Memory researcher Herman Ebbinghaus performed classical overlearning studies in the late 1890's. He noticed that memory for learned material decreased over time (see also forgetting curve). Ebbinghaus recognized that lists of nonsense syllables became more difficult to recall over time, and some lists required more review time to regain 100% recall. He defined overlearning as the number of repetitions of material after that material can be 100% recalled.
Ebbinghaus measured the benefits of overlearning by comparing the ratio of overlearning to saving. "Saving" referred to the number of nonsense syllables recalled from the list, 24 hours after participants first learned the list. He found that participants recalled, or "saved", one additional nonsense syllable for every three repetitions of the items after participants had mastered the list. For example, reviewing a list 12 times after mastering the list resulted in an average recall of 4 items greater than participants who did not overlearn the items.
Meta-analysis suggests that overlearning does significantly affect recall over time. However, the size of this effect is moderated by the amount of overlearning, task type, and length of retention. Researchers included 15 studies in the meta-analysis. These 15 studies tested overlearning effects on physical and cognitive tasks. Both types of task showed an effect of overlearning. The effect size for physical tasks was smaller than the effect size for cognitive tasks. The amount of overlearning affected retention: more overlearning led to more retention on both types of task. The length of the retention interval also affected overlearning, but the effects were different for physical and cognitive tasks. Whereas participants overlearning physical tasks increased in ability during the retention interval, participants who overlearned cognitive tasks decreased in recall ability over time.
Overlearning geography facts and word definitions
In one study, researchers examined the effects of overlearning geography facts or word definitions. After one week, overlearners recalled more geography facts and word definitions than non-overlearners, but this improvement gradually disappeared after the study. This research suggests that overlearning may be an inefficient study method for long-term retention of geography facts and word definitions. Overlearning improves short-term retention of material, but learners must also spend more time studying. Over time the improvements created by overlearning fade, and the learner is no better off than someone who did not spend time overlearning the material.
In another study, researchers examined the effects of overlearning and distributed practice on mathematics knowledge. Distributed practice refers to practice that is spaced over time. In Experiment 1, participants completed 10 math problems either all at once or distributed across two sessions. Participants in the distributed practice condition performed no differently from participants in the single-session condition one week later, but distributed practice participants did perform better than single-session participants four weeks later. In Experiment 2, participants completed either three or nine practice problems in one sitting. When participants were retested one or four weeks later, no differences were found between three-problem and nine-problem participants. Researchers found no effect of overlearning on mathematics retention.
- Murphy, Gardner (1929). An Historical Introduction to Modern Psychology. International Library of Psychology.
- Driskell, James E.; Willis, R. P., & Copper, C. (1992). "Effect of Overlearning on Retention". Journal of Applied Psychology 77 (5): 615-622.
- Rohrer, Doug; Taylor, K. (2006). "The Effects of Overlearning and Distributed Practise on the Retention of Mathematics Knowledge". Applied Cognitive Psychology 20: 1209-1224. doi:10.1002/acp.1266.
- Rohrer, Doug; Taylor, K., Pashler, H., Wixted, J. T., & Cepeda, N. J. (2004). "The Effect of Overlearning on Long-Term Retention". Applied Cognitive Psychology 19: 361-374. doi:10.1002/acp.1083.