I think that the simulation will run at whatever the visualization threshold is set to. Before and after you run it, you can change it to anything you want. You are correct on your other statements.

doug

On 9/17/13 3:24 PM, Anita van Loenhoud wrote:

Hi Doug,

It is finally starting to make sense to me. It was indeed the case that before Monte Carlo, within a cluster there were different values for each vertex, and after Monte Carlo, each vertex had the same value. I had difficulty understanding the different functions of the color scale min/max threshold and the vertexwise threshold. Is the following correct:

The min/max threshold is nothing more than a visualisation threshold; it does not change anything in subsequent analyses that are performed (such as Monte Carlo), it just decides which results are significant enough to be colored or opaque.

The vertexwise threshold (the one that is next to Monte Carlo) is the one that decides which vertices should be considered in the simulation. So a vertexwise threshold of 2 only takes vertices into account that have a p value < 0.01.

After Monte Carlo, the min/max (visualisation) threshold now decides which cluster wise p-values (cwp's) should be shown in color. So, compared to before Monte Carlo, it now has a slightly different function: it does not decide which individual vertices, but which clusters should be colored/opaque.

One last question (if the above is correct): could one decide to set the vertex wise threshold at 1.3 for example, and the visualisation threshold at 2? Or do both threshold always have to be the same?

Thanks a lot for your help,

Anita _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer