Hi Jack,
One answer is to use Melodic to discover the truth that is really out
there....
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.
:)
On Fri, 18 Apr 2003, Jack Grinband wrote:
> Hi Steve,
> Is there any way for FSL to model a monotonic increase across several
> conditions without specifying exactly what that increase will be? For
> example, let's say that in my task there are 3 conditions each representing
> increasing levels of difficulty. Unfortunately, I don't know if the
> relationship between the conditions is linear, exponential, etc. All I
> know is that the signal should increase as difficulty increases.
>
> Along the same lines... is a (-2,0,2) contrast different from a (-1,0,1)
> contrast? That is, in a (-1,0,1) contrast will Feat find voxels with any linear trend or
> only those voxels that have a linear trend with a slope of 1 (as specified by the
> contrast vector)?
>
> thanks a lot,
>
> jack
>
> --
> Jack Grinband, PhD
> Mind and Brain Institute
> Columbia University
> NYSPI Annex Room 518
> 1051 Riverside Drive
> New York, NY 10032
>
> email: [log in to unmask]
> phone: 212-543-6931 ext 306
> fax: 212-543-5816
>
Stephen M. Smith MA DPhil CEng MIEE
Associate Director, FMRIB and Analysis Research Coordinator
Oxford University Centre for Functional MRI of the Brain
John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
|