Explain it to someone who isn't there

You have probably had the small, humbling experience of thinking you knew something until someone asked you to explain it. The word was right there — you'd read it a dozen times, you could pick it out of a line-up — and then a friend said "wait, what does that actually mean?" and you found yourself reaching for a definition that wouldn't come. Recognising a word and being able to teach it are two very different depths of knowing, and most studying quietly leaves you in the shallower one.

Which is a shame, because a long line of research points at a fix that costs nothing and feels almost too simple to bother with: don't study a word to know it. Study it to teach it. The odd part — the part that took psychologists a while to believe — is that you don't actually have to teach anyone. You just have to be getting ready to.

A finding hiding in plain sight

Teachers have always said they only really learned their subject once they had to teach it, and for a long time that was filed under folk wisdom. The first careful test came in a 1980 paper in the Journal of Educational Psychology, where John Bargh and Yaacov Schul had people study material under one of two mindsets: some prepared for a test, others prepared to teach what they'd read to another student. The teachers-to-be came away understanding the material better. Preparing to pass something on, it seemed, was a better way to take it in than preparing to be examined on it.

That could have been the whole story — teaching helps because teaching is hard work, and hard work helps. But in 2014, Julie Nestojko, Dung Bui, Nate Kornell and Elizabeth Bjork ran a version that pulled the surprise out into the open. In two experiments published in Memory & Cognition, students read a passage; half were told they'd be tested on it, half were told they'd have to teach it to another student. Then everyone was simply tested, and nobody taught anyone anything. The teaching was a rumour, a frame of mind, and nothing more.

The expect-to-teach group still won. They produced more complete and better-organised recall of the passage, and answered more questions correctly — especially the questions about the main points, the load-bearing ideas rather than the trivia. The teaching never happened; the expectation of it was enough to change how they read. Somewhere between "I will be tested" and "I will have to explain this," the mind shifts into a different gear.

The students who learned the most were preparing for a lesson that never took place. The expectation did the work; the audience was optional.

What a listener makes you do

Why should an imaginary student change anything? Because the moment you plan to explain something, you start reading it differently — and every one of those differences is a known friend of memory.

You select: teaching forces you to decide what actually matters, because you can't pass on everything, so you hunt for the spine of the thing. You organise: a good explanation has an order, so you start arranging ideas into a shape someone could follow — which is exactly the better-organised recall Nestojko's group measured. You elaborate: to explain a word you reach for an example, a contrast, a plainer synonym, a little story, and each of those is a new thread tying the word to something you already know. And you rehearse the moment of retrieval — the anxious, useful rehearsal of "could I actually produce this if I had to?"

Studying to be tested, oddly, invites none of this as strongly. A test, in the mind's shorthand, is something you survive; a lesson is something you have to make. One asks you to hold the material still long enough to be checked. The other asks you to take it apart and rebuild it in your own words — and rebuilding is how things get learned. This is the same reason saying a word aloud and tying it to your own life both help: they all drag a word out of passive recognition and make you do something with it.

The protégé effect

There's a second force at work, and it has a lovely name. In 2009, Catherine Chase, Doris Chin, Marily Oppezzo and Daniel Schwartz built a piece of software called Betty's Brain, in which children learn a science topic by teaching it to a cartoon pupil — the agent reasons out loud based on what you taught it, and gets things gloriously wrong when you've taught it badly. Across two studies in the Journal of Science Education and Technology, the children worked harder and learned more for their pupil than they did for themselves: they spent longer reading, took more care, and came away knowing more. The researchers called it the protégé effect, and found it was strongest in the lower-achieving students — the ones who most needed a reason to keep going.

The mechanism is almost embarrassingly human. We shrug off our own failures more easily than we'd let down someone in our care. A word you're learning for yourself can slide; a word you're responsible for explaining to someone — even a cartoon, even an imaginary someone — carries a small, motivating weight of obligation. You look at it more closely because you're about to be on the hook for it.

The honest catch

Now the part that keeps this from being a magic trick. If teaching works, why does it work — is there something special about the act of teaching, or is it doing something more ordinary in disguise? In 2018, Aloysius Wei Lun Koh, Sean Hang Kai Lee and Stephen Wee Hun Lim went looking, in a study in Applied Cognitive Psychology, and their answer is worth taking to heart.

They split learners into four groups. One taught the material from memory. One taught it while reading from notes — teaching, but with the answers in front of them. One skipped the teaching and simply tried to recall the material. And one did unrelated arithmetic as a control. A week later, everyone sat a comprehension test. The two groups that came out on top were the one that taught from memory and the one that just retrieved — and they were roughly level with each other. The group that taught with their notes, never having to pull the material back out of their own heads, did no better than the people doing arithmetic.

The lesson is bracing: much of the fabled power of teaching is retrieval practice wearing a friendlier costume. Teaching helps because a real explanation makes you fetch the material from memory rather than read it off a page. Take the fetching away — let someone teach with the notes open — and the benefit largely evaporates. So "prepare to teach" is not a spell; it's a reliable way of tricking yourself into doing the retrieval that actually teaches.

Which points at one more refinement. Logan Fiorella and Richard Mayer, in Contemporary Educational Psychology in 2013, had students learn a lesson on the Doppler effect either expecting to teach it or actually delivering the explanation to a camera. Merely expecting to teach lifted their scores on the test taken right away. But the students who actually gave the explanation held onto the material longest, doing best on a delayed test days later. Intending to teach primes you; going through with the explanation is what makes it durable. So don't stop at the intention. Say the thing.

Worth remembering: preparing to teach a word makes you learn it better than preparing to be tested — the effect shows up even when the teaching never happens, because the expectation alone changes how you read. But the engine underneath is mostly retrieval: teaching helps when it forces you to pull the word back out of your own memory, and much less when you can lean on notes. So explain each word in your own words, from memory, to a listener real or imagined — and actually say it, don't just plan to.

What to actually do

For foreign vocabulary — where a word is a fragile new sound as well as a meaning — this is one of the cheapest upgrades available. A few ways to put it to work:

Study each new word as if you'll teach it tonight. Not "will I recognise this?" but "could I explain it to someone who's never met it?" That single shift in framing is the whole Nestojko finding, and it costs nothing.
Explain the word out loud, in your own words, without looking. Give its meaning, an example sentence, maybe a near-synonym and how it differs. Doing it from memory is the part that counts — that's the retrieval Koh's study found doing the real work.
Teach a listener who isn't there. A rubber duck, a pet, an empty chair, the voice memo app, a patient friend you're texting. The audience can be imaginary; the explanation has to be real.
Recruit a real protégé when you can. A study partner one step behind you, a curious family member — you'll try harder for them than for yourself, and the questions they ask will find the holes in what you thought you knew.
Finish the sentence, don't just intend to. Expecting to teach warms you up; actually delivering the explanation is what makes the word last. When a word comes up, take the extra ten seconds to say the little lesson aloud.

Where vocabla fits

You can't teach a word you can't get back out of your memory, and you can't get a word back if you've never practised reaching for it. That's the quiet dependency underneath all of this: teaching is powerful because it's built on retrieval, and retrieval is a muscle that only grows when a word is handed back to you at the right moment — just as it's about to slip — and you have to produce it yourself.

That's the part vocabla is built to handle. It brings each word back on the schedule where reaching for it does the most good, inside a real sentence rather than stranded on a card, so the act of recall is already happening every time you practise. What you can add, for free, is the last move the research keeps pointing to: when a word surfaces, don't just tap the answer and move on. Explain it to the room. Teach it to nobody in particular. The word you can teach is the word you've actually got.