Spaced Repetition: The Complete Guide

The Forgetting Curve

Hermann Ebbinghaus used self-experiments with memorised syllables to study forgetting over time. His results helped establish that forgetting is often rapid after initial learning, but they do not provide a universal percentage that applies to every student or subject.

Later spacing research shows that distributing practice across time can improve delayed retention compared with concentrating the same practice into one session. The size and duration of that benefit depend on the material, retrieval quality, feedback and the delay before the final test.

How Spaced Repetition Works

The core idea is to revisit information after a delay, when retrieval takes some effort but remains productive. There is no observable universal optimal moment; a practical scheduler estimates the next interval from your previous response and adjusts it over time.

An illustrative starting schedule might look like this:

• Day 1: First exposure — learn the material.
• Day 2: First review (1 day later).
• Day 5: Second review (3 days after that).
• Day 12: Third review (7 days after that).
• Day 25: Fourth review (15 days after that).
• Day 55: Fifth review (30 days after that).

Treat these intervals as a starting point, not a promise. Shorten the next interval when recall fails, lengthen it after confident recall, and continue occasional maintenance when the material still matters.

The SM-2 Algorithm

Piotr Woźniak’s SuperMemo work produced the influential SM-2 algorithm, which helped popularise adaptive flashcard scheduling. Modern systems use different implementations, but many retain the basic idea of updating an interval from the learner’s response.

SM-2 works with three simple parameters:

• Interval: How many days until the next review.
• Ease Factor: How easy this card is for you (higher = longer intervals).
• Repetitions: How many times you have successfully recalled this card.

After each review, you rate your performance:

• Known (instant recall): The interval increases and the card moves further into the future.
• Hard (recalled with effort): The interval grows more cautiously so the card returns sooner.
• Review (could not recall): The interval becomes short and the card returns again soon.

This three-level rating system is simpler than the original SM-2 (which used six grades) but captures the essential distinction: instant recall, struggling recall, and no recall. The algorithm adjusts the schedule based on your actual performance, not a fixed calendar.

Why Spaced Repetition Is So Effective

Three cognitive mechanisms work together:

1. Desirable Difficulty: When recall feels slightly effortful, your brain strengthens the memory trace more than when recall is easy. Reviewing just before you forget creates optimal difficulty.
2. Consolidation During Sleep: Each sleep cycle consolidates and reorganises memories. Spacing reviews across multiple nights gives your brain more consolidation cycles.
3. Contextual Variation: Recalling information at different times, in different moods, and in different surroundings creates multiple retrieval paths — making the memory more robust.

Practical Implementation

You do not need to calculate every interval manually. Several approaches can manage the schedule:

• Physical flashcards + Leitner boxes: Five boxes. Cards start in box 1. If you get it right, it moves to the next box. Get it wrong, it goes back to box 1. You review box 1 daily, box 2 every 3 days, box 3 weekly, etc.
• Digital flashcard apps: Tools like Knowyio update review timing from your recall rating. The schedule reduces manual date tracking, but honest ratings and accurate cards still matter.
• Spreadsheet method: For manual control, keep columns for Topic, Last Reviewed, Interval and Next Review. Sort by Next Review to build each day’s study list.

Common Mistakes to Avoid

• Letting the queue grow unchecked: Missing a day does not invalidate the method, but a large backlog can make the routine hard to resume. Reduce new cards until the queue is manageable.
• Over-rating yourself: If you reveal the answer before producing it, mark the card for an earlier return instead of treating recognition as recall.
• Too many new cards per day: Adding 100 new cards daily creates an impossible review load within a week. Start with 10-20 new cards per day and adjust based on your review time.
• Over-reviewing secure material: Cards recalled reliably may need less frequent review. Keep more time for shorter-interval cards and exam-style application.

Getting Started Today

Start with one subject and a small set of focused cards. Use the example intervals only as an initial schedule, rate recall honestly and compare delayed performance with the first session. Add new cards slowly enough that scheduled reviews remain manageable.

Create your first AI-powered flashcard deck on Knowyio — paste your notes, pick your exam, and let the system generate recall questions with built-in spaced repetition scheduling.