Hi there -
A while ago I got extensive help from some Steve and some other folks on here
on how to set up a design for a rather complicated study I'm working on.
There were 16 subjects, with 15 conditions per subject as part of a 3x5 full
factorial, within subject design. In the end, we ended up using a single
second level analysis which took all 240 copes in as input at once (!)
Well, just for kicks, I wanted to see, at a single subject level, how my
results might change if I were to take the two scans which composed each
subject's data and actually analyze them separately. Mind you, the session
was split purely because of data collection limitations that prevented me
from running it as one huge scan ... the full balancing of all
conditions/images is achieved only over both scans. This is why I had been
analyzing them as a single concatenated dataset previously.
Anyway, I have two questions.
1. If I want to look at just a single subject's data and see how it differs
when I do all the processing on the scans separately vs. together, I guess I
would use a higher level analysis which uses only a single EV, with ones for
both inputs, to get an average. This single analysis should be applied to
all the contrasts I specified in each scan separately. Yes?
2. When I try to actually do this, FEAT complains that my data hasn't been
registered. Well, I don't need to register it since they are in the same
data space. Is there any way to turn off this complaint, or do I have to
create some type of fake example_func2standard.mat file in a fake /reg
directory anyway?
thanks,
Ed
--
Ed Vessel
U. of Southern California [log in to unmask]
Dept. of Neuroscience
HNB, 3641 Watt Way http://geon.usc.edu/~vessel
Los Angeles, CA 90089-2520
(213) 740-6102
|