Well that's a bunch of things to answer ..
1st the easy ones "contrast estimate and 90% C.I." is your effect size
with confidence interval, that show you how big is your effect ; "fitted
respone" is the modeled response (=XB once you ran the glm) but best is
to use 'adjusted response' which is the original data minus the
parameters not used in the contrast you are looking at (it's like the
effect size but in your data rather than the parameters).
eigenvariate if I am right that's the time course of your signal but
instead of the mean that the 1st eigen value across voxels
** as for the resolution best is to use the resel count (nb of
independent element in a smooth image) written a the bottom the results
.. as you can also refer to the FWHM you used to smooth your data
good luck
c
> Hellp all
> I have some questions about SPM, Who can Help me with answer?
> please tell me how can "eigenvariate", "contrast estimate and 90%
> C.I." & "fitted respone" help me? what results I can get from them?
> they show me what???
> another question is "how can I obtain contrast, smoothing or
> resolution of images with SPM analysis???
>
> Your sincerely
> Fatemeh Shahmardan
>
>
--
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.
|