Learning by Thinking: How Reflection Can Spur Progress Along the Learning Curve

Abstract

It is common wisdom that practice makes perfect. And, in fact, we find evidence that when given a choice between practicing a task and reflecting on their previously accumulated practice, most people opt for the former. We argue in this paper that this preference is misinformed. Using evidence gathered in ten experimental studies (N = 4,340) conducted across different environments, geographies, and populations, we provide a rich understanding of the conditions under which the marginal benefit of reflecting on previously accumulated experience is superior to the marginal benefit of accumulating additional experience. We show that reflection has the potential to generate spillover effects to different but related tasks, and that reflection is mostly beneficial at the beginning of the learning curve, as long as one has accumulated a sufficient amount of experience on which to reflect. Interestingly, our study results also suggest that the way in which one engages in reflection may play a major role in its effectiveness as a learning tool. We test the robustness of the reflection effect to different tasks and its persistence over time in a series of additional studies.

Keywords: learning, reflection, knowledge articulation, knowledge codification, field experiment, laboratory experiment