (low
contrast value can lead to huge statistics because of extremely low ResMS).
Does anyone know the function of these instructions in spm_spm (line 910
in rev 4191)?
%-Modify ResMS (a form of shrinkage) to avoid problems of very low variance
%--------------------------------------------------------------------------
try
if ~strcmpi(spm_get_defaults('modality'),'fmri')
ResMS = spm_read_vols(VResMS);
ResMS = ResMS + 1e-3 * max(ResMS(isfinite(ResMS)));
spm_write_vol(VResMS, ResMS);
clear ResMS
end
end
They seem to be designed exactly for this problem. However, they do not
appear to be executed because, at least in my system, this instruction
spm_get_defaults('modality')
gives an error (reference to non-existent field 'modality').
Cheers,
Bruno
|