Hi SPMers,

          I have a few questions regarding multiple regression in SPM 5.Is there a method to model nuisance variables for regression and look at correlations with behavioral scores in the same design at second level? What is the difference between the two options for covariates given in the multiple regression model? How should the contrasts be made if a variable (age, gender etc) is to be regressed and the effect of another (behavioral score) is to be correlated? If one had to perform correlations with behavioral scores, will one sample t test do if covariates are also modeled alongside?

Thanks,

--
Sarika Cherodath
Graduate Student
National Brain Research Centre
Manesar, Gurgaon -122050
India