Practice at Retrieval
The single most important variable in promoting long-term retention and transfer is "practice at retrieval"--learners generate responses, with minimal retrieval cues, repeatedly, over time. "Practice at retrieval" simply means that the learner accesses their memory to retrieve some particular information without relying on external memory aids. In other words, the learner attempts to recall this information in some form given minimal cues. By doing so repeatedly, especially in varied contexts, the learner strengthens access to this information, facilitating later recall (i.e., long-term retention) and transfer of knowledge across contexts.
Several studies (e.g., Cull, 2000; Glover, 1989; Wheeler & Roediger, 1992) demonstrate the beneficial effects of "practice at retrieval" on learning. In these studies, practice at retrieval has been shown to be more effective than merely spending more time studying the material without actively engaging in memory retrieval. Practice at retrieval can occur during either review or test sessions.
Related to practice at retrieval are two different effects. One is the "testing effect," in which intervening tests improves learning of concepts that are retrieved from memory. That is, later recall of information is enhanced on a final test--but only for those concepts that were previously tested, or recalled. This testing effect is in turn influenced by the "spacing effect." The spacing effect implies that when review/testing sessions occur close in time, they are not as effective in enhancing learning as when these sessions are distributed, or spaced out, over time.
The importance of spacing tests is highlighted in a series of experiments examining the testing effect in educational settings (Glover, 1989). In these experiments, undergraduates and seventh graders were tested on their recall of either essay passages or labels for diagrams. When a single intervening test was administered immediately after initial learning, the beneficial effects of testing were minimized, relative to when the intervening test was administered two days later. Furthermore, when intervening tests are spaced, two tests were more effective than a single test in improving long-term retention of material.
These experiments (Glover, 1989) also demonstrate the impact of retrieval cues. Compared to a cued-recall or recognition intervening test, a free-recall test produced better performance on a final test, regardless of the format of the final test. It is important to note that free recall involves minimal retrieval cues as opposed to the other two types of intervening tests. These findings suggest that the act of retrieving from memory given minimal cues enhances the testing effect.
This demonstration illustrates the testing effect.
Educational Applications Align lectures, assignments and tests, so that important information will have to be remembered at different times distributed throughout the course, enhancing long-term retention. Have students retrieve this information in multiple ways by either varying the questions or context in which it is assessed:
During lectures, ask students questions to elicit responses that reflect understanding of previously introduced course material. This serves the dual purpose of probing students' knowledge, so that misconceptions can be directly and immediately addressed in the lecture. On homework assignments, have students retrieve key information from lectures and readings. Chapter summaries, for instance, may include study questions that ask students to recall major points or conclusions to be drawn from the reading. Encourage group studying in which students actively discuss course topics. In these groups, students have an opportunity to explain difficult course concepts to one and another, engaging in "practice at retrieval." As with probing questions during lectures, test questions offer another opportunity for "practice at retrieval," thus, potentially enhancing knowledge of the material being tested. Ideally tests should be cumulative and test items should probe for understanding of the material. Suggested Readings Cull, W. (2000). Untangling the benefits of multiple study opportunities and repeated testing for cued recall. Applied Cognitive Psychology, 14, 215-235. A series of experiments demonstrates that learning is improved when learners are required to actively retrieve information from memory (i.e., multiple testing) rather than passively studying the correct response. These beneficial effects of testing on long-term retention are enhanced when testing, or "practice at retrieval," is distributed over time.
Dempster, F. N. (1988). The spacing effect: A case study in the failure to apply the results of psychological research. American Psychologist, 43, 627-634. This paper discusses the history of the spacing effect and why it has failed to be broadly applied in classrooms. Although the theoretical basis of the spacing effect may not be fully understood, the author concludes that the implications of this robust effect should be utilized and further tested in educational settings. The findings yielded from applied research, therefore, can shed light on the conditions that do and do not enhance the effect.
Dempster, F. N., & Perkins, P. G. (1993). Revitalizing classroom assessment: Using tests to promote learning. Journal of Instructional Psychology, 20, 197-203. This paper examines the potential of classroom tests to improve learning and ways to maximize the beneficial effects of testing, including spacing tests over time.
Glover, J. A. (1989). The "testing" phenomenon: Not gone but nearly forgotten. Journal of Educational Psychology, 81, 392-399. A series of experiments illustrate the "testing effect" and how it is enhanced when tests are spaced and the process of retrieval is complete (e.g., free recall vs. cued recall).
Wheeler, M. A., & Roediger, H. L. (1992). Disparate effects of repeated testing: Reconciling Ballard's (1913) and Bartlett's (1932) results. Psychological Science, 3, 240-245. Another series of experiments demonstrates the importance of timing, or spacing, for testing. Whereas Bartlett had found that repeated testing decreased memory for story details over time, Ballard had found that repeated testing improved memory for lines of poetry. Attempting to reconcile these disparate findings, the authors conclude that recall of information is optimally benefited by intervening tests when they occur close in time.