Jiansong,
To address design optimality in general, I recommmend that you look at
optseq (as Ken Rogers mentioned) or Tor Wager's GA toolbox
http://www.columbia.edu/cu/psychology/tor/software.htm
But, just to narrowly answer your question, on the efficiency of a
particular design, this is easy to do for a completed SPM analysis.
Just do
% define your contrast of interest
contrast = ;
load SPM
Var = contrast'*SPM.xX.Bcov*contrast;
E = 1/Var
Just note that efficiencies don't have absolute units and are really
only useful for comparing different designs created with the same
HRF, predictor scaling, etc.
-Tom
-- Thomas Nichols -------------------- Department of Biostatistics
http://www.sph.umich.edu/~nichols University of Michigan
[log in to unmask] 1420 Washington Heights
-------------------------------------- Ann Arbor, MI 48109-2029
