Dear SPMs,
I have two questions:
(1) If I have data smoothed with a FWHM of x mm and want to resmmooth
them with y mm using the s*-files, is it correct that I have to resmooth
the s*-files with a factor of the third root of y-x so that "x times
third root of y-x equals y"? If this is not correct what is the
algorithm?
(2) Consider a study with 2 groups, say I and II. The experiment has two
conditions A and B which were acquired in each individual in two
separate sessions. For individual analysis I use "Scaling" in order to
make a global normalisation. Now suppose I want to do a second level
analysis comparing condition A between the two groups. Am I wright NOT
to use the contrast-images of the scaled indivdual, first level-analysis
because they are normalized with condition (B)? If I am NOT supposed to
use the scaled contrast images and rather use unscaled
(session-specific-scaled) invidual contrast images which option do I
have to use "Grand Mean Scaling" in the second level analysis or not?
And why?
Thanks for your help
Henrik Walter
--
Dr. Dr. Henrik Walter
Abteilung Psychiatrie III
Universitätsklinikum Ulm
Leimgrubenweg 12
89075 Ulm
fon: 0731/502 1489
fax: 0731/502 1549
e-mail: [log in to unmask]
|