Hi,
I just realised that I should not use FLAME for FLOBS, but non-parametric tests (randomise). I'll have a look at that; sorry.
Stefanie
From: FSL - FMRIB's Software Library [[log in to unmask]] On Behalf Of Stefanie Becker [[log in to unmask]]
Sent: Thursday, 17 November 2011 1:07 AM
To: [log in to unmask]
Subject: [FSL] ME analysis with FLOBS
Dear all,
I'm trying to compute a higher-level analysis (Flame1) over results from first-level analysis with Flobs. I'd like to feed the results of a summary statistics up to the higher-level analyses, and following previous discussion in this forum, I've computed the copes by summing the (unthresholded) zstats together, dividing them by sqrt(3), concatenated them, and now I'd like to run a mixed effects analysis from the command line. The problem I have is that FLAME seems to require varcopes as well (=> syntax of flameo command): Can I get the varcopes in a similar way, i.e., sum(varcopes)/sqrt(n) + concatenate the averaged varcopes across sessions? Or should I not use Flame in this case?
Thanks,
Stefanie
|