The Art of Remembering: Why Spaced Repetition Works

There’s something strangely frustrating about learning a language.

You spend an evening learning twenty new words. The next morning, you feel like you know them. A week later, someone asks you a simple question and suddenly that word is nowhere to be found.

You have seen it. You have written it down. You have highlighted it. You may even have memorised it. And yet, you forgot it.

For a long time, I thought this was simply part of learning a language. Then I discovered spaced repetition — a deceptively simple idea that completely changed the way I think about learning.

The idea is this: you don’t need to repeat everything constantly. You need to repeat the right things at the right time. And there is a lot more science and history behind that idea than most people realise.

We are not designed to remember everything

Human memory is selective. If you think about everything that happened to you last Tuesday, you probably can’t remember most of it. But you might remember one conversation, a strange smell, a song you heard, or something that made you laugh. Our brains are constantly deciding what is worth keeping and what can disappear.

One of the earliest people to study this process scientifically was the German psychologist Hermann Ebbinghaus. In the late nineteenth century, he conducted experiments on his own memory, memorising meaningless combinations of syllables and measuring how much he could retain over time. His work became an important foundation for the scientific study of forgetting — and it is where the famous idea of the forgetting curve comes from.

The basic observation is quite intuitive: after we learn something, our memory of it tends to weaken over time if we don’t encounter or retrieve it again. But there is an important detail. Forgetting isn’t necessarily the enemy of learning. In fact, a small amount of forgetting can make the next attempt to remember something more useful.

That is where spaced repetition becomes interesting.

So, who actually invented spaced repetition?

This question has a slightly complicated answer. It is tempting to say that Ebbinghaus invented spaced repetition, but that isn’t quite correct. He studied memory and forgetting, and his research provided a foundation for later work; the idea of spacing learning over time had also been discussed by other researchers and thinkers.

A major practical step came in 1972, when German researcher Sebastian Leitner introduced what became known as the Leitner system. His idea was beautifully simple. Imagine you have a pile of flashcards. Every time you remember a card correctly, it moves into a different box and is reviewed less frequently. If you forget it, it moves back and gets more attention. You can do the entire system with nothing more than paper, cards and a few boxes.

Then, in the 1980s, another important development happened. Polish researcher Piotr Woźniak began experimenting with computer-based repetition. In 1985 he developed an approach based on increasingly long intervals between reviews, and in 1987 the first version of his SuperMemo program appeared. His work eventually led to algorithms that could adapt repetition schedules to individual pieces of information and to the learner’s performance.

So there wasn’t really one person who woke up one morning and “invented” spaced repetition. It was more like a long chain of discoveries:

research into memory → research into forgetting → practical flashcards → algorithms → modern learning systems.

But how does spaced repetition actually work?

The easiest way to understand it is to imagine learning a new word. Let’s say you are learning Russian and your new word is встреча — meeting.

If you immediately see it again ten times, it will probably feel very familiar. But familiarity is not necessarily the same thing as being able to remember something later. Instead, imagine this:

• Day 1: Learn встреча
• Day 2: Try to remember it
• Day 4: See it again
• Day 8: Retrieve it again
• Day 16: Review it
• Day 30: Review it again

The exact intervals don’t have to be these numbers — in modern systems they adjust depending on how difficult the item is and how well you remember it. The important part is the space between repetitions. And there is another important ingredient.

You have to try to remember

Spaced repetition works particularly well when the learner isn’t simply looking at the answer, but actively tries to retrieve it. If I show you:

встреча = meeting

you might think, oh yes, I know that. But if I ask, “How do you say ‘meeting’ in Russian?” and you have to produce встреча yourself, you are doing something different. You are retrieving information from memory. That little struggle is useful.

Why not just repeat everything every day?

Because that would be exhausting. Imagine learning 2,000 Russian words and reviewing every single one every day — you would spend your entire life reviewing vocabulary.

The beauty of spaced repetition is that not every piece of information needs the same amount of attention. A word you consistently remember can appear less often. A word you keep forgetting should come back sooner. In other words, the system gradually learns something about you.

This is why modern spaced-repetition systems can feel very different from traditional flashcards. Algorithms use your previous answers to estimate when an item is likely to become hard to recall again, and schedule another review around that point. The exact algorithms differ from system to system, but the underlying principle remains the same.

The broader research supports it. A major 2006 review by Cepeda and colleagues analysed hundreds of assessments across hundreds of experiments and found substantial evidence for the benefits of distributed practice — while also showing that the optimal spacing depends on how long you want to retain the information.

The strange thing about good learning

There is a paradox here: good learning doesn’t always feel easy.

When you recognise a word immediately, everything feels comfortable — I know this already. But when you have to pause for a moment and search your memory — what was that word again? — something more interesting is happening. You are exercising the connection between the concept and the information stored in your memory.

That is why spaced repetition isn’t simply about showing you the same flashcard again and again. It’s about creating the right amount of difficulty. Too easy, and you’re mostly recognising. Too difficult, and you have no chance of retrieving. The useful space is somewhere in between — and that is where learning happens.

What does this have to do with learning Russian?

Almost everything. Learning a language isn’t one big skill. It’s thousands of small pieces of information that have to become available quickly: words, phrases, verb forms, genders, endings, pronunciation, expressions, patterns.

You don’t want to spend ten seconds remembering how to say a basic word when you’re in a conversation. Eventually, you want it to appear naturally. And that requires something more than simply seeing Russian — you need to encounter it repeatedly, over time, and retrieve it when you need it.

Let’s say you learn the word забывать — “to forget”. You might understand it perfectly when your teacher explains it. But understanding a word when you see it is only one step. You also want to hear it, recognise it, remember its meaning, produce it yourself, and eventually use it naturally in a sentence. Repetition helps move information from “I think I’ve seen this before” toward “I actually have access to this.”

This is why we use it at Putnik

When I started Putnik, I didn’t want the platform to become another place where students simply collect information. There is already an enormous amount of it online: vocabulary lists, grammar explanations, videos, textbooks, apps, podcasts. The real problem isn’t always access to information. Sometimes the problem is remembering it.

That’s one of the reasons we built spaced repetition into the Putnik learning experience. Instead of asking you to memorise something once and hoping it stays there forever, we bring important material back to you over time. Something you struggle with receives more attention; something you’ve mastered doesn’t need to constantly interrupt you. The system becomes part of the learning process rather than another task you have to remember to do.

The system should remember what you are learning, so that you can focus on actually learning.

Learning is not about never forgetting

I think this is the part people often get wrong. The goal isn’t a perfect memory where you never forget anything — that’s not how human memory works. The goal is to make forgetting useful.

You forget something. You try to retrieve it. You remember it again. Then you leave it alone for a little longer. And when it starts becoming difficult again, you bring it back. Over time, the intervals grow longer and the information becomes more stable.

It is almost like building a path through a forest. The first time, there is barely a path. You walk it again. Then again. Then again. Eventually you don’t have to think about where the path is anymore — you just walk. That is what I like about spaced repetition: it doesn’t promise a shortcut around learning. It gives repetition a structure. And when you’re learning something as complex as a language, that structure makes an enormous difference.

One last thing

If there is one idea I’d like you to take away, it is not a particular algorithm or a particular number of days between reviews. It’s much simpler: don’t just learn something once. Meet it again later. And then meet it again. Give your brain a reason to remember.

Because learning a language isn’t about how much you can memorise today. It’s about what you can still remember six months from now. That is the real test. And that is why, at Putnik, we don’t just teach Russian — we build systems that help you remember it.