If you have ever used a flashcard app, you have probably encountered spaced repetition without knowing its name. The idea is simple: review information at increasing intervals, timed so that each review happens just before you would forget it. The results, in study after study, are dramatic compared to other study methods.
But most language apps implement it badly. Understanding why requires understanding the actual science first.
The forgetting curve
In 1885, German psychologist Hermann Ebbinghaus spent months memorising nonsense syllables and then testing how quickly he forgot them. He was the only subject in his own experiments, which is either very dedicated or very unusual depending on your perspective.
What he found became known as the forgetting curve: newly learned information decays rapidly at first, then more slowly. Without any review, most people forget roughly half of new information within a day and the majority within a week.
But here is the crucial finding: each time you successfully recall something, the forgetting curve resets — and it resets at a shallower angle. After your third successful recall, you might not need to review again for a month. After your fifth, maybe three months.
This is the mechanism that spaced repetition exploits. You are not trying to prevent forgetting — you are using the act of remembering to make future forgetting harder.
How the spacing effect works in practice
The modern understanding of spaced repetition comes from researchers like Sebastian Leitner, who in 1972 devised the Leitner box — a physical system of index cards divided into compartments based on how well you know each card.
The key principle: the difficulty of retrieval matters. When you struggle to remember something and then successfully recall it, your brain encodes it more deeply than if you reviewed it while it was fresh and easy. This is called the desirable difficulty effect.
This means the optimal review moment is not "as soon as possible" — it is "just before you would forget." Review too early and you are wasting effort on something you already know well. Review too late and it has already slipped away.
Modern spaced repetition systems (most famously Anki) use algorithms to calculate the optimal review interval for each item individually, adjusting based on how quickly or slowly you recall it.
Why language learning is a particularly good fit
Vocabulary is almost perfectly suited to spaced repetition for three reasons:
1. The items are discrete. Each word is its own unit. It is easy to test and easy to measure. You either produce the word or you do not.
2. The volume is large. Getting to conversational fluency in a language requires several thousand words. No one can memorise several thousand things through brute force. You need a system that manages the review schedule for you.
3. The payoff is immediate. Unlike learning, say, calculus — where you need many concepts before anything is useful — each vocabulary word you learn gives you an immediate small increase in your ability to communicate. This makes motivation easier to sustain.
Where most apps go wrong
The widespread adoption of spaced repetition in language apps is broadly a good thing. But many apps distort the technique in ways that undermine its effectiveness.
Streaks punish the forgetting curve, not you. The point of spaced repetition is that review intervals get longer as you learn something better. If you are reviewing a word every single day because your streak requires daily practice, you are reviewing it far too often. You are doing easy, low-value repetitions instead of the spaced, challenging ones that build long-term memory.
Daily reminders create the wrong habit. Real spaced repetition is demand-driven, not schedule-driven. You review items when they are due — which might be tomorrow, or in three weeks, or in two months. An app that nudges you to practice every day at 7pm is not implementing spaced repetition. It is implementing a daily ritual that happens to include some flashcards.
Gamification rewards time spent, not learning. XP points, level-ups, and leaderboards measure engagement. They do not measure whether you have actually learned anything. Worse, they can reward easy, comfortable review (items you already know well) over the effortful retrieval that actually builds memory.
What genuinely effective vocabulary practice looks like
Review when items are due, not when a notification tells you to. A well-designed spaced repetition system will accumulate a queue of items that are due for review. Work through that queue when you sit down to study — not on someone else's schedule.
Make retrieval effortful. If an answer comes too easily, the review interval should be extended. If you are breezing through every card without real effort, you are not doing spaced repetition — you are doing comfortable self-congratulation.
Learn new words in context, review them in isolation. Encountering a word in a sentence or conversation creates the initial anchor. Spaced repetition then keeps it alive. The two work together, not in competition.
Accept that some forgetting is part of the process. If you are never getting any cards wrong, your intervals are probably too short. Forgetting a card and then correctly recalling it on the next review is one of the most effective moments in the whole system. It hurts a little — which is exactly the point.
The bottom line
Spaced repetition works because memory is not a filing cabinet — it is a muscle that gets stronger through use at the right moments. The science has been replicated dozens of times across multiple decades and multiple languages.
But the technique only works if it is implemented faithfully. Streaks, daily reminders, and gamification are engagement mechanics. They are not memory mechanics. The best thing you can do for your language learning is to find a system that lets you review what you actually need to review, when you actually need to review it — and then trust the process.