GLM with continuous variable (e.g. error in seconds)

AFNI Message Board
1

What is the best way to run 3dDeconvolve when you have a continuous variable (e.g. error in seconds). Each trial has an error value from 0-1900. I also have baseline trials.

2

Does that describe the duration of each event?
Or as a continuous variable - is it something that
you would expect to scale the magnitude of the
BOLD response as an amplitude modulator?

  • rick
3

The latter. I guess I should really describe it as a precision metric. So we want to find regions that are more engaged for trials where there is more precision. and the precision “value” is continuous

4

ah, apparently I should be using stim_times_AM2

5

Are you all set then? Sorry for the down time
while we changed to a new website.

  • rick