If this makes sense, there aren't any topological defects in the regions of other similar results.

On Tue, Mar 6, 2012 at 4:05 PM, Allie Rosen <rosen.allie@gmail.com> wrote:
Hi Doug,

Yes, I think there actually may be a topologic defect in this area, although not directly where the cluster is (attached). Could this be the cause of the thickness result? If I go back and try to repair the defect, will that make the results more typical and, more importantly, correct?

Thanks again,
Allie



On Tue, Mar 6, 2012 at 3:57 PM, Douglas N Greve <greve@nmr.mgh.harvard.edu> wrote:
Hi Allie, yes, it means that each group is affected differently by age. The controls are increasing their thickness with age in this area (which seems strange, usually you expect general decreases with age). You might check the quality of the surfaces in this area.
doug

Allie Rosen wrote:
Hi Doug,

Thanks for the response. I've actually already run an interaction contrast and the same area is still significant (attached). I'm not sure how to interpret this either, unfortunately. Does it say that both groups are affected significantly differently by age? Can you tell me what it means? I'm having problems trying to do this myself.

Thanks very much,
Allie

On Tue, Mar 6, 2012 at 3:47 PM, Douglas N Greve <greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu>> wrote:

   Allie, to be more precise, you have selected control<patient
   regressing out the effects of age. This second part is the key.
    If you were to trace the regression lines back to age=0 (the
   meaning of "regressing out age"), then control<patient by a wide
   margin. But you have a problem here in that the regression lines
   are crossed. This means that you have an interaction between age
   and patient (no interaction means that the lines would be
   parallel). An interaction means that you cannot assess whether
   there is a difference in the thickness because this difference
   changes depending upon the age. One thing you can try is to
   actually compute an interaction contrast (something like [0 0 1
   -1]). If it is significant in this area, then there's not much you
   can do (though it may be interesting in itself). If it is not
   significant, then you can switch to a DOSS model which forces the
   lines to be parallel and makes the distance between the lines be
   independent of age. Does this make sense?
   doug



   Allie Rosen wrote:

       Hi All,

       I'm having trouble reconciling the cluster colours with the
       graph that can be made when you load the group descriptor file
       and click on a point. I've attached an example.
       In this study, I'm comparing thickness in a patient and
       control group, with age regressed out. The contrast I've used
       means that blue means control<patient. Therefore, patients are
       thicker than controls. But in the graph, patients don't seem
       to be thicker. I've seen the same results with red clusters
       (i.e. they should mean that controls are thicker but the graph
       shows otherwhise). Please let me know how I can interpret my
       findings given the graph. Which group is actually thicker?

       Thank you,
       Allison

       ------------------------------------------------------------------------

       ------------------------------------------------------------------------

       _______________________________________________
       Freesurfer mailing list
       Freesurfer@nmr.mgh.harvard.edu
       <mailto:Freesurfer@nmr.mgh.harvard.edu>

       https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer


   --     Douglas N. Greve, Ph.D.
   MGH-NMR Center
   greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu>
   Phone Number: 617-724-2358 <tel:617-724-2358> Fax: 617-726-7422
   <tel:617-726-7422>

   Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting
   <http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting>
   FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html
   <http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html>




   The information in this e-mail is intended only for the person to
   whom it is
   addressed. If you believe this e-mail was sent to you in error and
   the e-mail
   contains patient information, please contact the Partners
   Compliance HelpLine at
   http://www.partners.org/complianceline . If the e-mail was sent to
   you in error
   but does not contain patient information, please contact the
   sender and properly
   dispose of the e-mail.



------------------------------------------------------------------------