Principle 02 · When to review
Spaced repetition: review at the right moment so you do not forget
8 minute read · Reviewed by [REVIEWER NAME] · Updated on
Short answer
Spaced repetition means reviewing content on several separate occasions, with gaps in between, instead of studying it all at once. Meta-analyses published since 2020 found a strong benefit from spacing out quiz sessions, and one of them, on language learning, showed that longer gaps beat shorter ones when you need to remember later.
What it is
In spaced repetition, the same content is reviewed several times, with gaps between reviews. It rests on the spacing effect: spreading study over time leads to better retention than packing it into a single session.
The idea is simple, but doing it by hand is hard: every person forgets at a different pace, and so does every piece of content.
What the research says
Spacing out quizzes gives a strong benefit
In a meta-analysis of 29 studies, answering the same questions in spaced sessions led to better retention than answering them back to back (g = 0.74, a benefit the authors describe as strong). A schedule of ever longer gaps was no better than equal gaps (g = 0.034), although the advantage of expanding gaps grew with the number of times each item was tested.
Latimier, Peyre and Ramus (2021), Educational Psychology Review.
Longer gaps for remembering later
In a meta-analysis of 48 experiments on second language learning, with 3,411 participants, spaced practice had a medium to large effect. Shorter gaps were as effective as longer ones on an immediate test, but less effective on a delayed test. Equal and expanding gaps were equivalent.
Kim and Webb (2022), Language Learning.
It works, but is underused
A review of research on spacing and retrieval practice, with an emphasis on real educational settings, describes how both strategies improve learning across domains and ages. The authors note that they remain underused and point to possible causes: mistaken beliefs about how learning works, lack of awareness of these strategies and the fact that they feel counter-intuitive.
Carpenter, Pan and Butler (2022), Nature Reviews Psychology.
Scheduling reviews with a memory model
Researchers at the language app MaiMemo published an algorithm that schedules each learner’s reviews so as to minimise the cost of memorising. The algorithm is used in that app.
Ye, Su and Cao (2022), KDD.
The g value measures the difference between the two groups in standard deviations: the larger it is, the bigger the advantage for the group that spaced its practice.
Where the idea comes from. A 2006 synthesis of 317 experiments from 184 articles had already found that studying on two separate occasions beats studying once (Cepeda and colleagues, 2006). In 2008 the same team showed that the optimal gap between sessions grows with the retention interval: about 20 to 40% of that interval for a test one week later, about 5 to 10% for a test one year later (Cepeda and colleagues, 2008). In a review of ten learning techniques, distributed practice was one of only two rated as highly useful (Dunlosky and colleagues, 2013).
Why it works
- Each review takes some effort. Recalling something that has started to fade is harder than repeating something you just saw, and conditions that make practice more demanding can improve long-term retention (Bjork, 1994). See desirable difficulties.
- The gap adapts to the goal. To remember for months, the optimal gap between reviews is longer than to remember for days (Cepeda and colleagues, 2008).
- It pairs well with retrieval practice. Reviewing by answering questions combines both effects. See retrieval practice.
The memory half-life model
The half-life of a memory is the time after which the probability of recalling it drops to half. From each person’s history of answers to each question, that half-life can be estimated and used to decide when the question should come back (Settles and Meeder, 2016).
How the game applies it
Academia AtlasLaunch 2026
2Level4 more correct answers to level up and 9 to the next medal
- Every participant has a memory half-life model for every question.
- A correct answer lengthens the gap before the question returns; a wrong one shortens it so the question comes back soon.
- The gap does not follow a fixed calendar: it adapts to each person’s answers. The meta-analyses found no advantage for a fixed schedule of expanding gaps over equal gaps; what they showed is the benefit of spacing, and of spacing further when you need to remember for longer.
- Once the estimated memory passes seven days, the question counts as mastered and rarely appears.
- We explain the full life cycle of a question in how it works.
References
- Carpenter, S. K., Pan, S. C. & Butler, A. C. (2022). The science of effective learning with spacing and retrieval practice. Nature Reviews Psychology, 1(9), 496-511. doi:10.1038/s44159-022-00089-1
- Kim, S. K. & Webb, S. (2022). The effects of spaced practice on second language learning: a meta-analysis. Language Learning, 72(1), 269-319. doi:10.1111/lang.12479
- 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 (KDD 2022), 4381-4390. doi:10.1145/3534678.3539081
- Latimier, A., Peyre, H. & Ramus, F. (2021). A meta-analytic review of the benefit of spacing out retrieval practice episodes on retention. Educational Psychology Review, 33(3), 959-987. doi:10.1007/s10648-020-09572-8
- Settles, B. & Meeder, B. (2016). A trainable spaced repetition model for language learning. Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (ACL 2016), 1848-1858. doi:10.18653/v1/P16-1174
- 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
- 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
- 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
- Bjork, R. A. (1994). Memory and metamemory considerations in the training of human beings. In J. Metcalfe & A. Shimamura (Eds.), Metacognition: knowing about knowing. MIT Press, pp. 185-205.