Dear SPMers.
It does not seem that regressors of interest (e.g. box-car regressors)
is centered in spm2b. I want to know the influence of it on the estimated parameters.
Suppose that we conduct a simple on-off block design
and set up the design matrix with regressors of a box-car * hrf and constant term.
In spm99, the box-car regressor has been centered, but not in spm2b.
I think that it gives a different sense to constant term.
Namely, in spm99, the constant term is the mean amplitude of the time-series in a voxel,
in spm2b, that is the amplitude of rest (off) condition. Is my understanding true?
If it is true, the baseline (response = 0) of "predicted" time series
adjusted by appropriate contrast is rest-condition and
the unit of the response is the percent signal change in terms of global signal?
Thank you for your help.
------------------------------------------------------
Takanori Kochiyama
Graduate School of Human and Environmental Studies
Kyoto University
E-mail [log in to unmask]
Tel +81 75 753 7862
|