I have a basic question about spm 1st level analysis.
Lets say you have one event in your model (Event A). You also have a continuous variable (V) that coincides at every time point with event A.
Is there a difference between parametrically modulating event A by V (columns in design matrix would be A V) and creating contrasts for both a positive (0 1) and negative (0 -1) linear relationship between V and A
versus modelling event A as two events with parametric modulators for both V- and V+ (columns in design matrix would be A+ V+ A- V-) with contrasts for both positive (0 1 0 0) and negative (0 0 0 1) linear relationships?
Thanks in advance,
David
|