> I would like to be able to determine whether a given variable
> (such as age) is best fit to the task response with a linear or a
> non-linear model. How might I go about doing this? It has been
> suggested that I enter both the independent variable and the
> squared independent variable into 3dRegAna (in 2 separate
> columns). Is this the appropriate way to examine this question?
Check out examples 2 and 3 on pages 17-22 in 3dRegAna manual:
http://afni.nimh.nih.gov/pub/dist/doc/manual/3dRegAnam.pdf
> how might I determine whether the linear or non-linear model
> best fits the data?
Add option -fcoef in 3dRegAna for each coefficient of the model, and the F value is the proportion of variation in the data that can be explained by that regressor.
Gang