Dear Christopher,
open the batch editor in SPM and select "SPM"->"Stats"->"Bayesian Model
Selection"->"BMS: DCM"
You will see that you can enter DCMs from multiple subjects and if you
check "Load family" you will get the context menu which is quite
explicit about the structure array you have to create for family-level
inference.
Good luck,
Thilo
On Tue, 2012-01-10 at 19:55 +0000, Christopher Lee Friesen wrote:
> I've created 9 families of models for a DCM study 10 subjects, i'm wondering about the course of action in the model selection phase. Is it possible to test which family best captures the data across all my subjects at once, or can family-level inferences only be tested on one subject at a time? If it's possible to test the families all at once, how would one go about setting this up in SPM? Thank you,
>
> Christopher Friesen.
|