David,
Cohen et al. showed that a parametric manipulation of a working
memory produced differentiated temporal BOLD responses. Hence your question
is a complicated one: For anything other than primary cortices (and
very simple tasks) the hemodynamic delay will confounded with the
temporal dynamics of the network of neuronal activation.
But ignoring the confound, I have several tidbits that might help:
First, there will systematic variation in the delay between slices
unless you account for it otherwise (e.g. with phase shifting).
Including the derivative of the smoothed box car will model this to
some extent.
For a very simple finger moving task, Kim et al. reported that the
delay varied between subjects, but was consistent within subjects;
this paper has references to other literature on the hemodynamic
delay.
In contrast to the Kim paper, in high-temporal resolution single trial
data, for a visual perception task, I have seen the delay vary upto 5
seconds within subject, both within a slice and between slices
accounting for the acquisition delay. (unpublished results).
Hope this helps.
-Tom
References:
Cohen JD. Perlstein WM. Braver TS. Nystrom LE. Noll DC. Jonides
J. Smith EE. Temporal dynamics of brain activation during a working
memory task [see comments]. Nature. 386(6625):604-8, 1997 Apr 10.
Kim, SG, Richter, W, Ugurbil, K. (1997). Limitations of Temporal
Resolution in Functional MRI. Magnetic Resonance in Medicine
37:631-636.
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|