Browse the wiki

Spaced repetition

From the YapLearn Wiki · Memory technique · Updated 2026-10-28

Spaced repetition is a way of studying in which the same material is reviewed several times with gaps between reviews, and the gaps grow longer as the material becomes better known. It relies on the spacing effect: information studied in sessions spread out over time is remembered longer than the same amount of study crowded into one session [1][2]. In language learning it is used mostly for vocabulary, usually with flashcards on paper or in software such as SuperMemo and Anki [3].

The idea traces back to Hermann Ebbinghaus, who in 1885 published the first measurements of how memory fades over time [4]. Later research on hundreds of experiments showed that the best gap between reviews depends on how long the material must be remembered [1][5]. A 2013 review of ten common study techniques rated distributed (spaced) practice as one of only two techniques of "high utility", together with practice testing [6].

Contents

Key facts

  • In 1879 and 1880 Ebbinghaus learned lists of nonsense syllables and relearned them later; after 24 hours he still saved about one third of the original learning time, and after a month about one fifth [4].
  • A 2015 replication by Jaap Murre and Joeri Dros, with one subject who spent 70 hours learning and relearning lists, produced results similar to Ebbinghaus's curve [7].
  • A 2006 review found 839 assessments of distributed practice in 317 experiments from 184 articles [1].
  • In a study of more than 1,350 people, the best gap between study sessions was about 20 to 40 percent of the test delay for a 1-week test, and about 5 to 10 percent for a 1-year test [5].
  • In a 9-year study of 300 English and foreign-language word pairs, 13 review sessions spaced 56 days apart gave retention comparable to 26 sessions spaced 14 days apart [8].
  • The SM-2 algorithm, used in SuperMemo from 1987, sets the first interval at 1 day, the second at 6 days, and multiplies each later interval by an "easiness factor" that starts at 2.5 [9].
  • Anki's FSRS scheduler targets a desired retention of 90 percent by default [10].

Ebbinghaus and the forgetting curve

The original experiments

Hermann Ebbinghaus, a German psychologist, studied his own memory using series of meaningless syllables so that prior knowledge would not affect results. In 1879 and 1880 he ran 163 double tests: he learned eight series of 13 syllables until he could recite them twice without error, then relearned them after about 20 minutes, 1 hour, 9 hours, 1 day, 2 days, 6 days or 31 days [4].

Savings and the curve

He measured memory by savings: the share of the original learning time that was saved when relearning. One hour after learning, about half of the original work had to be repeated; after 24 hours about one third was still remembered, after 6 days about one fourth, and after a month about one fifth [4]. Plotted over time, these values form the forgetting curve: rapid loss at first, then slower loss. His book *Memory: A Contribution to Experimental Psychology* appeared in German in 1885 and in English in 1913 [4].

The 2015 replication

Murre and Dros repeated the experiment in 2015 with intervals of 20 minutes, 1 hour, 9 hours, 1 day, 2 days and 31 days. Their results matched the original closely, although the curve was not completely smooth and probably shows a jump upward starting at the 24-hour point [7].

The spacing effect in research

The spacing effect is the finding that spreading study over time improves long-term memory compared with massed study (repetitions in one block). Kang's 2016 review notes that hundreds of studies in cognitive and educational psychology show the effect, that it applies to memory, problem solving and transfer to new situations, and that adding tests to spaced practice increases the benefit [2].

The 2006 review

Nicholas Cepeda and colleagues analyzed 839 assessments of distributed practice and found that the gap between study sessions and the time until the test work together: the gap that produced the best retention grew as the retention interval grew [1].

The 2008 gap study

The 2008 study by Cepeda and colleagues taught more than 1,350 people a set of facts, gave a review after a gap of up to 3.5 months, and tested them up to a year later. Performance first rose and then slowly fell as the gap increased, tracing a "ridgeline" of optimal gaps [5].

Best gap between sessions in Cepeda et al. (2008)
Time until final testBest gap as share of that time
1 weekabout 20 to 40 percent [5]
1 yearabout 5 to 10 percent [5]

Gaps for short and long goals

In practical terms, a review about a day or two after first study suits a test a week away, while knowledge meant to last a year benefits from gaps of several weeks [5]. The authors concluded that the interaction between gap and test delay implies that many educational practices are highly inefficient [5].

Foreign vocabulary studies

The Pimsleur memory schedule

Paul Pimsleur, an applied linguist, proposed a fixed "memory schedule" for language learners in 1967, in which each recall comes after a longer interval than the one before. The schedule usually reproduced from the article runs 5 seconds, 25 seconds, 2 minutes, 10 minutes, 1 hour, 5 hours, 1 day, 5 days, 25 days, 4 months and 2 years*, each interval about five times the previous one [11].

The nine-year word pair study

In a 9-year study led by Harry Bahrick, four subjects learned and relearned 300 English and foreign-language word pairs in 13 or 26 sessions spaced 14, 28 or 56 days apart, and were tested 1, 2, 3 or 5 years after training ended. Longer gaps slowed learning slightly during training but produced substantially higher retention; 13 sessions at 56-day intervals matched 26 sessions at 14-day intervals [8]. The benefit held for both easy and difficult words [8].

Flashcard systems and algorithms

Spaced repetition became widespread through flashcard systems that decide when each card is due.

Paper systems

Leitner box

In 1972 Sebastian Leitner, a German author, popularized a paper method in which cards move to a later box after a correct answer and back after a mistake, so that well-known cards are reviewed less often. It gives only a rough schedule and handles material of varying difficulty poorly [3].

Software schedulers

SuperMemo and SM-2

Piotr Wozniak wrote the first SuperMemo program in December 1987 [9]. Its SM-2 algorithm works as follows:

  • Each answer gets a grade from 0 (complete blackout) to 5 (perfect response) [9].
  • Intervals grow from review to review [9]:
    • the first interval is 1 day;
    • the second interval is 6 days;
    • each later interval is the previous one times an easiness factor that starts at 2.5 and never falls below 1.3.

Anki

Anki is a free flashcard program built on active recall and spaced repetition. Its legacy scheduler is based on SM-2 [3][10].

FSRS

FSRS (Free Spaced Repetition Scheduler) is an alternative to SM-2 supported since Anki 23.10. It fits its parameters to the user's own review history and models each card with three values [10][12]:

  • difficulty;
  • stability;
  • retrievability, the probability of recall at a given moment.

Its developer, Jarrett Ye, based it on research at the vocabulary app MaiMemo, where a scheduler built from 220 million students' memory logs performed 12.6 percent better than earlier methods [13][12].

Comparison of schedulers

Main spaced repetition schedulers
SystemYearHow intervals are set
Pimsleur schedule1967Fixed, each about five times the last [11]
Leitner box1972Card moves between boxes [3]
SM-219871 day, 6 days, then times an easiness factor [9]
FSRS2020sFitted to the learner's review history [12]

Intervals in practice

Software schedulers grow intervals automatically, so a new word may be seen again after a day, then after several days, then after weeks or months [9][3]. FSRS lets the learner set a desired retention, the target probability of recall at review time; the Anki manual describes the default of 90 percent as offering a good balance of retention and workload [10]. Bahrick's results suggest that for long-term retention, fewer reviews at wider gaps can match more reviews at narrow gaps [8].

Because each review asks the learner to produce the answer, spaced repetition combines spacing with Retrieval practice, and Kang notes that the two together give larger gains than either alone [2].

Limitations

Spaced repetition strengthens memory for items that have already been met, but a flashcard shows a word out of context. Knowing a word well enough to understand it in speech and text requires meeting it many times in real use, which is the role of Comprehensible input. Kang also discusses barriers that have kept spaced practice from wide use in schools, despite its low cost [2]. Cards that are too hard or too easy waste time, so decks work best when new cards sit at the learner's Knowledge frontier.

See also

References

  1. Cepeda, N. J., Pashler, H., Vul, E., Wixted, J. T., Rohrer, D. (2006). Distributed practice in verbal recall tasks: A review and quantitative synthesis. Psychological Bulletin, 132(3), 354-380. doi:10.1037/0033-2909.132.3.354
  2. Kang, S. H. K. (2016). Spaced repetition promotes efficient and effective learning: Policy implications for instruction. Policy Insights from the Behavioral and Brain Sciences, 3(1), 12-19. doi:10.1177/2372732215624708
  3. Anki Manual (2026). Background. docs.ankiweb.net. docs.ankiweb.net/background.html
  4. Ebbinghaus, H. (trans. Ruger, H. A., Bussenius, C. E.) (1913). Memory: A Contribution to Experimental Psychology (1885; English translation 1913). Teachers College, Columbia University; Classics in the History of Psychology. psychclassics.yorku.ca/Ebbinghaus/index.htm
  5. Cepeda, N. J., Vul, E., Rohrer, D., Wixted, J. T., Pashler, H. (2008). Spacing effects in learning: A temporal ridgeline of optimal retention. Psychological Science, 19(11), 1095-1102. doi:10.1111/j.1467-9280.2008.02209.x
  6. Dunlosky, J., Rawson, K. A., Marsh, E. J., Nathan, M. J., Willingham, D. T. (2013). Improving students' learning with effective learning techniques: Promising directions from cognitive and educational psychology. Psychological Science in the Public Interest, 14(1), 4-58. doi:10.1177/1529100612453266
  7. Murre, J. M. J., Dros, J. (2015). Replication and analysis of Ebbinghaus' forgetting curve. PLOS ONE, 10(7), e0120644. doi:10.1371/journal.pone.0120644
  8. Bahrick, H. P., Bahrick, L. E., Bahrick, A. S., Bahrick, P. E. (1993). Maintenance of foreign language vocabulary and the spacing effect. Psychological Science, 4(5), 316-321. doi:10.1111/j.1467-9280.1993.tb00571.x
  9. Wozniak, P. A. (1990). Application of a computer to improve the results obtained in working with the SuperMemo method (SM-2 algorithm, from Optimization of learning, master's thesis). SuperMemo. super-memory.com/english/ol/sm2.htm
  10. Anki Manual (2026). Deck Options (FSRS). docs.ankiweb.net. docs.ankiweb.net/deck-options.html
  11. Pimsleur, P. (1967). A memory schedule. The Modern Language Journal, 51(2), 73-75. doi:10.1111/j.1540-4781.1967.tb06700.x
  12. Ye, J. (2026). ABC of FSRS (open-spaced-repetition wiki). GitHub, open-spaced-repetition/awesome-fsrs. github.com/open-spaced-repetition/awesome-fsrs/wiki/ABC-of-FSRS
  13. Ye, J., Su, J., Cao, Y. (2022). A stochastic shortest path algorithm for optimizing spaced repetition scheduling. Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 4381-4390. doi:10.1145/3534678.3539081

Tags: Memory and review · Vocabulary · Practice and drills

Spaced repetition | YapLearn Wiki