Thanks for your reply. For the cases in which I have 3 time points for a subject I actually do have 3 sw*.nii files, not four, which I think makes sense. You mentioned I should have only three C1 files, three wc1 files, etc. However, only one c1avg*nii file was outputted for each subject (no matter if the subject had three or four time points). Furthermore, no wc1 files were given, instead I have rc1avg* files. I think this is because I used the serial registration option in SPM 12. Can you verify that this is correct, I'm a little confused now. Thanks & Happy 2015.

p.s. thanks for the stats advice!

On Tue, Dec 30, 2014 at 7:38 AM, H. Nebl <[log in to unmask]> wrote:

> This has resulted in 4 sw*.nii files for each subject.
If you only have three volumes for some of the subjects then you should also have only three c1 files, three wc1 files, ...

> 2) can I perform either a paired sample t-test or GLM (flex factorial) using the final output sw*.nii files, to for example, look
> at changes in gray matter across time?
As you have only one factor (time) you would simply use a "One-way ANOVA - within subject". Alternatively you can also run several "paired t-tests" to compare e.g. time 1 and 2. "Flexible factorial" is possible as well, but if you specify everything correctly it's identical to the "One-way ANOVA - within subject". However, within SPM you can only compare those subjects for whom data is available for all the investigated time points. There's no option for missing data/empty cells.