Dear SPMers,
I am fairly new to SPM, so I have a couple of basic questions. I am using
SPM2. I am doing a working memory task where I have a one-back condition
(A), a two-back condition (B) and a control condition (C). Each condition
is preceded by a short set of instructions (D), letting the participant know
which type of condition is coming up. So, the model is: DADBDC DADBDC
DADBDC. The instructional blocks are 7.5 seconds, and the WM and controls
blocks are each 62.5 seconds. Using the rule of thumb of SOA*2 for the high
pass filter, the high pass filter I have been entering is 210*2, or 420
seconds. This, of course, is way higher than the SPM2 default of 128. My
questions are:
1) Am I calculating this correctly?
2) Is 420 too high? The activation maps I get when I use 420 look
reasonable...not too much noise...but I would like to make sure.
3) What is the basis for the rule of thumb of SOA*2? Why not just SOA?
4) I do realize that using the frequency display on the design matrix also
provides a guide to where to set the high pass filter cutoff, but I don't
really understand how to translate the frequency display into an actual high
pass filter value. If anyone can explain this in an email, that would be
great.
Thanks for your help!
Wendy
*********************************************
Wendy R. Kates, Ph.D.
Associate Professor
Department of Psychiatry and Program in Neuroscience
Director, Center for Psychiatric Neuroimaging
SUNY Upstate Medical University
750 East Adams Street
Syracuse, NY 13210
phone: 315-464-3270
fax: 315-464-3163
email: [log in to unmask]
|