Eugene

--

Centre for Functional MRI of the Brain (FMRIB) | University of Oxford
John Radcliffe Hospital | Headington
OX3 9DU | Oxford | UK

Ph: +44 (0) 1865 222 523 | Mob: +44 (0) 7946 362 059 | Fax: +44 (0) 1865 222 717

--

On 6 April 2010 00:48, Libby O'Hare <[log in to unmask]> wrote:

Hello FSLers,

I’m trying to understand some discrepancies I’ve noticed in my single subject analyses. I am attempting to generate parameter estimates (PE) for all of my subjects within an ROI.

I have a simple A r B r C r blocked design (conditions A, B, and C with interspersed rest blocks (r)). I’ve used featquery to pull the PEs for A > rest, B > rest, and C > rest contrasts.

A > rest [ 1 0 0 ]
B > rest [ 0 1 0 ]
C > rest [ 0 0 1 ]

Given that these are simple contrasts, PE should equal COPE. However, this relationship only holds for condition A. Furthermore, I have found that the PE value for condition C is consistently identical to the COPE value for condition B (for every subject, for all runs).

Clearly something is going on with my data, but I’m stumped. My design matrices look correct and I’ve double-checked the stimulus timing files that I’ve used to model each EV.

Any advice would be most appreciated!

Thanks,
Libby