A couple of comments on your design:
(1) The design looks fine. It will be singular for a within-subject design.

(2) You can only evaluate the statistics for GTE or TE effects because this is a within-subject design because the error generated from this model corresponds to this effect. For more details see previous emails from Tom Nichols and myself. The other thing to remember about this model is that the assumption is that levels of each factor have equal variance, which I think is safe in a 2x2 within-subject designs.

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
Office: (773) 406-2464
This e-mail contains CONFIDENTIAL INFORMATION which may contain PROTECTED
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, Dec 23, 2011 at 11:35 AM, Hongyu Yang <[log in to unmask]> wrote:

From the manual, I need to replace EV 8-23 with one column for all, right?
But I found some informatin from FSL forum, which said:

Replace EV4 with a set of EVs, one for each
subject ID, in the manner of the paired t-test for a 2*2 design.

In our design, time points and emotion words are within fatctors,
group is beween factors, so it is a mixed design. Do I still need to use the design
mattrix listed in the manual to replace gloabal mean as one column?