Koa

Spaced Repetition: How to Remember What You Study

The science behind long-term retention - and how to build a spaced review system that actually works.

What Is Spaced Repetition?

Spaced repetition is a study method where you review material at increasing intervals over time. Instead of rereading your notes five times the night before an exam, you review them once after one day, again after three days, again after a week, and so on. Each successful review pushes the next one further into the future.

The idea is simple: you review material right before you're about to forget it. This forces your brain to actively reconstruct the memory, which strengthens it. Over time, the gaps between reviews grow longer, and the memory becomes more durable.

Cramming is the opposite approach. You mass all your practice into a single session, which creates a strong feeling of familiarity - but that familiarity fades fast. Within a week, most of what you crammed is gone. Spaced repetition trades that short-term intensity for lasting retention.

A Quick Example

Say you learn 30 terms for your biology midterm. With cramming, you'd drill all 30 the night before. With spaced repetition, you'd learn them on Monday, review on Tuesday, review again on Thursday, then once more the following Wednesday. By exam day, you've spent less total time - and you remember more.

The Science: Ebbinghaus and the Forgetting Curve

In 1885, German psychologist Hermann Ebbinghaus published the first experimental study of memory and forgetting. He memorized lists of nonsense syllables (like "DAX," "BUP," "ZOL") and tested himself at various intervals to see how quickly he forgot them.

What he found is now called the forgetting curve: memory decays exponentially after learning. Without any review, you lose roughly 50% of newly learned material within the first hour, about 70% within 24 hours, and nearly 90% within a week.

The key insight is that each review increases the stability (S) of the memory. After the first review, the curve still decays - but more slowly. After the second review, it decays even more slowly. By the fourth or fifth review, the information can persist for months or years with minimal effort.

This is the spacing effect: distributing practice over time produces stronger, more durable memories than concentrating it. Ebbinghaus discovered it in 1885, and it has been replicated in hundreds of studies since. A 2006 meta-analysis by Cepeda et al., reviewing 254 studies with over 14,000 participants, confirmed the spacing effect across all age groups, material types, and retention intervals.

How Spaced Repetition Works

The Basic Schedule

A standard spaced repetition schedule uses increasing intervals. After you first learn something, you review it at these approximate gaps:

These intervals aren't fixed rules. They're starting points. The real power of spaced repetition comes from adjusting the intervals based on how well you know each item.

The SM-2 Algorithm in Plain English

The SM-2 algorithm, created by Piotr Wozniak in 1987, is the most widely used spaced repetition scheduling system. It's the engine behind Anki, SuperMemo, and many other tools - including Koa. Here's how it works:

1. After each review, rate yourself from 0 to 5. 0 means complete blackout (you had no idea). 5 means perfect recall with no hesitation. Most useful ratings are 2 (wrong but recognized the answer), 3 (correct with difficulty), 4 (correct with some thought), and 5 (instant recall).
2. The algorithm adjusts an "easiness factor" (EF) for each item. Items you consistently rate 4-5 get a higher EF, meaning longer intervals. Items you rate 0-2 get a lower EF, meaning shorter intervals. EF starts at 2.5 and doesn't drop below 1.3.
3. The next interval is calculated as: previous interval multiplied by the easiness factor. So if your last interval was 7 days and your EF is 2.5, the next review is in 17.5 days (roughly 18).
4. If you fail an item (rating 0-1), the interval resets to 1 day. You essentially restart the learning process for that item, though the EF adjusts downward so future intervals grow more slowly.

You don't need to memorize this formula. The point is: easy items get reviewed less often, hard items get reviewed more often, and the whole system adapts to you automatically.

Spaced Repetition vs. Cramming

Cramming works - for about 24 hours. Massed practice creates a strong short-term memory trace that lets you pass a test the next morning. But research consistently shows that this memory decays far faster than spaced memories.

The Evidence

• Rohrer & Taylor (2006) found that students who spaced their practice over two sessions one week apart scored 64% better on a test given one week later compared to students who crammed in a single session.
• Karpicke & Roediger (2008) showed that repeated testing with spacing produced 150% better recall after one week than repeated studying without testing.
• Cepeda et al. (2006) meta-analyzed 254 studies: the spacing effect appeared in every single one, regardless of material type, age group, or retention interval.

Why Cramming Feels Effective

Cramming creates an illusion of competence. When you reread your notes, the material feels familiar. You recognize the concepts, and that recognition feels like understanding. But recognition and recall are different cognitive processes. Recognizing an answer on a multiple-choice test is easier than producing it from memory on a written exam.

Spaced repetition feels harder because you're deliberately reviewing when the material is starting to fade. That struggle is a feature, not a bug - it's what makes the memory stronger. Psychologists call this desirable difficulty.

For a midterm you need to pass tomorrow, cramming might be your only option. But for final exams, cumulative courses, professional certifications, or anything you want to retain after the course ends - spaced repetition is the clear winner.

How to Build a Spaced Review System

Step 1: Turn Your Material into Reviewable Units

Spaced repetition works best with discrete items - individual facts, concepts, or procedures that you can test yourself on. Flashcards are the classic format, but the unit could also be a practice problem, a definition, or a concept explanation.

Good flashcards follow the minimum information principle: each card tests one thing. Instead of a card that asks "Explain the stages of mitosis," make four separate cards - one per stage. Smaller cards are faster to review, easier to rate accurately, and allow the algorithm to track each fact independently.

Step 2: Set Up a Review Schedule

You need a system that tracks when each item is due for review. You can do this manually with a spreadsheet, a physical card box (the Leitner system), or with software that handles the scheduling automatically.

If doing it manually, create five groups:

1. Box 1 (Daily) - New cards and cards you got wrong
2. Box 2 (Every 3 days) - Cards you got right once
3. Box 3 (Weekly) - Cards you got right twice
4. Box 4 (Biweekly) - Cards you got right three times
5. Box 5 (Monthly) - Cards you've mastered

Get a card right, it moves up a box. Get it wrong, it goes back to Box 1. This is the Leitner system, and it's a manual approximation of what SM-2 does automatically.

Step 3: Rate Difficulty Honestly

After each review, rate how hard the recall was. This is what drives the entire system. Two guidelines:

• If you looked at the card and immediately knew the answer, rate it easy (4-5)
• If you got it right but had to think about it, rate it medium (3)
• If you got it wrong or couldn't recall it at all, rate it hard (0-2) and review it again soon

Step 4: Integrate with Your Course Schedule

The best time to create cards is right after class, while the material is fresh. Then your first review happens naturally within 24 hours. As the semester progresses, your daily review load stays manageable because older cards are on longer intervals.

Before an exam, you don't need to cram - you've been reviewing all along. Your pre-exam study shifts from re-learning to sharpening: focusing on the items the algorithm has flagged as weak.

Combining with Active Recall

Spaced repetition tells you when to review. Active recall tells you how. Together, they form the two most evidence-backed study techniques in cognitive science.

Active recall means testing yourself - pulling information out of your memory rather than passively rereading it. When you look at the front of a flashcard and try to produce the answer before flipping it over, that's active recall. When you close your textbook and try to write down everything you remember from the chapter, that's active recall.

The combination is powerful because each technique amplifies the other. Active recall makes each review session more effective (retrieval strengthens memory more than re-exposure). Spacing makes sure those effective sessions happen at the optimal time (right before you'd forget).

For a deeper dive into retrieval practice techniques, see our guide to the active recall study method.

Tools for Spaced Repetition

Anki

Anki is the most well-known spaced repetition tool. It's free (desktop and Android), open-source, and uses a modified SM-2 algorithm. You create your own decks of flashcards, and Anki schedules your daily reviews. It's powerful but has a steep learning curve - the interface is utilitarian, and setting up cards with the right formatting takes time.

Anki works well if you're willing to invest the setup time and you want maximum control over the algorithm's parameters. Medical students, in particular, have built enormous shared Anki decks (like AnKing) that cover entire curricula.

Koa

Koa takes a different approach. Instead of making you create every card from scratch, Koa connects to your university's LMS (Canvas, Moodle, Brightspace) and uses AI to generate review questions from your actual course material - lecture notes, assignments, syllabi. The SM-2 algorithm handles scheduling, but you spend your time reviewing, not building cards.

Koa also goes beyond simple flashcards. It supports multiple question formats (multiple choice, short answer, concept explanations, problem solving) and adapts the format based on the material. Reviewing a calculus course? You get practice problems. Reviewing a psychology course? You get concept-application questions.

Getting Started This Week

You don't need to overhaul your study system overnight. Here's a practical plan to start using spaced repetition this week:

Day 1: Pick One Course

Choose the course where you most need better retention - something with lots of facts, terms, or concepts to remember. Don't try to set up spaced repetition for all your courses at once.

Day 2: Create Your First 20 Cards

Go through your most recent lecture notes and create 20 flashcards covering the key concepts. Follow the minimum information principle: one fact per card. If you're using Koa, connect your course and let the AI generate them.

Day 3: First Review

Review all 20 cards. Rate each one honestly. Cards you got wrong go back to the top of the queue. Cards you got right are scheduled for review in 3 days.

Day 4-5: Add New Material

After your next lecture, add 10-15 new cards. Review any cards that are due from Day 3. Your total daily review time should be about 10-15 minutes at this point.

Day 6-7: Review and Adjust

By the end of the week, you'll have ~40-50 cards, with the oldest ones coming up for their second review. Notice which cards you keep getting wrong - these are the concepts you need to understand more deeply, not just memorize. Consider rewriting those cards to be clearer, or adding more context.

After two weeks of daily 15-minute reviews, spaced repetition becomes a habit. After a month, you'll have a growing library of material that you can recall on demand - without having to cram before every test.

Let the algorithm remember for you

Koa uses SM-2 spaced repetition to schedule your reviews automatically - so you study the right material at the right time.

Try Koa - It's Free

Frequently Asked Questions