Print

Print
The interaction contrasts tests:
A1B1>A1B2 >> A2B1>A2B2

It's hard to test for equivalence between conditions, so testing A1B2=A2B1=A2B2 will be difficult. However, you could test:
A1B1 > average(A1B2,A2B1,A2B2)

This is close to the contrast you generated. The correct contrast would be:
A1B1: 1 0 1 0 1 0 0 0
A1B2: 1 0 0 1 0 1 0 0
A2B1: 0 1 1 0 0 0 1 0
A2B2: 0 1 0 1 0 0 0 1

The average of the last three would be:
(1 2 1 2 0 1 1 1)/3

So A1B1 - the average would be:

2/3 -2/3 2/3 -2/3 1 -1/3 -1/3 -1/3

Alternatively, you could multiple the first contrast by 3:
2 -2 2 -2 3 -1 -1 -1

Both contrasts will produce identical values.


Best Regards, Donald McLaren
=================
D.G. McLaren, Ph.D.
Research Fellow, Department of Neurology, Massachusetts General Hospital and
Harvard Medical School
Postdoctoral Research Fellow, GRECC, Bedford VA
Website: http://www.martinos.org/~mclaren
Office: (773) 406-2464
=====================
This e-mail contains CONFIDENTIAL INFORMATION which may contain PROTECTED
HEALTHCARE INFORMATION and may also be LEGALLY PRIVILEGED and which is
intended only for the use of the individual or entity named above. If the
reader of the e-mail is not the intended recipient or the employee or agent
responsible for delivering it to the intended recipient, you are hereby
notified that you are in possession of confidential and privileged
information. Any unauthorized use, disclosure, copying or the taking of any
action in reliance on the contents of this information is strictly
prohibited and may be unlawful. If you have received this e-mail
unintentionally, please immediately notify the sender via telephone at (773)
406-2464 or email.


On Thu, May 29, 2014 at 6:37 PM, Angela J <[log in to unmask]> wrote:
Dear SPMers,

With a 2x2 design my design matrix looks like this:

A1 A2 B1 B2 A1B1 A1B2 A2B1 A2B2

I'm interested in the interaction contrast so I did 0 0 0 0 1 -1 -1 1 (or the opposite contrast).
If I understand correctly this will test voxels which show A1B1>A1B2 but A2B1<A2B2. But what if my hypothesis is that A1B1 > A1B2=A2B1=A2B2 (or A1B1 > A1B2 > A2B1=A2B2). I came up with the contrast 0 0 0 0 3 -1 -1 -1 but SPM says it's an invalid contrast.

Is it not possible to test an interaction such as the above?

Thank you in advance,
AJ