Lee Kar Ho Francis wrote:
> Dear Sirs/madam,
> I would like to post a question on spm. For fMRI model, if i want
> to pull out the error variance of each voxel after parameter estimation,
> which file should i look into? or is there any built in function that
> can plot out such information easily? Actually i would like to calculate the
> denoised input (Y-e) or XB. Can anybody give me any advice on this?
> Thanks.
> Regards,
> FRANCIS
Given
Y = X b + e
where X=[X0 X1], b =[b0;b1] and X1 are the effects of interest (eg.
your stimulus), X0 the effects
of no interest (eg. a basis set modelling physiological drifts) then
Y_fitted = X1 b1
and
Y_adjusted = Y_fitted + e
In SPM, if you set up a contrast to look at
your effects of interest you can then
press the *plot button* in the bottom left window. This plots the
fitted and adjusted responses for a selected voxel. If you then select
'fitted and adjusted responses' and 'plot against .. scan' you'll
get a plot of Y_fitted in red, and Y_adjusted in blue (in the matlab
workspace these are the variables Y and y respectively)
Hope this helps.
Will.
--
William D. Penny
Wellcome Department of Cognitive Neurology
University College London
12 Queen Square
London WC1N 3BG
Tel: 020 7833 7478
FAX: 020 7813 1420
Email: [log in to unmask]
URL: http://www.fil.ion.ucl.ac.uk/~wpenny/
|