Print

Print
Kathleen - here are my 2 cents.  First, the timing of the peak in the HRF is not really grounded in theory, it's more of an empirically derived parameter (Karl can say more about where they got their particular parameter estimates from).  Second, I think that there is substantial disagreement over whether measurement of the HRF in a reference area (eg visual cortex) should be used to determine the timing of the regressor for the whole brain, given that there is substantial variability across brain regions in the shape of the HRF (e.g., Handwerker et al,. 2004, Neuroimage).  

it seems to me that the best way to convince your committee member is to estimate the entire response empirically using a finite impulse response basis set.  By fitting such a basis set and then using an F test across all regressors, you can detect activation with no assumptions about the shape of the response.  then you could look in some of the regions that come out and examine the timing of the response.  A downside of this is that you risk overfitting, but for the purposes of convincing your committee member it should be suitable.  Another intermediate option is to use an optimized basis set, like those created using the FLOBS tool in FSL.

cheers
russ

On Jun 15, 2007, at 6:02 PM, Kathleen W. Smith wrote:

Dear Dr. Friston, Dr. Worsley, and other SPM experts,
 
One of my committee members is setting me a very interesting challenge:
 
"...my primary concern is confirming that your analysis is not
compromised by using an incorrect hrf peak. We spent a lot of work deciding on
the correct oxygenation peak delay to use in our analyses, and perhaps now
(many years later) 6s is considered stable across studies. I've never seen that
statement made, so my concern is that each study may need a different hrf peak.
This is critical since all your analyses are based on the assumption that
oxygenation will peak at 6s, and if it doesn't (or more troubling, if the peak
differs by condition, which I sincerely hope it doesn't and I don't expect),
your analyses are severely compromised. All that said, you can placate my
concerns by choosing one set of predictions (your choice, but choose the most
critical hypothesis), and showing that the observed hrf peak is the same for
all conditions, and that the peak is close to 6s."
 
 
Having read Kalina Christoff's procedures on identifying individual hrf's, I don't believe that I can establish the hrf for individual Ps retroactively from my existing data. Christoff indicates that such a study would be designed so that no stimulus would evoke a new hrf until the previous one had run its course.
 
I know that the canonical hrf peaks at six seconds and that this is grounded in theory.
 
I have been analysing my data using SPM2, using hrf+time derivative.
 
I wonder whether there is an indirect way to address this question. Would any of you mind pointing me in the right direction?
 
Many thanks!
 
Kathleen Smith


---
Russell A. Poldrack, Ph.d.
Associate Professor
Wendell Jeffrey and Bernice Wenzel Term Chair in Behavioral Neuroscience
UCLA Department of Psychology
Franz Hall, Box 951563
Los Angeles, CA 90095-1563

phone: 310-794-1224
fax: 310-206-5895