Steve,
>One answer is to use Melodic to discover the truth that is really out
>there....
So the monotonic increase in activation will appear as a separate
component? But the identification of this component will be very difficult
since we don't know how it varies across the conditions, is that right?
>But, using FEAT; [-2 0 2] will give the same answer as [-1 0 1], ie,
>looking for a linear increase across the different conditions.
>
>For a positive quadratic effect, use [1 -2 1] etc.
I guess I'm still a bit confused about what the contrast vector really
signifies. I did a simple finger tapping experiment in which the rate of
tapping was 0.5, 1.0, or 2.0 Hz. When I set up the contrast vector, I
assumed a linear relationship and set it up as (-0.5, 0, 1). Is this
correct? (I also compared it with a (-1,0,1) contrast and the (-0.5, 0, 1)
contrast gave me a higher level of significance for the areas I was
expecting to be active.)
If I had 5 conditions (e.g. finger tapping at 1, 2, 3, 4, and 5 Hz), would
my contrast vector be (-2,-1,0,1,2) for a linear trend in rate? And if I
expected an exponential increase then the contrast vector would be
(e^1,e^2,e^3,e^4,e^5)?
thanks,
jack
|