playing around with the function spm_hrf, I noticed the following:
The resulting hrf peaks one second earlier than I would expect given the
specified delay of response (relative to onset), i.e. p(1).
E.g., issueing the following command, shows the above mentioned.
figure; hold on; TR = 1; peak = 6; plot(0:32, spm_hrf(TR, [peak 16 1 1 6
0 32])); plot([peak peak],[-1 1],'--r'); grid on; xlim([-1 35]);
I was under the impression, that the delay of response effectively is
the peak of the hrf. Probably I'm missing something obvious here, I
would be thankful for any guidance on this.