T here is relatively strong empirical evidence that an effective means of enhancing learner performance is to pose ‘‘deep questions’’ (1). Examples of deep questioning include asking learners to explore the causes of a sequence of events, the motivations of the people involved, or the quality of the evidence behind a particular practice or theory. By their nature, deep questions invite learners to do more than merely to recall a fact or recognize a finding. They invite them to ponder why we accept or do not accept a particular hypothesis, to compare two alternative points of view, and to ask, What if? The principal purpose in asking deep questions is to promote deep understanding. Learners taking an exam who rely on nothing more than mnemonic devices and checklists may be able to regurgitate basic information, but they are less likely to be able assess the strengths and weaknesses of this information or use it effectively in problem solving. They are also less likely to be able to adapt what they know to new situations in which key parameters have changed. When their diagnostic knowledge is tested, they are more likely to provide ‘‘laundry lists’’ than targeted differential diagnoses. The goal in asking deep questions is not merely to promote deeper understanding of particular concepts, such as the relationship between inflammation and contrast enhancement, but to help learners develop the habit of asking deep questions about whatever they learn. A passive learner merely records what a lecturer is saying, but a deep learner is writing down not only what the lecturer is saying but questions to which it gives rise. The deep learner is the one who asks the most probing and illuminating questions at the end of the lecture. Consider the difference between the conditions of the books in the libraries of John Adams and Thomas Jefferson, the second and third presidents of the United States. The latter’s collection formed the nidus for what we today call