Hi Katharina,
Yes, the directories are exactly the same - I had saved them on an external drive and then loaded them back on the computer.
I am wondering if it has something to do with the pmod structure (which I know little about), but which is talked about in item#040060 from 2010-02-03. There is a change suggested for spm5, but i cannot find the equivalent statement for spm8.
"I figured out the problem with loading parametric modulators as part of the multicondition file in SPM5. I'm not sure why it happened, but probably at some point Matlab changed the behavior of the switch function so it no longer works well with cells.
To get this to work you need to make the following 2 changes to spm.
spm_config_fmri_spec.m
line 880: I have: P(q1).name = cond.pmod(q).name;
it should be: P(q1).name = cond.pmod(q).name{:};
spm_config_fmri_design.m
line 865: I have: P(q1).name = cond.pmod(q).name;
it should be: P(q1).name = cond.pmod(q).name{:};
It should work after you make these changes."
Rebecca
|