Well I guess you could. Just before estimating the model write
something like the following on the command line. (in the example there
is one regressor of interest (the first) and 6 motion parameters.)
load SPM.mat % be sure to be in the right directory
% Specify collumns of interest (iC) and nuisance (iG)
SPM.xX.iG=2:7 %columns of no interest
SPM.xX.iC(SPM.xX.iG)=[] % erase these columns from the effects of
interest
save SPM SPM % save the changed SPM struture.
% Estimate parameters by pressing the estimate button as usual
best
torben
P.S. You would probably want to include at least the squared movement
parameters or 24 movement related regressors as in Friston et al. 1996
MRM.
Torben E. Lund
Danish Research Centre for MR
Copenhagen University Hospital
Kettegaard Allé 30
2650 Hvidovre
Denmark
email: [log in to unmask]
webpage: http://www.drcmr.dk
On 15 Feb 2005, at 22:38, Cyril PERNET wrote:
> Hi Torben,
>
> I agree with you that most of the time, the p-values dercrease with
> motion covariates. Do you know if it exists another way than the batch
> mode to specify these regressors as covariates of no interest ... (I'm
> not so familiar with the batch mode .. I know it's easier .. it works
> well ...)
>
> Thank u,
> cyril
|