 |
 |
On Fri, 9 Sep 2005 11:56:54 -0400, Anna Barnes <[log in to unmask]> wrote: >Is there a way to get the SPM(t) maps into standard anatomical space >after doing the analysis in the subjects space so that I can use the >MSU utility other than going back to the original images and >reprocessing them and running the stats analysis again. One technical issue about applying spatial normalization to SPM(t) maps: Spatial normalization always involves interpolation between grids (or lattices). Interpolation can be viewed, very loosely, as a linear averaging procedure. So it's not true that the interpolation of a t-map is the t-map of the interpolation, because the t-statistic is a nonlinear function of the data. Whether this really matters in your particular case is another matter. Though in my opinion it's better to put the data into canonical space before running statistics. An exception: you can create contrast maps in non-canonical space and normalize *them* (then e.g. bring those to the second (group) level), because contrasts are (unlike t-maps) a linear function of the data. > >Thanks >anna > > >Anna Barnes, PhD >fMRI Research Laboratory >Department of Radiology >Columbia University College of Physicians and Surgeons >710 W168th St >New York, NY10032 >USA >+1212 342 0293 > >