Hi Peter,
Thanks for your info last week. I did follow the manual steps as you recommended and things are working good. I am able to extract VOI's (by adjusting F contrast to the desired condition), design models and estimate through scripts. You advises were helpful to understand certain aspect of DCM which i was confused on.
Finally, i have one more question and i would really appreciate if you can answer that.
I designed like 30 models per session (I have four sessions). I am planning to average models AND parameters across sessions ( i.e. average model 1 across session 1, 2,3 and 4 and also its parameters). When I review the probability (Pp) for any given single model, many times its around 0.5-0.7. Not all models reach the CI > 0.9. For visualizing my data, I am using Ep.A,Ep.B and Ep.C.
Should the models Pp<0.9 CI should be dropped as they suggest not the strong probability of direct effect they depict?
for example:
model Ep.B Pp.B
1 0.025 0.51
2 0.018 0.61
3 0.215 0.87
4 0.015 0.92
Do i drop models 1,2 and 3 as they have low Pp.B.
Please let me know
Thanks
Kushal
|