Dear Feng,
You can easily check the values of your inputs by looking at
DCM.U.u (try plotting them).
What Klaas meant is that you could have for example a stimulus in which a singe variable increases (such as loudness of the same audio stimuli where frequency and stimulus lateralization (left right etc) are the same). Then you could define your regressors in correspondence with your stimuli such that u1 has e.g. a value 1 where SPL was 30dB and value 2 where it was 60 dB.
The same could be done for u2 etc.
Kind Regards
Branislava
|