Dear George,
if you look at the code in spm_regions.m (the function that performs VOI
time-series extraction), you will notice that the data in xY.u have been
projected, whitened and filtered, and I would suggest you do the same on
your resting state components, ie run:
y = spm_filter(SPM.xX.K,SPM.xX.W*y);
as well as some of the code in the "Remove null space of contrast"
section depending on what your design matrix actually contains.
Best regards,
Guillaume.
On 08/05/15 07:24, George Chahine wrote:
> Dear Peter and SPM experts,
>
> The analysis so far has ran smoothly. However, I have a further question concerning the DCM structure components. I'm modifying the time series in DCM.Y.y to reflect that of the components I am interested in analyzing rather than an anatomically defined VOI. What about DCM.Y.x0? Does it have any influence over the subsequent analysis, and if so, what should those values be?
>
> Thanks a lot in advance,
>
> George
>
--
Guillaume Flandin, PhD
Wellcome Trust Centre for Neuroimaging
University College London
12 Queen Square
London WC1N 3BG
|