Hi Everyone:

I'm looking into doing a kind of dynamic functional connectivity analysis on some resting state data that has already been preprocessed using afni_proc.py. I think I'm going to take it over to FSL's MELODIC to try to run an ICA on it and then hopefully find some ICs where I can investigate how they vary across time.

It seems like the most common way to go about this is with a sliding window approach where you specify a window size (e.g. 30s/15TRs with a TR of 2s) then look at the BOLD response of the IC network there. Then move the window over 30s and do the same there, etc. to the end of the run. I feel like this is a profoundly stupid question but how would one go about making these sliding windows to be able to investigate responses within these smaller periods? Is it as easy as making a "stimulus" timing file with the timing window specifications (e.g. 30.0 60.0 etc.)? That was my first thought but it seemed too simplistic.

Thanks for any input you guys can provide.

Lauren