Dear Giancarlo
> Dear JB,
> To be sure that you can remember my experimental design I attached two
> files:
> 1. a jpeg file which containes the configured matrix of my exp design
> (factorial design, two factors: A (3 levels) B (4 levels)
> 2. the SPM_fMRIDesMtx.mat (if you have time enough to run it on your SPM
> and look at it - but I don't think so!)
>
> According to your suggestions, this should be the contrasts' setting in
> order to perform the F-test:
>
> assuming three variables a,b,c, where
>
> a =
>
> 1 -1 0 0
> 0 1 -1 0
> 0 0 1 -1
>
> b =
>
> -1 1 0 0
> 0 1 -1 0
> 0 0 1 -1
> c =
>
> 0 0 0 0
> 0 0 0 0
> 0 0 0 0
>
> the combination A=[a b c; c a b; b c a] should give me the vector
> containing the appropriate weights for the F-test.
> Could you please explain to me why?
>
well, there is (3-1)x(4-1) interaction terms in a 3x4 factorial design.
Let say you had a simpler 2x3 design with conditions A1 A2 A3, B1 B2 B3.
The interaction would be the F contrast containing
1/ the difference between 1 and 2 depending on A or B :
[1 -1 0 -1 1 0]
2/ the difference between 2 and 3 depending on A or B :
[0 1 -1 0 -1 1]
(note that the third direction : difference between 1 and 3 depending on
A or B is redundant)
so the F contrast in that simpler case is
[1 -1 0 -1 1 0; 0 1 -1 0 -1 1]
I think you can generalise this to your more complex case ...
hope this cast some light....
very best
jb
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~+
| Jean-Baptiste Poline [log in to unmask] |
| 4 Pl General Leclerc |
| CEA/DRM/SHFJ Tel +33 (0) 1 69 86 78 52 |
| 91406 ORSAY FRANCE Fax +33 (0) 1 69 86 78 68 |
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~+
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|