Hello Angela,
not that I know too much about this specific matter, but given your
desparate call for help ;) let me have a try:
> If I include the spatial and temporal derivatives in my design
> matrix, and tell SPM to generate the contrasts for me (1st level), it
> shows me two types of contrasts: F-tests, which include condition
> plus derivatives, and t-tests which are defined separately for a
> given condition and for its derivatives. Does it make sense to
> perform t-test on, say, the spatial derivative alone? What would this
> tell me?
I am not exactly sure what you mean with spatial derivatives, but the
idea of temporal derivatives is that they may explain variability in,
e.g., the bold response. If you explain this additional variance with an
extra regressor, the estimation of the main effect is going to be less
variable, hence stronger. This is the logic behind including them in the
design (when specifying a t-test), but not in the contrast (they take
away noise, but you do not necessarily want to look at it :)
The idea behind an omnibus F-test is that you look at wherever in the
brain is variance explained in whichever direction (positive, negative)
by whatever regressor you entered. This may then also include the
derivatives, but I am not sure it makes sense to look at them only
(although you could conceivably come up with a study examining, say, the
difference in variability in the bold response between atherosclerosis
patients and healthy controls - in such a comparison, looking at the
variance explained by the derivatives might make sense).
> I've also seen some researchers that just include the derivatives in
> the design matrix but do nothing with them, do we gain anything by
> doing this, or could we just as well not include the derivatives if
> they aren't going to be used anyway?
See above. They are meant to explain variability that otherwise may
muddy the water of your main effect, so including them, but not looking
at them, may make sense.
Hope this helps,
Marko
--
____________________________________________________
PD Dr. med. Marko Wilke
Facharzt für Kinder- und Jugendmedizin
Leiter, Experimentelle Pädiatrische Neurobildgebung
Universitäts-Kinderklinik
Abt. III (Neuropädiatrie)
Marko Wilke, MD, PhD
Pediatrician
Head, Experimental Pediatric Neuroimaging
University Children's Hospital
Dept. III (Pediatric Neurology)
Hoppe-Seyler-Str. 1
D - 72076 Tübingen, Germany
Tel. +49 7071 29-83416
Fax +49 7071 29-5473
[log in to unmask]
http://www.medizin.uni-tuebingen.de/kinder/epn/
____________________________________________________
|