Hi Shukti and Stephen,

Regarding DynamicFC analyses in the context of multiple conditions, you should simply create one additional condition (e.g. named 'rest') encompassing all of your sessions/conditions, and then select that condition in the 'rest condition' field in the FirstLevel.DynamicFC tab. This will use your entire dataset (both conditions in your case) for the dynamic FC analyses. Then, when clicking 'Done', if you want to obtain condition-specific measures (e.g. look at the components separately between your two conditions) simply check the 'All conditions' checkbox in that gui before clicking on 'Start'.

Regarding DynamicFC analyses in the context of multiple groups (e.g.controls vs. patients) that is perfectly fine. We are hoping to have a manuscript out shortly detailing the specificities of these analyses, but both conceptually and in practice the way these analyses work is very similar to ICA analyses. You first estimate a number of dynamic components using your entire dataset (all subjects across multiple subject groups) and then those global components are "backprojected" into subject-specific and condition-specific factors, allowing you to then compare how these components are expressed across different conditions or across different subject groups. Compared to simply running the analyses separately across different subject groups, which would then require users to manually identify the potentially-similar components across the different analysis results, this methodology allows you to more meaningfully and automatically identify individual components across all subjects, while still allowing to investigate how these individual components might be expressed differently across different subjects or conditions. 

Hope this helps
Alfonso
 
Originally posted by Stephen L.: