Hi Brad, this looks fine. The ony possible gotcha is in the interpretation
of the final contrast - note that it is actually asking A-B-C+D - so you
may want to use the "contrast masking" function in the poststats tab to
make sure perhaps that both A-B and C-D are positive in the areas shown by
the final contrast.
Cheers, Steve.
On Tue, 6 Sep 2005, Brad Goodyear wrote:
> I have a somewhat simple analysis to perform, but I'm not quite sure
> how to set it up.
> Let's say I have 3 subjects (S1, S2, S3), each scanned under conditions
> A,B,C, and D.
> I am interested in comparing A vs B and C vs D. But I am also
> interested in the comparison (A-B)vs(C-D).
> Would the following capture this?
>
> Input EV1(S1 mean) EV2(S2 mean) Ev3(S3 mean)
> EV4(AvsB) EV5(CvsD)
> S1:A 1 0
> 0 1 0
> S2:A 0 1
> 0 1 0
> S3:A 0 0
> 1 1 0
> S1:B 1 0
> 0 -1 0
> S2:B 0 1
> 0 -1 0
> S3:B 0 0
> 1 -1 0
> S1:C 1 0
> 0 0 1
> S2:C 0 1
> 0 0 1
> S3:C 0 0
> 1 0 1
> S1:D 1 0
> 0 0 -1
> S2:D 0 1
> 0 0 -1
> S3:D 0 0
> 1 0 -1
>
> with contrasts:
>
> C1 (A vs B): 0 0 0 1 0
> C2 (C vs D): 0 0 0 0 1
> C3 (a-b vs c-d): 0 0 0 1 -1
>
> or do I need an EV to code the a-b vs c-d directly?
>
> -BRad
>
--
Stephen M. Smith DPhil
Associate Director, FMRIB and Analysis Research Coordinator
Oxford University Centre for Functional MRI of the Brain
John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
|