Hi,
> I have been looking at plots of PSTH for my event-related responses and
> noticed that they do no start at 0 at time zero. Why is that? Is it because
> there is some scaling of the initial value with respect to some overall
> brain mean?
>
> How exactly is the PSTH curve obtained and does the fitted time course
> relate to PSTH?
You can think of the PSTH as calculating the best estimate of the
signal 0 seconds, 1 seconds, 2 seconds (if your time bin is one
second) after the event has occurred, and after adjusting for other
effects in the model. There is no constraint that the signal should
be at zero at 0 seconds. Just for example, random noise will mean
than the average signal at 0 seconds will not be exactly zero.
For more information on the FIR method used for the PSTH, you might
want to have a look at these papers:
Ollinger JM, Shulman GL, Corbetta M.
Separating processes within a trial in event-related functional MRI.
Neuroimage. 2001 Jan;13(1):210-7.
Dale AM.
Optimal experimental design for event-related fMRI.
Hum Brain Mapp. 1999;8(2-3):109-14.
Best,
Matthew
|