Most students know the pattern: an all-nighter before the exam, a decent grade, and almost nothing left three weeks later. Spaced repetition is the best-documented answer to that problem. The idea fits in one sentence: review the same material several times, leaving more and more time between reviews. This article walks through what the research shows, with the numbers and the limits. If you want to apply it without working out dates yourself, Mindlet’s AI flashcards schedule your reviews for you.
What is spaced repetition?
There are two ways to spend three hours on a chapter: three hours in a row the night before, or one hour today, one in three days and one in ten days. Same time, different outcome. Researchers call this the spacing effect: for the same amount of study, spreading it out leads to more and longer-lasting memory than cramming it together.
Spaced repetition adds one more idea: the gaps grow. After the first review you wait a day, then three, then a week, then a month. Each review lands just as you are starting to forget.
What do the major studies say?
The spacing effect is one of the most reliable results in psychology. These are the studies worth knowing.
| Study | What it measured | What it found |
|---|---|---|
| Cepeda, Pashler, Vul, Wixted and Rohrer (2006), Psychological Bulletin 132(3) | Meta-analysis of 839 assessments, 317 experiments, 184 articles | Spaced study beats massed study; the best gap grows with the delay before the test |
| Cepeda, Vul, Rohrer, Wixted and Pashler (2008), Psychological Science 19(11) | Over 1,350 participants, tests up to a year later | Best gap is about 20 to 40 percent of a one-week delay, 5 to 10 percent of a one-year delay |
| Kornell (2009), Applied Cognitive Psychology 23(9) | Vocabulary learned with flashcards | Spacing beat massing for 90 percent of participants, yet 72 percent believed the opposite after the first session |
| Dunlosky, Rawson, Marsh, Nathan and Willingham (2013), Psychological Science in the Public Interest 14(1) | Review of ten common study techniques | Only two rated high utility: spaced practice and practice testing |
| Lindsey, Shroyer, Pashler and Mozer (2014), Psychological Science 25(3) | Personalized review scheduled by a model, in a real classroom, over a semester | 16.5 percent better retention than massed study |
| Donoghue and Hattie (2021), Frontiers in Education 6 | Meta-analysis of 242 studies, 169,179 participants | Spaced practice ranked first of ten techniques (effect size d = 0.85), ahead of practice testing (0.74) and well ahead of highlighting and summarizing (0.44) |
Sources: Cepeda et al., 2006; Cepeda et al., 2008; Kornell, 2009; Dunlosky et al., 2013; Lindsey et al., 2014; Donoghue and Hattie, 2021.
Kornell’s result deserves a second look. Nine people out of ten learned more by spacing, and most of them were convinced the opposite was true. Cramming feels fluent: the answer comes back quickly because you just read it. That fluency fades within days, which is why students keep cramming even though the evidence has been around for over a century.
Why does spacing work?
When you review something right after learning it, it is still in your head. Retrieving it takes no effort, and effortless retrieval does little for memory. When you review it a few days later, you have to dig for it. That harder retrieval strengthens the memory.
Elizabeth and Robert Bjork call this a desirable difficulty: a condition that slows learning in the moment but makes it last (Bjork and Bjork, 2011, in Psychology and the Real World). Spacing is the textbook example. It works best when every review is a real attempt to recall, not a reread. That is where spacing meets active recall.
What does it do for your studying?
- More retention for the same time. This is the core finding of the meta-analyses above.
- No more starting from scratch. What you know well comes back less and less often, so your time goes to what is still shaky.
- You find your gaps early. Every spaced review is a small test. You discover what you do not know in November, not the night before the exam.
- One routine for every deadline. Next week’s quiz and the final in June use the same habit; only the intervals change.
What are the limits?
A strong method is not a magic one. Three honest caveats:
- Most studies measure facts. Donoghue and Hattie point this out themselves: most experiments use vocabulary, definitions and dates. Effects on deep understanding of complex reasoning are less documented. Spacing helps you keep a proof in memory; it does not make you understand it.
- You need time ahead of you. Spacing means starting early. With a week left you can still space over a few days (see our one-week exam study plan), but the gain is smaller than with three months.
- Problem solving needs more. In math or physics, mixing problem types (interleaving) matters as much as spacing them. In Rohrer and Taylor (2007), Instructional Science 35(6), mixed practice clearly outperformed blocked practice on a test one week later.
How do you apply it without losing your evenings?
There are three ways to space your reviews, from manual to automatic.
| Method | How it works | Strength | Weakness |
|---|---|---|---|
| Fixed schedule | Review on day 1, 3, 7, 15, 30 | Simple, works on paper | Same intervals for everything, known or not |
| Leitner boxes | A card you get right moves up a box; a miss sends it back | Easy cards come back less | You manage the boxes by hand |
| Algorithm | A model estimates how likely you are to remember each card | Intervals tuned per card and per person | Requires an app |
Modern schedulers model memory card by card. FSRS, for instance, grew out of a paper published in 2022 (Ye, Su and Cao, KDD ‘22, pp. 4381-4390). It estimates how stable each memory is and how likely you are to recall it, then schedules the next review for when that likelihood drops below a target.
Mindlet uses FSRS-6. By default it aims for a 90 percent chance that you remember a card when it comes up, and it re-tunes its parameters every week from your own answers. You add your course material, the AI turns it into cards, and the app shows you each day what is due. One thing to know before you start: the Mindlet app interface is in French, but it can generate your cards in nine languages, English included. Our spaced repetition guide explains the method in more depth.
Without an app, one rule is enough to get started: whenever you review something, close your notes and try to say it before you check. If it came easily, double the interval. If it was a struggle, look at it again tomorrow. Our guide on how to make good flashcards shows how to write cards that work this way.
FAQ
Frequently asked questions
Does spaced repetition really work?
Yes. It is one of the most replicated findings in learning research. Cepeda et al. (2006) pooled 839 assessments from 317 experiments, and the review by Dunlosky et al. (2013) rated spaced practice as one of only two high-utility techniques, alongside practice testing.
How long should I wait between reviews?
It depends on when you need to remember. In Cepeda et al. (2008), the best gap was roughly 20 to 40 percent of the time until a test one week away, and 5 to 10 percent for a test one year away. The further away the exam, the further apart your reviews can be.
Does spaced repetition work for math?
It helps you retain definitions, formulas and methods. For problem solving, research also supports interleaving: mixing problem types instead of practicing them in blocks. Rohrer and Taylor (2007) found a clear advantage for mixed practice on a test one week later.
Do I need an app for spaced repetition?
No. A Leitner box or a simple review calendar works. An app calculates the dates for you and adjusts them to your answers, which removes the planning effort. Mindlet does this with a scheduler that learns how fast you forget each card.
