I had a couple of spm questions and was wondering whether anyone might
possibly be able to help? The two questions are about two different studies.
1) I have a study that has been partially analysed using spm5b, with
some .mat files prepared for the analysis. I've noticed that for
the TA in the slice timing set was set to be 1, while more accurately it should
be 1.047.. Do you know how the TA is used in this step? Will it create a
cumulative error? Should I redo the preprocessing with the more accurate
TA?
2) My second question relates to an analysis using SPM2. It is of a
rather long task using eight 9 minute blocks. I have an analysis with global scaling but the
estimates for smoothness at the 2nd level are very high (over 30) for several of
the contrasts I am interested in (although not for all of the contrasts).
If the same analysis is done without global scaling the smoothness estimates are
around 10. I was wondering whether anyone had any suggestion as to why this
might be? At the first level the smoothness tends to be lower with the global scaling.
Thanks very much for your help.
Joana
|