Ahh, so zstat has no post stats run on it? That makes a lot of sense now. Thanks!
________________________________________
From: FSL - FMRIB's Software Library [[log in to unmask]] On Behalf Of David V. Smith [[log in to unmask]]
Sent: March 25, 2011 1:35 PM
To: [log in to unmask]
Subject: Re: [FSL] zstat1 vs thresh_zstat1
I suspect you want the threshold map, unless you want to report uncorrected statistics.
You're right about the cluster significance contributing to the differences you're seeing. If you want to replicate the thresh_zstat1 map from the zstat1 map, then you need to look at the commands that were executed in the post-stats part of your log file.
Cheers,
David
On Mar 25, 2011, at 12:09 PM, Sean McWhinney wrote:
> After conducting a second level analysis to look at group activation maps in fslview, I'm noticing some differences between "zstat1" and "thresh_zstat1". I ran FEAT with a Z threshold of 2.3, but when comparing "thresh_zstat1" with "zstat1" (which I threshold at 2.3, just as FEAT was run with) there is less activation in the former. I was wondering if this might be related to cluster significance, or some other factor in the way the maps are computed? Would one map be more statistically valid than the other in this situation? The activation I was expecting to find is in fact present in zstat1, but not in thresh_zstat1, even under the same threshold. Any insight is greatly appreciaed on this topic.
|