rick reynolds wrote:

> Hi Joel,
>
> You might note a post about similar questions regarding
> datasets created from FIASCO data using pghtoanfi. It
> seems like that program may be creating mal-formed AFNI
> datasets (under some circumstances). Is that what you are
> using?

Hi Rick;
I'm pretty sure Jeffrey is using an older version of pghtoafni, since I can do the operation he is describing with the most recent version. This is indeed the sort of situation we're trying to straighten out- I believe Jeffrey's problem is caused by the functional volume specification failing to overlap the structural volume specification once both have been translated to AFNI format. Unfortunately the negative DELTA value we've been discussing in this thread is an example of the sort of change I've had to make to avoid problems like Jeffrey's.
OK, I've just done what Jeffrey describes, and it works with my version of the FIASCO code. Jeffrey, if you're reading this, try treating the functional GrandMean dataset just as you treated the Tmap. If you can't see any 'activation' when you overlay the GrandMean on your structural image, the problem is a coordinate system problem like Rick and I have been discussing here.

-Joel