[1 0 -1 1 0 -1] is correct in this case. [.5 0 -.5 .5 0 -.5 would produce the same results as well (except for the con file being different by a factor of 2).

The reason is that both runs have the same number of trials for the left and right side of the equation. If you write the null hypothesis: A1=A3, you want both sides to have equal weights (e.g. 1*A1=1*A3). You want to represent the average of A1 equalling the average of A3. The average, when you have different number of trials is not averging the 2 runs, but weighting by the number of trials.

Hopefully that cleared up the issue.

Best Regards, Donald McLaren
=================
D.G. McLaren, Ph.D.
Postdoctoral Research Fellow, GRECC, Bedford VA
Research Fellow, Department of Neurology, Massachusetts General Hospital and
Harvard Medical School
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 Fri, May 4, 2012 at 11:35 AM, Martín Martínez <[log in to unmask]> wrote:
ty for the response, i think it is clear.

Regarding Donald´s response, i wanted to ask a new one considering different conditions and weights:
A1: 6 trials
A2: 6 trials
A3: 18 trials

In session 2, the design is the same (same weights for conditions). 

If i want to define the contrast A1 - A3, is it correct to define it as [1 0 0 -1 1 0 0 -1]. Is the design correct? If something is wrong, i can detail it better.

Thank you.
Martin


2012/5/4 MCLAREN, Donald <[log in to unmask]>

The other thing to consider it to weight the contrast by the number of trials in each run:

A1 - 20 trials
B1 - 15 trials
A2 - 30 trials
B2 - 5 trials

[.4 -.75 0 .6 -.25 0]

Best Regards, Donald McLaren
=================
D.G. McLaren, Ph.D.
Postdoctoral Research Fellow, GRECC, Bedford VA
Research Fellow, Department of Neurology, Massachusetts General Hospital and
Harvard Medical School
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 Fri, May 4, 2012 at 9:33 AM, Jonathan Peelle <[log in to unmask]> wrote:
Dear Julie,

> I am currently processing fMRI data that includes 2 sessions.
>
> Can anyone provide more info or a reference regarding Martin's statement to
> "Be aware when choosing the constrast because you´ll have both sessions in
> your design matrix."?
>
> For example, if I have 2 sessions and 3 conditions per session:
> 1: A B C
> 2: A B C
>
> If I wanted to define the contrast A -B, would the contrast weights be:
>
> 1 -1 0 1 -1 0
>
> or
>
> 0.5 -0.5 0 0.5 -0.5 0

These two contrasts will give you equivalent t statistics, but the
contrast estimates will be different (though the relative size is the
same), because the error scales with the contrast weights. The second
one - where the positive and negative contrast weights sum to 1 - is
technically the average of the conditions. But because the statistics
come out the same, I don't think people usually get too bothered about
it.

Hope this helps!

Best regards,

Jonathan

--
Dr. Jonathan Peelle
Center for Cognitive Neuroscience and
Department of Neurology
University of Pennsylvania
3 West Gates
3400 Spruce Street
Philadelphia, PA 19104
USA
http://jonathanpeelle.net/