Dear spm list,
I'm writing for clarification with analysis using the Volterra series
option. I have an event-related design with 4 types of events randomly
intermixed. The events are close enough together in time that I think
it's necessary to model nonlinear (saturation) effects - thus the Volterra
convolution.
My question is about how to interpret the results of the analysis, and
which contrasts to use. Say my four event types are A,B,C, and D. The
model includes regressors for each type, plus A x A, A x B, etc.
interactions for all permutations. I'm interested in areas where
activation due to A & B is greater than C & D - i.e., a 1 1 -1 -1 contrast
over these four. I'm not sure what contrast weights to use for the
interaction regressors, though. If I leave them as zeros, it seems like
I'm doing something like finding linear effects after accounting for
"saturation" effects. But these nonlinear terms might be an important
part of the response - so I'd want something like A + B with their linear
AND nonlinear components vs C + D linear & nonlinear. How might I enter a
contrast that tests the difference between A + B and C + D, where what is
tested are predictors that are essentially linear responses adjusted for
saturation effects?
Thanks for your help!!
Tor
_____________________________
Tor Wager
Department of Psychology
University of Michigan
Cognition and Perception Area
525 East University
Ann Arbor, MI 48109-1109
Office: 734-936-1295
Home: 734-995-8975
Email: [log in to unmask]
_____________________________
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|