Thank you Alfonso, this makes sense. By "and also repeating your main second-level analyses but adding an additional second-level covariate to account for potential differences between these four subjects and other controls" do you mean to perform second level analysis with all subjects and then with those 4 excluded? Best regards