Dear Bas,
I am so thankful for your help. :)
I have read the paper you told in the last letter.
The temporal deviation is much easier to understand. :)
I have some thoughts and questions about the dispersion deviation. Would you please help me to figure out?
By tracing the SPM code about spm_hrf function, there are some parameters seems very important.
% p(1) - delay of response (relative to onset)
% p(2) - delay of undershoot (relative to onset)
% p(3) - dispersion of response
% p(4) - dispersion of undershoot
% p(5) - ratio of response to undershoot
% p(6) - onset (seconds)
% p(7) - length of kernel (seconds)
The questions are I cannot understand what is the dispersion of response (p(3)) and the dispersion of undershoot(p(4))?
Besides, can I account the "dispersion" is just like the width of the HRF curve?
I know how to calculate the dispersion deviation curve by tracing spm_get_bg.m code,
and I would like to know if I have the dispersion deviation curve,
can I obtain the dispersion (width) of HRF by estimating the difference of two minimum turning points (or I can just call it undershoot?) which are beside the peak of the dispersion deviation curve?
Is the dispersion deviation curve be used as this way?
I am sorry for the unfamiliar terminology.
Thanks again for your help.
Best Regards,
Debbie
|