Hi David,

To look at the correlation between functional connectivity and a subject-level variable (e.g. entered as a second-level covariate named "mycovariate") simply select 'AllSubjects' and 'mycovariate' in the between-subject effects list and enter a between-subjects contrast [0 1] (in the conditions and sources  list simply select your condition -e.g. rest- and seed -e.g. atlas.Hippocampus.l-, respectively, and leave the default 1 contrasts there).

In general CONN second-level model definitions work somewhat similarly to SPM's model specification. In the between-subjects effects list you have a list of all possible between-subjects effects and you select there which ones you want to include into your second-level model (in our case we are choosing 'AllSubjects' -a constant term- and 'mycovariate' -your covariate effect- so your second-level model defines a simple regression equation). Then in the between-subjects contrast field you enter which particular contrast (across the selected subject effects) you want to evaluate (in our case we are choosing [0 1] to evaluate the effect of the regression with "mycovariate" term). Last, in the conditions and sources list you simply select the desired conditions and sources that you would want to evaluate (and if you select multiple conditions and/or multiple seeds you can also specify your desired between-conditions and/or between-sources contrasts there and these will be always treated as within-subject effects -e.g. if you want to compare the connectivity between two conditions-)

Hope this helps
Alfonso
 
Originally posted by David White: