There are some questions as following,

1. First we use 'Check Reg' button and 'Display' button to make sure the origin is approximately aligned to the template. But in the New segment step the image was aligned to MNI, after DARTEL and Normalise step the image was aligned to MNI again. I found that there are a few differences if I  manually manipulated a single image or not, but I have no idea about  is manually manipulating a single image needed?

2. DARTEL step use the rc1 and rc2 images which were realigned in the previous step to create templates and flow field. But why can't I use the origin image which has already manually manipulated in the first?

3. I use two sample t test to compare if there is any difference between two group in GM. I check my image and set 0.2 to threshold masking to exclude voxels outside GM. Is the setting right? Do I need to input any implicit mask?

4. The global normalisation is about dealing with brains of different sizes. So I use a matlab scrip which I found on the web and sum of the of c1, c2 and c3. But I have no idea where should I put the vector ?

I think there are three choices. (a.) Covariates, and the centering choose 'As implied by ANCOVA' (b.) Global calculation, choose user and specify my TICV vector. (c.) Global normalisation, 'Overall grand men scaling' choose yes and the thus 'Grand mean scaled value' enter the TICV vector. The 'Normalisation' choose ANCOVA.  I found there are a few differences between the three setting but still don't know what is the result mean?

Many thanks of your help

Best

ZK
--