The Leitner System

The 1972 cardboard-box spaced repetition system invented by German science journalist Sebastian Leitner — the direct ancestor of every modern flashcard app, including Neurako.

The Leitner System is a spaced repetition method using physical flashcards sorted into a series of boxes by familiarity. Cards you answer correctly move to a box reviewed less frequently; cards you miss return to the daily-review box. German science journalist Sebastian Leitner introduced it in his 1972 book So lernt man lernen ("How to Learn to Learn"). Every modern spaced repetition app, including Anki and Neurako, descends from this idea.

Key Takeaways

Leitner was a journalist, not a learning scientist — he built on Ebbinghaus's forgetting curve research with a practical at-home system - The classic version uses five boxes with intervals roughly 1, 2, 4, 7, 14 days; many simpler three-box adaptations exist - The core insight: spend more review time on the cards you don't yet know, less on the cards you've mastered - Modern algorithms like SM-2 and FSRS apply the same principle with per-card adaptive intervals instead of fixed box schedules - You can run Leitner with literal index cards today — it costs nothing and works

Origin

Sebastian Leitner was a German science journalist, not an academic researcher. In 1972 he published So lernt man lernen — translated as How to Learn to Learn — a popular book on study methods. The book was a synthesis of existing research on memory, including Ebbinghaus's late-1800s work on the forgetting curve, into practical advice anyone could apply at home with index cards and a divided box.

The system was widely adopted in German schools and spread internationally over the following decades. By the 1990s, it had become the dominant method for vocabulary study among serious language learners. When personal computers became common, the natural next step was to digitize the box-shuffling. The first widely-used spaced repetition software — SuperMemo, released in 1987 by Piotr Wozniak — formalized the underlying scheduling math, but the Leitner box was its conceptual ancestor.

The classic five-box version

Leitner's original system used five boxes:

Box 1 — "Every day." All new cards start here. Reviewed daily.
Box 2 — Reviewed every 2 days.
Box 3 — Reviewed every 4 days.
Box 4 — Reviewed every 7 days.
Box 5 — Reviewed every 14 days. The "mastered" box.

The rules:

Start every new card in Box 1. Review Box 1 daily.
When you answer a card correctly, move it forward one box. A correct Box 1 card moves to Box 2. A correct Box 5 card stays in Box 5 (or moves to a "retired" pile in some variants).
When you answer a card incorrectly, move it back to Box 1. This is the punishing step. A card you've nearly mastered in Box 5 falls all the way back to daily review if you fail it once.
Review each box on its scheduled day. Box 2 every other day, Box 3 every four days, and so on. Use a calendar or notebook to track which boxes are due.

The principle is concentrating review time on the cards you don't know. Easy cards drift into long-interval boxes and rarely consume study time. Hard cards bounce around the early boxes, getting frequent practice until they stabilize.

A simpler three-box version

Many practitioners use a simplified three-box variant that's easier to manage with physical cards:

Box 1 — Daily. New cards and cards you recently got wrong.
Box 2 — Every 3 days.
Box 3 — Every week (or longer).

The rules are the same: correct → forward, incorrect → back to Box 1. The reduced complexity makes the system easier to maintain without a digital scheduler.

Why it works

The Leitner System is a practical implementation of three well-evidenced principles:

The spacing effect. Cepeda, Pashler, Vul, Wixted, and Rohrer's 2006 meta-analysis of 254 studies confirmed that distributing study sessions across time produces stronger long-term retention than massed practice. The Leitner box schedule operationalizes spacing: cards you know better get longer gaps, cards you know less get shorter gaps.

Active retrieval. Every box review is a retrieval test. Roediger and Karpicke (2006) showed that retrieval strengthens memory more than restudy. Flipping a Leitner card is a retrieval event — the system is essentially structured testing practice.

Adaptive scheduling. Cards you struggle with get reviewed more; cards you've mastered get reviewed less. This matches the "desirable difficulty" principle (Bjork): productive practice happens at the edge of capability, not below it or beyond it.

Limitations of the original system

Leitner's box schedule uses fixed intervals — every card in Box 3 is reviewed on the same schedule, regardless of how the individual learner remembers that specific card. This causes two problems:

Lumpy review days. If you add 30 cards on the same day, all 30 will hit Box 2, Box 3, and so on in lockstep. Some review days are empty; others are crushing. Modern algorithms add a small random "fuzz" to intervals to spread the load.

One-size-fits-all intervals. A card you can recall easily after a week may sit in Box 3 with cards you barely remember. The box doesn't know that this specific card has stable memory and could go to Box 4 sooner.

The SM-2 algorithm (1987, used in early Anki) addressed this by calculating per-card intervals based on rating history. FSRS (Free Spaced Repetition Scheduler), released in the early 2020s, took it further — modeling each card's stability, difficulty, and current retrievability based on the actual review history. For a detailed comparison, see FSRS vs SM-2.

In Neurako

Neurako uses FSRS, the most advanced descendant of the Leitner principle. Where Leitner sorted cards into discrete boxes with fixed intervals, FSRS calculates the optimal next-review date for each card individually, targeting a 90% recall probability at review time. The underlying idea — review hard cards more, easy cards less — is identical. See The FSRS Algorithm for how it works in detail.

Should you use the original Leitner system today?

It depends on what you're optimizing for.

Use physical Leitner boxes when:

You're studying offline and want zero screen time
You're tutoring a child and want a tactile system
You're learning something small (under 200 cards) where the per-card scheduling precision doesn't matter much
You're a deliberate minimalist who wants to avoid app-based studying

Use a digital spaced repetition app when:

Your card collection is large (hundreds to thousands of cards)
You want statistical accuracy in scheduling
You want to add cards from your phone or sync across devices
You want AI-assisted card creation, audio support, or image attachments

Both work. The 1972 cardboard-box version is not obsolete — it's a fully functional spaced repetition system. Modern algorithms simply optimize the scheduling math while keeping the same core principle.

A note on history

It's worth noting that spaced repetition as a concept predates Leitner by nearly a century. Hermann Ebbinghaus described the forgetting curve in 1885 and explicitly suggested that distributed review could counteract memory decay. Cognitive psychologists in the 1960s and 1970s (Melton, Landauer, Bjork) explored the spacing effect in controlled experiments. Leitner's contribution was packaging existing knowledge as a system anyone could implement with cardboard and index cards.

Each major step in the history:

1885: Ebbinghaus describes the forgetting curve
1932: F. C. Bartlett's work on memory reconstruction
1960s–70s: Cognitive psychologists formalize the spacing effect (Melton, Landauer, Bjork)
1972: Leitner publishes the box system
1987: Wozniak releases SuperMemo with the SM-2 algorithm
2006: Anki releases, popularizing SM-2 widely
2022–: FSRS becomes widely available in modern flashcard apps, including native support in Anki 23.10+

The line from Ebbinghaus to Neurako runs through Leitner's cardboard box.

Spaced Repetition Explained

The full memory science behind why spaced review works.

FSRS vs SM-2

How the modern algorithm compares to the original SuperMemo approach.

The FSRS Algorithm

How Neurako's scheduler works in detail — the modern successor to the Leitner box.

Sources

Leitner, S. (1972). So lernt man lernen: Der Weg zum Erfolg [How to Learn to Learn: The Way to Success]. Herder.
Ebbinghaus, H. (1885/1913). Memory: A Contribution to Experimental Psychology (H. A. Ruger & C. E. Bussenius, Trans.). Teachers College, Columbia University.
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. https://pubmed.ncbi.nlm.nih.gov/16719566/
Roediger, H. L., & Karpicke, J. D. (2006). Test-enhanced learning: Taking memory tests improves long-term retention. Psychological Science, 17(3), 249–255. https://pubmed.ncbi.nlm.nih.gov/16507066/

Ready to turn this into a study ritual?

Start studying with Neurako