Dear Zude,
SPM performs a serial (Gram–Schmidt) orthogonalisation of all your
parametric variates (and their polynomial expansion) within a condition,
starting with the main trial effect, see spm_orth.m and spm_get_ons.m.
Thus, the orthogonalisation is made automatically and will depend on the
order with which you entered your parametric modulations.
Best regards,
Guillaume.
zhu zhu wrote:
> Dear SPMers,
>
> I am wondering how to handle two parametric modulation in SPM. For one
> parametric parameter, the SPM will take care the orthoganolization
> problem, but I am wondering whether it's also true for two parametric
> parameters. Should I do ortoganolization first for the HRF convolved
> regressors for the two parametric parameters as well as the onset event?
> If it's, how to do it? Thank you very much!
>
> Best regards,
> Zude Zhu
--
Guillaume Flandin, PhD
Wellcome Trust Centre for Neuroimaging
University College London
12 Queen Square
London WC1N 3BG
|