Hi,
COPE stands for Contrast Of Parameter Estimates, and are only generated when you specify contrasts. If you are really just interested in the PEs then put in simple contrasts (all zeros but with a single one on the parameter of interest - e.g. 0 0 1 0 0 0) and this will give you COPEs that correspond to the selected parameters.
All the best,
Mark
On 22 Oct 2013, at 21:58, SUBSCRIBE FSL Peter <[log in to unmask]> wrote:
> Greetings,
>
> Higher-level statistics in FEAT seem to require COPE files as inputs. I ran stats on previously-smoothed and processed nifti files for my various subjects, in which the explanatory variable of interest continuously varies over the course of the fMRI session. An additional list of confounds is also included in the GLM where the rows represent TRs and the columns are the regressor values.
>
> Running FEAT in this way outputs several PE files, but no COPE files. How does one create these? I went back through the FEAT tutorial data (the fmri fluency data set), and walking through this analysis also does not seem to produce COPE files for me. What could I be doing wrong?
>
> -Peter
|