Dear Andy,
I'm not familiar with 3dclustsim but if you do a search on this term on
the mailing list, you should find advices from Donald and others
regarding this tool. See also:
https://gate.nmr.mgh.harvard.edu/wiki/whynhow/index.php/3dClustSim
SPM smoothness estimate is available in:
>> load SPM.mat
>> SPM.xVol.FWHM % {voxels}
>> M = SPM.xVol.M(1:3,1:3);
>> SPM.xVol.FWHM .* sqrt(diag(M'*M))' % {mm}
That said, may I ask you what is your rationale for wanting to use
3dclustsim? As you are asking a question about AFNI on the SPM mailing
list, I thought it would be appropriate to add a quote from someone
working with FSL:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=spm;2f091e87.1001
Anyway, this is a genuine question: queries about 3dclustsim/alphasim
occur regularly and I'm therefore curious of what your answer is.
Best regards,
Guillaume.
On 08/05/15 08:25, Andy Yeung wrote:
> Dear all,
>
> I'd like to use 3dclustsim from AFNI.
> Documents say we need to input the estimated smoothness, and experts say
> estimate by using residual image. I wonder what values should I input
> for smoothness? Those from beneath the SPM GLM results table after
> clicking a T-contrast?
>
> Best,
> Andy
--
Guillaume Flandin, PhD
Wellcome Trust Centre for Neuroimaging
University College London
12 Queen Square
London WC1N 3BG
|