Thanks Jo/Tim. In fact, testing for whether a 3-1 contrast is due to a
primarily linear trend is actually quite easy, by including also the
quadratic contrast as well (-1 2 -1 and its inverse)- this would
immediately show deviations from the linear trend. Hope this makes sense!
Thanks, Steve.
On Fri, 6 Jun 2003, Joseph Devlin wrote:
> I just wanted to add a brief note to Tim's response.
>
> >The only real way to be more confident about your trend is to collect data
> >at more points on the line!
>
> If you do only have three conditions and you are interested in places that
> show a linear increase/decrease across those three, then one method for
> checking this in your data is to compute the I3 > I1 contrast and then
> within all areas identified, compute the mean signal change across the
> three conditions and plot them. That way you can see immediately if I2
> screws things up by being outside the I1 ... I3 range. It's just about the
> only way I can think of given only three conditions to demonstrate a
> monotonic increase/decrease.
>
>
> Joe
>
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
|