Try starting spm beforehand. This runs the spm_defaults.m function, which
sets up the defaults. The crash appeared because there didn't seem to be any
default settings.
Best regards,
-John
On Wednesday 31 August 2005 16:29, Claus Lamm wrote:
> Dear all,
>
> I have a readily specified design (SPM.mat) for an ER-fMRI study.
> The design has been specified with canonical hrf only.
>
> Inititial analyses now suggest that it might make sense to add
> the temporal and dispersion derivative.
>
> Is there any way to change the existing SPM.mat to include those additional
> regressors (so I don't have to enter all the onsets etc. again?)?
>
> I already tried something like:
>
> load SPM.mat
> SPM.xBF.order = 3;
> SPM.xBF.name = 'hrf (with time and dispersion derivative)';
> save(fullfile(SPM.swd,'SPM.mat'),'SPM')
>
> but this didn't work out (program always crashed when trying to estimate
> the design); error message says:
>
> Initialising parameters : ...computing???
> Attempt to reference field of non-structure array 'defaults'.
>
> Error in ==> c:\programme\spm2\spm_spm.m
> On line 464 ==> MAXRES = defaults.stats.maxres;
>
> I know that this kind of problem could be "easily" circumvented by setting
> everything up in a batch file but I had (and have) a hard time to specify
> the onsets and durations of my stimuli in the batch mode so I decided to
> enter them manually ...
>
> any help - thx a lot in advance
>
> claus
|