The Leitner System — Flashcard Boxes and Scheduled Review, Explained

From Tukuk-OS Wiki, the free knowledge base.

Sebastian Leitner's 1972 method turns a shoebox and rubber bands into a spaced-repetition machine: five boxes, fixed review intervals, and one rule — a card you get right graduates, a card you get wrong goes back to box one.

Definition

The Leitner system is a flashcard scheduling method in which cards move between a small number of boxes (traditionally five), each box carrying a fixed review interval. Newly learned cards sit in box 1 and are reviewed daily; a card answered correctly is promoted to the next box and its interval lengthens — box 2 every two days, box 3 weekly, box 4 biweekly, box 5 monthly, in the most common scheme. A card answered incorrectly is demoted back to box 1 regardless of where it sits. The physical version needs only index cards, a divided box and rubber bands; the algorithm fits on an index card because it is arithmetic, not software.

Origin

The system is credited to the German science journalist and author Sebastian Leitner, who described it in his 1972 book So lernt man lernen ("How to Learn to Learn"), building on the spaced-review idea that traces back to Hermann Ebbinghaus's forgetting-curve experiments of the 1880s. Leitner's contribution was the simplification: rather than asking learners to compute personal intervals, the box is the interval — a physical queue that decides, by which compartment a card rests in, when it is next due. The scheme spread through language-learning and medical-revision communities and is the direct ancestor of every digital flashcard scheduler that moves cards between "young" and "mature" piles.

How the scheduling works — and what it gets right

Two properties make the method work. First, review load follows difficulty: hard cards are seen daily because they keep returning to box 1, while a card in box 5 has already survived four spaced recalls and costs one look a month. The deck self-sorts; the learner does not choose what to review, the boxes do. Second, every review is a retrieval attempt — the card is answered before it is flipped — which is what separates the system from re-reading notes (see the companion essay on active recall).

The demotion rule looks harsh and is the point: a lapse on a box-4 card means the memory was not as stable as the schedule assumed, so the card returns to the short interval exactly as its performance claims. The known weakness is coarseness — five boxes give five intervals, so a card is either failing or about to be trusted; personal spaced-repetition algorithms add fine-grained grading between those poles.

Relationship to modern spaced-repetition software

Digital successors — from SuperMemo's SM-2 (1987) to Anki and its descendants — generalise Leitner in one specific way: instead of fixed box intervals, each card carries a computed interval and an ease factor updated by the difficulty grade you assign. The skeleton is unchanged: review due cards, lengthen the interval on success, shorten it on failure. A browser study tool implements the Leitner logic directly without any boxes: a flashcards generator separates "knew it" from "later" on every revealed card, so the later pile is box 1 — run the deck again tomorrow and the schedule re-forms itself; planning the review dates around an exam is the job of the spaced study planner. For a printable version of the same idea, the Cornell notes generator cue column serves as the recall half of the page.

Practical notes

The method assumes cards are atomic — one fact per card, question and answer on opposite faces — because a card mixing one known and one unknown half can neither be honestly promoted nor honestly demoted. It also assumes the deck is written by its user: the boxes schedule review of your formulations, and a borrowed deck imports someone else's encoding, which is why the essays recommend building cards from your own notes, one line per card.

See also

• Pomodoro Technique — Definition, Origin and How the 25-Minute Timer Works
• Spaced-Repetition Algorithms — How SM-2 and Its Successors Compute the Next Interval
• The Forgetting Curve and How Spaced Repetition Beats It
• Memory Tricks That Actually Work — Acronyms, Chunking and the Ridiculous Image
• Active Recall — Why Testing Yourself Beats Re-Reading, Every Single Time
• What Sleep Debt Actually Does to Your Brain and Body