Spaced repetition

What is spaced repetition?

Spaced repetition is a learning technique in which material is reviewed at deliberately increasing intervals rather than in one concentrated session. Instead of reading a fact ten times in an hour, a learner reviews it today, again in a few days, then after a week, then after a month, with each successful recall pushing the next review further into the future. The method is most commonly associated with flashcards, but the underlying principle applies to any kind of information that needs to be retained over the long term.

The technique works by timing each review to occur roughly when the information would otherwise start to be forgotten. Reviewing too early wastes effort on something already well known, while reviewing too late means relearning from scratch. By targeting that window, spaced repetition strengthens a memory at a point where the strengthening has a large effect, producing durable retention from a comparatively small number of reviews.

Spaced repetition is closely tied to what the research literature calls distributed practice. The benefit of distributing practice over time, known as the spacing effect, is one of the best-replicated findings in the science of learning, replicated across vocabulary, mathematics, medicine, and procedural skills.

• Core idea: review at expanding intervals, not all at once.
• Goal: catch each item roughly when it would start to be forgotten.
• Closely related to distributed practice in cognitive psychology.
• Applies far beyond flashcards to any long-term retention task.

The spacing effect and the science of intervals

The foundation of spaced repetition is the spacing effect: information is retained more effectively when study sessions are spread out over time than when they are massed together. The effect traces back to Hermann Ebbinghaus, who from roughly 1880 to 1885 ran memorization experiments on himself using nonsense syllables and published his results in 1885 in Über das Gedächtnis. His work produced the now-famous forgetting curve, which shows recall dropping sharply soon after learning and then leveling off.

Each successful review tends to slow later forgetting, which is part of why intervals can keep growing. A landmark meta-analysis by Cepeda, Pashler, Vul, Wixted, and Rohrer, published in Psychological Bulletin in 2006, drew on 839 assessments across 317 experiments located in 184 articles and found that spaced practice reliably outperformed massed practice. A key nuance from that work is that the optimal gap between study sessions grows with how long you need to remember: the longer the target retention interval, the longer the ideal spacing between reviews.

• Spacing effect: distributed study beats concentrated study for retention.
• Ebbinghaus mapped the forgetting curve in self-experiments around 1880 to 1885.
• Cepeda et al. (2006) confirmed the effect across 317 experiments in 184 articles.
• Optimal interval scales with how long you need to retain the material.

Spaced repetition vs cramming and massed practice

Massed practice, commonly known as cramming, packs all study into a single block close to a deadline. It can produce a convincing feeling of fluency because the material is fresh, but that fluency is misleading: massed practice favors short-term performance and fades quickly. The harder, partly forgotten review that spacing creates is an example of a desirable difficulty, the term Robert Bjork coined for study conditions that feel harder in the moment yet produce stronger long-term learning.

Spaced repetition trades short-term comfort for long-term durability. Because each review is harder when the material has partly decayed, retrieval is more effortful, and that effort drives stronger encoding. The practical payoff is efficiency: comparable long-term retention can be reached with substantially less total study time when reviews are spaced rather than crammed.

• Cramming: high short-term recall, rapid forgetting afterward.
• Spaced repetition: slightly harder reviews, much stronger retention.
• Spacing reaches comparable retention with less total study time.
• Best for cumulative subjects, exams weeks away, and lasting knowledge.

Spaced repetition vs active recall, and how they combine

Active recall, also called retrieval practice or the testing effect, is the act of pulling information out of memory rather than passively rereading it. Closing the book and trying to answer a question tends to strengthen memory far more than reviewing the answer again. The testing effect and the spacing effect are distinct phenomena: one concerns how you study (retrieving versus rereading), the other concerns when you study (spaced versus massed).

The two combine powerfully. Spaced repetition answers the timing question, while active recall answers the method question, and well-designed flashcard systems use both at once: each scheduled review is itself a retrieval attempt. Reviewing by rereading on a spaced schedule captures only part of the benefit; reviewing by actively recalling on a spaced schedule captures more of the compounding effect, which is why flashcards, practice questions, and self-quizzing are the dominant formats.

• Active recall = retrieving from memory; the timing-independent lever.
• Spaced repetition = the timing lever; review at growing intervals.
• Combined, scheduled retrieval tends to beat either technique alone.
• Flashcards are popular because they bundle both by design.

Algorithms: Leitner, SM-2 and modern schedulers

An early widely used scheduling method was the Leitner system, introduced by German science writer Sebastian Leitner in his 1972 book So lernt man lernen. Cards move through a series of physical boxes: a card answered correctly is promoted to the next box and reviewed less often, while a card answered incorrectly drops back to the first box and is seen frequently again. It is simple, requires no computer, and still underlies the intuition behind every digital system.

Software made the intervals adaptive. The SM-2 algorithm, created by Piotr Wozniak for SuperMemo and first shipped in SuperMemo 1.0 for DOS in 1987, tracks an ease factor per card (starting at 2.5) that lengthens or shortens future intervals based on how well each review goes. SM-2 remains the classic algorithm and powered Anki for many years. More recently, the open-source Free Spaced Repetition Scheduler (FSRS) fits a statistical model of memory to a learner's own review history and schedules each card by its predicted probability of recall. Anki integrated FSRS as a built-in scheduling option in version 23.10, released on November 1, 2023.

AI-based memory tools and second-brain applications increasingly borrow these scheduling ideas to resurface saved notes and facts at spaced intervals rather than leaving them buried.

def sm2(q, n, ef, interval):
    if q < 3:                 # failed: restart the card
        return 0, ef, 1
    if n == 0:
        interval = 1
    elif n == 1:
        interval = 6
    else:
        interval = round(interval * ef)
    ef = ef + (0.1 - (5 - q) * (0.08 + (5 - q) * 0.02))
    ef = max(1.3, ef)
    return n + 1, ef, interval

• Leitner system (1972): physical boxes with escalating review frequencies.
• SM-2 (Wozniak, 1987): per-card ease factor adapts each interval.
• FSRS: models recall probability from a learner's own review data.
• Anki added built-in FSRS scheduling in version 23.10 (November 2023).

Beyond flashcards: reminders and surfacing in real life

Spaced repetition need not involve formal decks. The same principle applies whenever revisiting information at growing intervals matters: rehearsing a presentation over several days, periodically reviewing meeting notes, or scheduling reminders to revisit a saved article. The mechanism is similar even when the trigger is a calendar reminder rather than a flashcard queue.

This has made spacing a natural design pattern for note-taking and memory software. Rather than relying on a learner to manually re-open old material, a system can surface it on a spaced schedule, turning a passive archive into something that is actively reinforced. The strongest real-world use of spaced repetition tends to pair it with active recall: instead of merely re-reading a resurfaced note, the learner tries to recall its contents first, then checks.

• Spacing applies to rehearsal, review reminders, and resurfaced notes.
• Scheduled reminders can deliver the spacing effect without decks.
• Pair resurfacing with a recall attempt for the strongest effect.