Dear SPM epxerts,
I tried to use Fourier set to set up the first level analysis. The window length is 10 seconds (the presentation time of one trial) and the order is 1. I was told that this should generate two regressors (sine and cosine) for each of my experimental condition. And, for the second level analysis, the activation of each condition should be the root mean square of the sine and cosine regressors of each condition (i.e., (sine^2+cosine^2)^1/2. My questions are
1. the design matrix actually has three (rather than two) regressors for each condition. What do they represent?
2. is it possible to do the root mean square calculation in the first level analysis without ImCalc?
3. If the root mean square calculation can only be done by ImCalc, how should I see the results in the first level analysis? Or, I can only set up the contrast weight separately for sine and cosine regressor.
Thank you in advance for any suggestion!
Sincerely,
Veda
|