Dear Dana,
For reasons which are lost in the mists of time, the P value column does not appear in all cluster tables. So check both cluster_zstat1.html and cluster_zstat1_std.html (or you can click on the links between the voxel space and standard space versions at the top of the html pages). It is the column labelled "P" that contains the corrected p-value that you want. The cluster index value is an arbitrary number just used as a way of referring to that cluster (e.g. cluster number 1). You get one row per cluster in the cluster table, so if you only have one row then you only have one cluster.
If you have found the value in the "P" column to be 8.7e-06 then that means that this corrected p-value is 8.7 * 10^(-6) - i.e. 0.0000087.
As I said before, the corrected p-values apply to the whole cluster, not to individual voxels, so there is no such thing as the corrected p-values for each local maxima when you are using cluster-based inference.
All the best,
Mark
On 16 Feb 2013, at 19:30, Dana Wagshal <[log in to unmask]>
wrote:
> How exactly do I find the cluster table? Is it the cluster_zstat1_std.html? I also clicked on the activation image and got the same table. However, the table lists areas of local maxima but under the cluster index there is a value of 1. Should I take this to mean that the table is reporting one large cluster? And where would I find the corrected p-values for each local maxima? I see only 1 area that is labeled "P" and there's only 1 value associated with it, P = 8.7e-06
|