Hi
>
>
>  1. What is the difference in using -c 2 or -c 3 in randomise?
>  If this is related to the cluster size, as you write, how does
>  2 differ from 3? What is the unit?
>

the -c option is the cluster-forming threshold. The units are units  
of "t"
voxels with t-values above this will be clustered together spatially,  
and these
spatial clusters will then be considered by the statistical test.


>  If I overlay the results of these two runs, I get somewhat more
>  significant areas by using 3 than 2. Some of them are overlapping,
>  others not.


This is because you will need a larger number of voxels in your  
cluster if you threshold at a t-value of 2 than if you threshold at a  
t-value of 3 (because, in the null case, a large cluster with a t- 
threshold of 2 is more likely than a large cluster with a t-threshold  
of 3

>
>  I've gone through some of the papers by Nichols but I haven't
>  found any satisfactory answer.

I'll let him know!

>
>  2. In your technical paper you either 'correlate' covariates with
>  FA or 'regress them out'. What does this mean literally? Do you
>  compute voxelwise correlation coefficients (and if so, how do you
>  get them? How do you know whether they are positive or negative,  
> large
>  or small?
>

When we say "correlate with the covariate", we mean put this  
covariate into the GLM as a regressor of interest.
We can then compute the parameter-value and t-score for this variable  
and perform corrected thresholding using randomisation exactly as  
before.

> Are they ordinary correlations for linear dependence or some
>  nonparametric versions?).

Just the simple GLM, so linear as always.

>  Does 'regressing out' mean the usual 'adjusting for'? How do you
>  get the
>  coefficients of the covariates if GLM type estimation is used?

"Regressing out" means putting them into a GLM as "regressors of no  
interest"
we compute the linear coefficients using the GLM, and subtract these  
components from the data before we start the analysis.


>  Is there any paper you would recommend to clarify these otherwise
>  rather standard statistical ideas in this context?
>

Just think of these things exactly as you do in the GLM. There is no  
difference in the modelling. The only difference is in the inference,  
which we do by creating the null distribution from the data, rather  
than assuming a standard form.

>  Does 'correlating' technically mean that the covariates are in the
>  original design matrix whereas 'regressing out' means that they are
>  technically put to another design matrix with the prefix -x?
>

Yes

T


>  Thanks in advance for any clarification.
>
>  Best wishes,
>
>  Mervi