Given a situation and a “silly” suggestion for a solution, student will improve the suggestion to make it more useful with at least X% accuracy in X out of X trials.