Successful performance on Wason’s (1960) 2-4-6 task is typically poor at approximately 20%. One reliable way to enhance solution rates is to use Tweney, Doherty, Worner, Pliske, Mynatt, Gross, and Arkkelin’s (1980) logicallyidentical Dual Goal (DG) version of the task, where participants are requested to discover two complementary rules, one labeled ‘Dax’ (the standard ‘ascending numbers’ rule) and the other labeled ‘Med’ (i.e., any other number triple). Despite the robustness of the DG effect, the mechanism by which the DG paradigm facilitates performance has remained obscure. The present studies assessed various theoretical proposals by providing participants examples of Med triples of varying ‘usefulness’, as indexed by the cues that they provided for establishing a relevant contrast class to the Dax rule. Results showed that the usefulness of the Med exemplar had a significant effect on successful discovery of the Dax rule. We propose that the present DG results can best be accommodated by extending Oaksford and Chater’s (1994) Iterative Counterfactual Model beyond its current focus on the standard single goal 2-4-6 task.