Those are the dimensions of the beta image.
If you want the values from a specific cluster, you should search the list with the term "spm_clusters". There have definitely been posts that will tell you how to do it.
Chris
On 11/15/2013 10:53 AM, Julie Morgan wrote:
[log in to unmask]" type="cite">
Dear SPM experts,
I have activation in a particular cluster. I would like to correlate that activity with a behavioral score (from each participant).

I think perhaps I am on the wrong track altogether, and would really appreciate help with this.......

I am working at the group level of analysis.

I found a helpful  message ("Re: beta value") on the SPM list from Vadim Axel, dated 7 June 2009, and as a result, I entered the following information at the Matlab prompt:
vol = spm_vol('beta_0003.img');
[Vdata] = spm_read_vols(vol);

  The dimensions of the resulting Vdata variable are 79 x 95 x 68

I have scoured the list but unfortunately I am still puzzled by what these values refer to.
They do not match onto my design matrix dimensions (66 * 8). They do not match the number of images in the design, or number of conditions (8), or number of voxels in the glass brain (using not only the T but also the F contrast, as the beta surely comes from the F contrast)  for the particular contrast from which I noted the cluster of interest. None of these dimensions is a multiple of the number of participants (N = 33). I do not see any clues from reading the information below the glass brain.

Any advice will be appreciated. Thank you.
Julie