Dear Aser,
I think it was this paper:
http://www.sciencedirect.com/science/article/pii/S1053811998904190
In Fig. 3 for short activation time (1/3s ~= 0s) the response looks pretty much like double gamma function for parameters [6 16 1 1 6 0 32]. Please note that the comment in 'spm_hrf.m':
"p(1) - delay of response (relative to onset) 6"
is misleading. The peak occurs at alpha_1-1=6-1=5s. Many people think the peak of the canonical function is at 6s. I think it is because of this comment. Hopefully it can be changed.
Best,
Wiktor Olszowy
|