Dear Mike,
I'd recommend analyzing this using a FIR model, preferably using our
task-based multivariate analysis software fMRI-CPCA, like we did in
these papers:
http://www.cnoslab.com/pdfs/epoch-specific-functional.pdf
http://www.cnoslab.com/pdfs/decreased_efficiency_of_task_positive.pdf
you can download the software, free of charge, from here:
www.nitrc.org/projects/fmricpca
That way you just lock the timing to s1, and estimate any changes in
HDR that happen over the next 20 seconds. The software will give you
the 3-4 networks involved in these 20 seconds of HDR changes.
Typically we get one network involved in encoding, another network
involved in response, and a third that combines the maintenance period
with other stages. The software provide a HDR shape for each subject,
so differences between groups and conditions can be testing using
mixed-model ANOVAs. All the details are listed in the papers I linked
you to above.
Best Wishes,
Todd Woodward
On Mon, May 25, 2015 at 7:35 AM, Mike <[log in to unmask]> wrote:
> Hi SPM experts,
>
> My experiment is a working memory ("wm") task, in which the first stimulus (s1) was presented, followed by a 5 sec wm interval and then the second stimulus (s2). Both s1 and s2 are events (electrical stimulus), each lasting for 10 ms only. Let's say that the onset time of s1 in a trial is at time 0 sec. Since its an event, I specify the onset time of s1 as 0 sec in SPM analysis, and its duration as 0. My question is, how should I specify the onset time and duration of the EV "wm" of this trial? Onset time = 0.01 s and duration = 4.91 s?
>
> Thanks
>
> Mike
--
Todd S. Woodward, Ph.D.
Cognitive Neuroscience of Schizophrenia Lab (CNoS)
http://www.bcmhari.ca/faculty/woodward-todd.php
http://www.cnoslab.com/donate.html
604-875-2000 x4724
*****
Associate Professor
Department of Psychiatry
University of British Columbia
BC Mental Health and Addiction Research Institute
Vancouver, Canada
|