Hi Matt,

I am not sure why you need to calculate all this. The image stats should be enough. Those numbers you are computing from functionally derived ROIs on the same subjects are going to be biased.

How to compute the standard deviation depends on what you are to do with it afterwards, or what it is you want it to describe. But to reiterate, I don't know why you will need to bother with it.

cheers,
ziad