間隔反復で記憶をハックする方法

Ben: Welcome back to Vocafy Unpacked. I'm your host, Ben, here again with Vocafy's language-learning expert, Clara.

Clara: Hi Ben. Great to be back.

Ben: Clara, today I want to talk about something every language learner knows.

You learn a bunch of new words. They feel familiar. Maybe you even test yourself and get them right.

Then a week later, somebody uses one of those words and you think, “I know this... I definitely learned this... what does it mean?”

What happened?

Clara: Something completely normal: you forgot.

Ben: That's reassuringly direct.

Clara: But it's important, because learners often treat forgetting as evidence that they're bad at languages.

Usually it isn't. Memory changes over time. If we encounter something once and then never use or retrieve it again, our ability to remember it often becomes weaker.

Ben: Is that the famous “forgetting curve”?

Clara: That's where the idea comes from.

In the nineteenth century, the psychologist Hermann Ebbinghaus carried out some of the earliest systematic experiments on memory. He learned lists of meaningless syllables himself and measured how difficult they were to relearn after different amounts of time.

His results showed a pattern that has become famous: forgetting was relatively rapid after learning, then slowed over time.

Ben: So that curve we see everywhere — the dramatic drop followed by a flatter line.

Clara: Exactly. But there's an important warning.

There isn't one universal forgetting curve that tells us exactly how much every person will forget after one hour, one day or one week.

Ebbinghaus studied one person — himself — using very artificial material. Real vocabulary learning depends on many things: how well you learned the word in the first place, how meaningful it is to you, how often you encounter it, what you already know, and what you're trying to remember.

Ben: So the useful lesson isn't “after exactly twenty-four hours, your memory will be at thirty percent.”

Clara: Definitely not.

The useful lesson is much simpler: one encounter is usually not enough, and the timing of later encounters matters.

Ben: Which brings us to spaced repetition.

Clara: Right.

Imagine you want to learn twenty new words. You could repeat each one ten times in a single evening.

That might make them feel very familiar.

Ben: Which is exactly when I confidently tell myself I've learned them.

Clara: And that's the trap. Familiarity right now isn't necessarily the same as being able to retrieve something days or weeks later.

Instead of doing all those repetitions at once, you can spread them across time.

You meet the word today. Return to it later. Then again after a longer interval. If it continues to be easy to remember, the gaps between reviews can become longer.

That's the basic idea behind spaced repetition.

Ben: And this isn't just an app trick?

Clara: No. The spacing effect is one of the most established findings in memory research.

Across many kinds of learning, information studied across separate sessions tends to be retained better over the long term than the same material studied repeatedly in one concentrated session.

Research specifically on second-language learning has found the same general pattern.

Ben: But why do we need an algorithm for that? Couldn't I just review everything every few days?

Clara: You could. And with ten words, that might work perfectly well.

The problem appears when you have fifty words, then five hundred, then several thousand.

Some are difficult. Some are easy. Some you've just learned. Others haven't troubled you for months.

Reviewing all of them equally would waste a lot of time.

Ben: So that's where an SRS — a Spaced Repetition System — comes in.

Clara: Exactly.

An SRS helps decide what should come back into practice and roughly when.

Items you're struggling with can return sooner. Items you keep remembering successfully can usually wait longer.

The point isn't to predict the exact second you're going to forget something. Human memory isn't that predictable.

The goal is much more practical: use your limited practice time where another retrieval is likely to be useful.

Ben: You just used the word “retrieval.” Is that different from repetition?

Clara: Very different — and this is one of the most important parts.

Suppose I show you a card that says:

“ephemeral — lasting for a very short time.”

You read it five times.

That's repetition.

But if I show you “ephemeral” and ask you to remember the meaning before revealing the answer, you're doing retrieval practice.

You're trying to bring the information back from memory.

Ben: So struggling for a few seconds isn't necessarily a bad thing.

Clara: Exactly.

Successful retrieval is itself a learning event. A large body of research shows that practising retrieval can improve later memory more than simply seeing the information again.

So a useful review system isn't just asking, “Have you looked at this word recently?”

It's giving you opportunities to actually retrieve what you know.

Ben: And presumably one kind of retrieval doesn't prove that I completely know a word.

Clara: That's especially important in language learning.

Think about what it means to “know” a word.

You might recognise it while reading but be unable to produce it yourself.

You might remember its meaning when you see the written form but fail to recognise it in speech.

Or you might understand it perfectly and still not know how to use it naturally in a sentence.

Ben: So vocabulary knowledge isn't really a simple known-or-not-known switch.

Clara: Exactly.

That's why varied practice can be useful.

You can recall a meaning, choose between alternatives, complete a sentence, write the word yourself, listen for it, or meet it again inside meaningful content.

These activities ask slightly different things of your memory.

Ben: Which sounds quite different from the classic flashcard experience of seeing the same isolated pair again and again.

Clara: It can be.

Flashcards themselves aren't the problem. They're a useful tool.

But a word becomes much richer when you also know how it sounds, where you've seen it, what kind of sentence it appears in, and how it connects to something meaningful.

Imagine you saved the word “ephemeral” from an article about a spectacular meteor shower.

Later, instead of seeing only:

“ephemeral = short-lived”

you can also reconnect it with the sentence and idea where you first encountered it.

Ben: So context gives the word somewhere to live.

Clara: That's a nice way to put it.

It doesn't mean that adding more information automatically makes every memory stronger. But learning vocabulary involves more than storing a translation. You're gradually building knowledge about the word's form, sound, meaning and use.

Meeting and retrieving it in different useful ways can strengthen those connections.

Ben: How does that fit into Vocafy?

Clara: Vocafy keeps the vocabulary you save connected with your learning context and history.

So a word or phrase you discover while reading, watching something or working with other content doesn't have to disappear when that activity ends.

You can return to saved vocabulary through review and practise it in different ways.

Ben: So spaced review becomes part of a bigger system rather than a separate deck of cards.

Clara: Exactly.

The system can help bring vocabulary back over time, while different activities let you work with that vocabulary through recall, listening, writing, sentence completion and other forms of practice.

And because saved words can keep their context, review doesn't have to mean repeatedly memorising disconnected word pairs.

Ben: What happens when I remember something easily?

Clara: In a spaced system, it generally makes sense for that item to appear less frequently.

Why spend tomorrow reviewing something that is already easy when another word keeps disappearing from memory?

As your memory becomes more reliable, the interval can grow.

Ben: And if I forget?

Clara: Then you've learned something useful about your learning.

The word probably needs more attention, so it can return sooner.

Forgetting during practice isn't a disaster. In fact, one reason we practise is to discover weaknesses while the stakes are low.

Ben: That's a much nicer way of looking at the red “wrong” answer.

Clara: It is.

A review isn't an exam. It's part of the learning process.

Ben: There's still one problem, though.

I've used spaced-repetition apps before. They're wonderful for a few days, and then suddenly I have a mountain of reviews waiting for me.

At that point, the scientifically optimal thing seems to be closing the app and pretending I never installed it.

Clara: And that's where learning science meets real life.

A theoretically efficient schedule isn't very useful if you stop using it.

The number of new items you add matters. The amount of review you can realistically sustain matters. And missing a day doesn't mean you've ruined the system.

Ben: So we don't need the famous “exactly ten minutes every morning forever” rule?

Clara: No magic number.

The broader principle is that distributed practice is usually better for long-term retention than cramming everything into one session.

A short session that you can realistically return to is often more useful than an enormous study session followed by a week of nothing.

But the best routine is one you can actually maintain.

Ben: And presumably we shouldn't spend our entire language-learning life doing reviews either.

Clara: Definitely not.

Remembering vocabulary is important, but language learning is bigger than memorising vocabulary.

Read. Listen. Have conversations. Write things. Encounter new language.

Spaced repetition is useful because it helps important words stay available while you do those other things.

It should support language use, not replace it.

Ben: So if we strip away all the hype, what's the real idea?

Clara: Don't expect one exposure to create a permanent memory.

Return to important things over time. Try to retrieve them instead of simply rereading them. Give difficult items more attention and easy ones increasingly more space.

And whenever possible, keep vocabulary connected to the language where it actually means something.

Ben: So we're not really defeating forgetting.

Clara: No. Forgetting is part of memory.

We're designing practice around it.

Ben: I think that's actually more interesting than “hack your brain and never forget anything.”

Clara: Much more realistic too.

Ben: Clara, thanks for helping us remember why we forget.

Clara: My pleasure.

Ben: And to everyone listening: if there's a word you learned recently and can still remember right now, maybe today is a good time to try recalling it again — before looking at the answer.