Hi - so it sounds like you want to do, for each subject, a second-level
cross-session analysis, and then a third-level analysis across subjects.
WRT the question of what kind of analysis to do at second level: FEAT is
(at second and higher levels) only offering the "complete" model, ie
mixed-effects (aka random-effects). This is the correct thing to do if you
are primarily interested in the third-level results. It is also the
correct thing to do if you are interested in looking, for a single
subject, at that subject's second-level-cross-session results from the
point of view of significant activation relative to session-effect
variances. The only remaining issue is: if you really wanted to interpret
second-level results whilst ignoring session variability, ie doing a
fixed-effects analysis, then that currently isn't an option.
Thanks, Steve.
ps - note also the recent posting about concatenation (which is generally
discouraged for signal-processing reasons), stating that this isn't
actually equivalent to either fixed- or random-effects analysis.
On Wed, 19 Mar 2003, Arun Bokde wrote:
> Hello,
>
> I have several runs for one subject and I would like to analyze them
> together at the fixed effects level. I seem to be missing something,
> because I cannot find a way to do it through the FEAT interface. I do
> not want to analyze the different runs together at the random effects
> level. How do I do this, without concantinating all the runs together ?
>
> TIA.
>
> Best, Arun
>
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
|