taking a few seconds to predict the answer before learning it, even when the prediction is incorrect, seemed to increase subsequent retention of learned material. This was true even when that prediction time substitutes for—rather than supplements—more conventional forms of studying.
