Hi CONN experts, 

I'm trying to find the best way to get a whole brain roi-roi connectivity matrix with task effects modeled (ie gPPI) for post-hoc analysis. Currently I have set up my imaging files and input my task parameters as conditions and run first-level analysis with analysis type as : gPPI, ROI-to-ROI analyses only and analysis options: correlation (bivariate). My understanding is this produces a connectivity matrix for each subject in conn*/results/firstlevel/ANALYSIS_01/resultsROI_Condition*.mat which should reflect the time series of each roi in the analysis correlated to the time series of every other roi independent of any specific seed. Is this the correct interpretation? Finally, if I want to get the same connectivity matrix but for a specific contrast modeled at the second level (ie Condition 1 > Condition 2) is it possible to produce this using the conn toolbox and would this be a valid approach to examine the whole-brain connectome under specific task conditions?

Thank you,
Isabella