Hi,

I have a quick question regarding a mathematical detail in FSL's
implementation of the GLM.  One can imagine at least two ways of
dealing with the mean of the dependent variable in a GLM:

1) Include a "mean" regressor in the model (with all 1's), and allow
this to model the mean.
2) Before running the regression, remove the mean from the dependent
variable (and presumably also from the regressors, to improve their
fit).

It seems to me, based on toying with models that include a mean term,
that FSL uses the second method.  However, since I haven't been able
to find this in the FEAT documentation, I wanted to make sure that
this is correct.  In particular, is the mean also removed from
regressors prior to running the GLM?  Also, assuming this is the case,
is there any reason that this method is used instead of #1?  Thanks,

Ben