Dear SnPM users,
It is my first time using the toolbox so I am still not sure how to setup my model and any help would be really welcome.
I would like to use SnPM to compare the connectivity of specific seeds between two groups (controls and patients). I have 26 controls but only 9 patients.
1. When I am setting up the model to the option ‘2 groups: 2 sample t-test; 1 scan/subject’, Which images should I enter? The ‘con’ images from the first level SPM t-test?
2. If I enter first the 26 controls and then the patients, would the subject index be a vector like: A A A A A A A A A A A A A A A A A A A A A A A A A A B B B B B B B B B?
3. I already smoothed my data on the first level, is the ‘FWHM for variance smooth’ option dependent on that or is an extra second level smooth step? As I have 33 degrees of freedom I would assume I don’t need it, am I right?
4. In the SnPM manual it says that work volumetrically is ‘more efficient’ but does it have any influence on the results?
5. When should I select global normalization or grand mean scaling?
6. Is the threshold regarding the gray matter mask?
Thank you in advance for your help,
Best regards, Kátia Andrade.
|