Hi Rick,
Thanks for your response! Yes the variation is only during the ITIs. On each trial, a stimulus is presented at 0s for 0.5s, and after a brief 0.4s, a second stimulus is presented for another 0.5s. After that, no stimuli is presented until 13s, and this stimulus at 13s stays on the screen for 0 to 5s (this interval is determined by subject's response time). The ITI starts at 18s, and ends between 22s to 28s.
For my experiment, I need the data from the first 13s. And I've tried using a gamma function before. But since another stimulus is presented at 13s, I was just wondering if I also need to model the signals for the stimulus at 13s.
Thanks so much for your help in advance!