How Effective Is Spaced Repetition? The Research, and Where It Fails

You've probably heard that spaced repetition is the most effective method for building vocabulary. Maybe you tried it and found the progress slower than you expected. Or maybe you're just skeptical of anything that gets called "scientifically proven" on the internet.

Fair enough. Here's what the research on spaced repetition language learning effectiveness actually says - including where the method has genuine limits that most enthusiasts don't mention.

Where spaced repetition comes from

The underlying science isn't new. Hermann Ebbinghaus spent years in the 1880s memorizing lists of nonsense syllables, then testing himself at different intervals to map how memory decays over time. What he documented became the forgetting curve: we lose information at a predictable rate, and that rate slows down each time we successfully review something.

The practical implication is significant. If you review a word right before you'd forget it, you reinforce the memory far more efficiently than if you review it randomly or repeatedly in one sitting. And crucially, the optimal review interval is calculable - it varies by word and by person, but it follows patterns that can be modeled.

Modern spaced repetition apps like MindDory are essentially implementations of this insight, running the scheduling math automatically so you don't have to. For a practical look at how to actually use the system day-to-day, the spaced repetition best practices guide covers exactly that.

What the studies actually found

Research on spaced repetition language learning effectiveness is consistent across decades. A few findings worth knowing specifically:

Studies comparing spaced practice against massed practice - cramming, essentially - reliably find that spacing wins for long-term retention. Not by a little. A landmark 2008 paper by Cepeda and colleagues found that spaced practice improved retention by 64% over massed practice when tested weeks later. The cramming group often performed comparably right after studying. A month later, the gap was substantial.

Research focused specifically on second language vocabulary acquisition has found that SRS-based study helps learners build larger vocabularies faster than wordlist memorization, particularly when cards include context like example sentences rather than isolated word-translation pairs. The context isn't just a nice-to-have - it measurably improves retention.

There's also a body of research on the testing effect: the act of trying to retrieve a memory is itself a more powerful consolidator than passive review. Flashcards do this automatically. Every time you see a card and try to recall the answer before flipping it, you're doing active retrieval. SRS combines retrieval practice with optimal timing - that's why the results are as strong as they are.

Where it works best

Spaced repetition is best suited for large sets of discrete facts that need to stay in memory long-term. Vocabulary is almost a perfect fit. Either you know a word or you don't. It needs to be accessible for years, not just for tomorrow's test. And there's a finite number of high-value words to learn, so progress compounds - every word you add makes the next one easier because you're building a richer context.

The high-frequency vocabulary range is where spaced repetition pays off most. The top 1,000 most common words in a language cover around 85% of everyday speech. Getting those locked in memory has an outsized effect on listening comprehension and reading - you stop constantly hitting words you don't know and can actually follow what's happening. Studying vocabulary faster with AI flashcards can help you reach that threshold significantly quicker.

Other things it handles well: irregular verb forms, grammar rules you keep forgetting, pronunciation pairs that sound similar, kanji readings if you're studying Japanese. Anything with a large set of things to memorize where accuracy matters.

Where it has genuine limits

This is where a lot of SRS advocates go quiet, so it's worth being direct.

Spaced repetition builds vocabulary. It doesn't build fluency. Those are different things. Knowing 3,000 words doesn't mean you can speak, write, or even listen effectively in a language. Fluency requires processing language in real time - understanding fast native speech, constructing sentences under pressure, navigating conversation where you don't control the topic. Flashcards don't develop any of that directly.

It also doesn't work well for things that aren't discrete retrievable facts. Developing an intuition for when to use the subjunctive, getting a feel for natural word order, understanding cultural references - these require immersion and practice, not more flashcards.

And the method only produces results proportional to consistency. Two weeks of intensive SRS followed by a month off will leave you with much less than someone who does 15 minutes every day. The spacing only works if there's actually space - meaning time and regularity.

Why it sometimes feels like it isn't working

Most people who try spaced repetition and don't see results are making one of a few specific mistakes.

Adding too many new words at once. Every word you add today generates review debt for weeks and months ahead. Add 50 words on day one and by week three your daily review pile is unmanageable. The sustainable range for most learners is 10–20 new words per day, and the lower end of that is smarter when you're starting out. The spaced repetition best practices go into exactly how to set these limits right.

Grading too generously. If you mark a card as "Good" when you barely remembered it, the algorithm schedules the next review too far out. When it comes back, you've forgotten. Honest ratings - even harsh ones - are what keeps the system accurate.

Using it as their only study method. SRS is a foundation, not a complete approach. People who combine flashcard review with actual reading, listening, or speaking in the language see dramatically better results than people who only do flashcards. The two reinforce each other: you review a word on a card, then encounter it in a podcast, and the memory solidifies in a way that neither would achieve alone.

Realistic numbers for what to expect

With consistent daily study - around 15–20 minutes - and 10–15 new words per day, most learners can build a working vocabulary of 1,500–2,000 words in a new language within six months. At that level, you understand the rough shape of most everyday conversations and can follow a significant portion of written text.

Two thousand words is approximately the threshold for 90% comprehension in common speech. That sounds abstract until you experience it: it's the point where you can follow a conversation without stopping every few seconds, where reading starts to feel like reading rather than decoding.

At a year of consistent study, 3,000–4,000 words is achievable. At that vocabulary size you can follow most news, watch TV with occasional pauses, and hold real conversations on most topics. Not native-level. But genuinely functional.

These numbers assume reasonable card quality, honest self-grading, and actual daily consistency. They're not a guarantee - but they're what the method reliably produces for people who use it well.

What is the optimal ratio of SRS to immersion?

This is one of the most searched questions in language learning communities - and the honest answer is that there is no single optimal ratio that works for everyone. But research and practical experience point to a useful framework.

Spaced repetition and immersion do different jobs. SRS builds your vocabulary store - the words you can recognise and retrieve. Immersion (listening, reading, speaking in real context) trains your brain to use that vocabulary fluidly, process it at speed, and infer meaning from surrounding context. Neither works well without the other.

For most learners, a practical research-informed split is roughly 20–30% structured SRS review and 70–80% comprehensible input and immersion. In time terms: if you study 60 minutes a day, that's 12–18 minutes of SRS and the rest on real-language exposure. The SRS keeps your vocabulary sharp; the immersion turns that vocabulary into usable fluency.

At beginner level, the ratio often shifts toward more SRS - you need a working vocabulary before immersion becomes comprehensible. Once you pass roughly 1,000–1,500 words, immersion becomes significantly more valuable and the ratio should shift accordingly. Advanced learners often drop structured SRS to maintenance level (5–10 minutes of reviews) and spend the bulk of their study time on immersion and production.

The key insight from the research: SRS and immersion are not competing strategies. They are complementary, and the learners who get the best results treat them as a system rather than choosing between them.

The honest summary

Spaced repetition language learning effectiveness is real and well-documented. It's one of the most evidence-backed approaches to vocabulary acquisition we have. Used correctly - good cards, honest ratings, daily consistency, combined with real language exposure - it produces results that are hard to match with other study methods.

Used badly - too many words, inflated ratings, treated as a complete substitute for actual language use - it produces a growing backlog and the feeling that you're going in circles.

The research tells you the method works. The habits you build around it determine whether it actually does for you. MindDory is designed to make the good habits easier - built-in sentence examples, audio, sensible new-card limits - but the daily commitment is still yours to make.