the ordering of EVs or order of trial types within EVs does not matter
- what matters is the order of the COPES. After specifying your
first-level design, create A-B, A-C and B-C copes for the first level
FEAT runs and stick to this order - all will be good then once you run
group-level FEAT. If you want to do fixed-effects only, just do the
same and average the copeX files from the command line
On 20 May 2004, at 15:28, Bradley Goodyear wrote:
> I think a similar question has been posted, but bear with me...
> I understand how to merge and average COPES for multiple sessions if
> the sessions
> are exactly the same, but what if the two sessions contain the same
> trial types but
> are in different order?
> For example, if you have in the first session, ABCABCABC, and in the
> session session you have ACBACBACB. I want to combine the two sessions
> compare A-B, A-C, and B-C.
> Can I combine COPES from each first level analysis somehow to avoid
> concatenating the two sessions?
Christian F. Beckmann
Oxford University Centre for Functional
Magnetic Resonance Imaging of the Brain,
John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK
Email: [log in to unmask] - http://www.fmrib.ox.ac.uk/~beckmann/
Phone: +44(0)1865 222782 Fax: +44(0)1865 222717