Hi Joe,
thanks. Very helpful and reassuring.
Arian
________________________________________
Von: FSL - FMRIB's Software Library [[log in to unmask]] im Auftrag von Joe Devlin [[log in to unmask]]
Gesendet: Donnerstag, 24. September 2009 11:33
An: [log in to unmask]
Betreff: Re: [FSL] double gamma hrf
Hi Arian,
I'm having trouble sending to the list at the moment so I've cc:ed you directly too. Hope this works...
Yes, the FSL double gamma function uses the parameters from the Glover 1999 paper and is the same as the default canonical HRF used by SPM. We've done some testing with it relative to the single gamma, and couldn't find a single case where the single gamma was a better fit. The paper is being revised at the moment so there is nothing to cite as yet, but you picked the better HRF function. I typically cite the glover paper. Interestingly, even the Boyton 1996 paper that introduced the single gamma clearly shows an undershoot in the data, although the authors didn't model it.
And isn`t it fair to say that it is / could be called a canonical HRF (I think the SPM canonical has also a late undershoot)?
Sure. Insofar as there is any meaning 'canonical' HRF, this is it.
What are lag and half-width of the first and second gamma. In case that comes up?
These come from the Glover 1999 article -- from the motor cortex ROI, if I remember correctly.
Hope this helps
Joe
Joseph T. Devlin, Ph.D.
Cognitive, Perceptual & Brain Sciences
University College London
Gower Street
London, WC1E 6BT
email: [log in to unmask]<mailto:[log in to unmask]>
|