Hi Claudia, thanks for sending the FEAT output.
In fact, because of the particular nature of the first-level design,
in this case it makes almost no difference whether you include a rest
EV or not.
The big difference in your second-level designs is because in one
case you used a fixed-effects model, and in the other you used a
mixed-effects model.
Cheers, Steve.
On 22 Aug 2006, at 21:48, Claudia Danielmeier wrote:
> Hi,
>
> Thanks to all who replied to the modeling questions.
>
> I doublechecked my analysis, but the results remain the same:
> plausible
> activations when I model the rest (e.g. 2 conditions involve hand
> movements
> and I find motor cortex activations) and no significant activation
> at all
> when I don't model the rest (not even in those condition where I would
> expect huge motor or visual activity).
>
> My EVs in the first level analysis look like this for the "modeled
> rest"
> version:
> contrast rest cond.A cond.B cond.C
> A-rest -1 1 0 0
> B-rest -1 0 1 0
> C-rest -1 0 0 1
>
> And they look like this for the version without the modeled rest:
> contrast cond.A cond.B cond.C
> A 1 0 0
> B 0 1 0
> C 0 0 1
>
> In the second level analysis I just want to average across
> sessions. And
> because I've defined all relevant contrasts on the first level
> already, my
> EV on the second level contains just a "1". My second level
> analyses look
> exactly the same for both versions. Might this be a problem?
>
> Thanks,
> Claudia
------------------------------------------------------------------------
---
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director, Oxford University FMRIB Centre
FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
------------------------------------------------------------------------
---
|