Dear Donald,
Thank you for your answer, but I would like to ask you some more questions, if you do not mind.
What kind of analysis do you think that is more accurate: the one with the raw regressor (A) or the one with the parametric modulator (C)?
I would like to know if my two groups differed in activity in condition 2. However, I am worried about whether or not the activity is correctly reflected in the regressor I choose to compare both groups.
I think that the fact of including more regressors reduce total error in the model, but as far as I understand, parametric modulators contain more information (C: reaction time) than the raw regressor (A) about how the response really was. Am I right?
However, I read that the order that parametric modulator was included in the model is important, since for example, the second parametric modulator would be explaining just the variance left by the first, and so on. In my case, I am interested in the 2nd condition parametric modulator, so maybe, it is not reflecting well real activity.
Note: I ordered conditions according to the speed of the process that I am measuring, so the 2nd is slower than the first. However, I am realizing that maybe it would be better to change the order, since I am only interested in the second condition.
Thank you for your time.
All the best,
CP
|