Hi Ruth,
If I understand correctly your mat file (and it is very likely that I
do not) you have 12 conditions with one parametric modulator per
condition. Paramteric modulators are supposed to be used to test for
modulation internally to a condition (that is, provide different values
for each onset within the same condition). I imagine that you had in
mind a parametric modulation across the conditions but since SPM
orthogonalizes the parametric modulators with repsect to the condition,
the way you set up your design will not be useful (and might be the
source of error). You might consider one of two alternatives:
1) Have all your conditions collapsed to a single one and then introduce
a parametric modulator with 12 values.
2) eliminate the modulators and test for aprametric effects in the
contrast definition (I have not tired this approach but it seems
resonable to me).
asaf
Ruth van Holst wrote:
>
> Dear SPM users,
>
> I hope you can help me with the following; In my first level analyse
> (in SPM5) I would like to incorporate parametric modulations and one
> temporal modulation (for habituating effects).
>
> I created a mat file with the cell arrays and structures described in
> the SPM outline. So I have onsets, durations, names, mod and tmod
> variables. When I upload this mat file in multiple regressors, I
> receive the error message: error running job, index exceeds matrix
> dimensions.
>
> I have checked everything I could think that could have gone wrong.
> But I am out of ideas.
>
> Can you please help me? What goes wrong?
>
> I have included my mat.file so it could be checked for abnormalities.
>
> Thank you in advance,
>
> Best regards,
>
> Ruth van Holst
>
|