Your hands remember what your eyes forget

Think about the way most vocabulary gets studied: seated, still, eyes moving across a list, the rest of the body switched off. It seems like the sensible arrangement — the mind is doing the work, so why involve anything else? And yet there is a small, slightly undignified thing you can add to that scene that makes the words stick harder. You can move your hands. When you meet a word for to throw, throw an imaginary something. For heavy, sag under an invisible weight. It feels theatrical, and it works, and the reason it works reaches all the way down into how memory is built.

Researchers call it enactment, and it has been quietly accumulating evidence for over forty years. The short version: a word you perform is remembered better than a word you only read or hear. The longer version is more interesting, because it turns out the body is not a bystander to memory at all.

A memory that lives in the muscles

The groundwork was laid in the early 1980s, well before anyone was thinking about language apps. Working independently, Johannes Engelkamp in Germany and Ronald Cohen in Canada handed people lists of simple action phrases — brush the teeth, shuffle the cards, lift the cup. Some participants just heard or read the phrases. Others were asked to actually perform each little action as it came up. Then everyone's memory was tested.

The people who had done the actions remembered reliably more of them, in both free recall and recognition. The finding was robust enough to earn its own name — the enactment effect, sometimes called the self-performed task effect — and it launched a whole subfield devoted to the memory of actions. Doing a thing, it turned out, laid down a sturdier trace than merely being told about it.

You can feel why this might be. When you enact a phrase, you don't just store its sound and its meaning; you store a motor record of your own body carrying it out. That is an extra, unusual handle on the memory — and, as with so much of how remembering works, the thing with more handles is the thing you can find again. This is the same currency of distinctiveness that makes the keyword method work and that rewards saying a word out loud rather than reading it in silence.

From remembered chores to foreign words

Brushing imaginary teeth is one thing. The real question, for anyone learning a language, is whether the same trick works on the genuinely hard part — attaching a brand-new, meaningless string of sounds to a meaning. Manuela Macedonia, working at the Max Planck Institute for Human Cognitive and Brain Sciences in Leipzig, set out to test exactly that.

Her clever move was to invent a language. Rather than use a real one, where learners might already know a few words, she and her colleagues built an artificial vocabulary called Vimmi — tri-syllabic words that obey the sound-rules of Italian but mean nothing to anyone. In one study, twenty volunteers spent six weeks learning Vimmi. For half the words, they read, heard and spoke them in the ordinary way. For the other half, they did all that and performed a matching gesture — a symbolic movement of the whole body that fitted the meaning. Same words, same time, same class; the only difference was the arms and the trunk.

The gesture-trained words won, and not by a whisker. Across her programme of experiments, words encoded with a self-performed gesture were remembered in larger numbers and held on longer than the words learned by eyes and ears alone. The body, it seemed, was not a distraction from learning the word. It was part of the word.

But the gesture has to fit

Before you start flailing at every flashcard, a crucial detail: not any movement will do. This was pinned down neatly by Spencer Kelly and colleagues in a 2009 study that taught English speakers a set of Japanese verbs. Learners saw the words paired with hand gestures — but for some words the gesture matched the meaning (a drinking motion for to drink), while for others the gesture was deliberately mismatched.

Only the congruent gestures helped. Words taught with a gesture that fit their meaning were learned best of all; words taught with a gesture that contradicted the meaning were learned no better — and sometimes worse — than words taught with speech alone. The lesson is that the movement isn't a magic charm you wave to summon memory. It works because it carries a second, redundant copy of the meaning, delivered through a different channel. Give the body the wrong message and you are simply adding noise.

A gesture doesn't help because you moved. It helps because your body said the same thing your voice did — meaning, told twice, through two different doors.

And it lasts

Encoding tricks are easy to demonstrate over an afternoon and much harder to demonstrate over a term. The most persuasive thing about the gesture research is that Macedonia and Wolfgang Klimesch took it into an ordinary classroom and left it there. In a study published in 2014, learners were taught foreign words over an extended period, and their memory was probed again and again across fourteen months. The words that had been learned with gestures stayed ahead of the audio-visual words the whole way down the curve — the advantage didn't just appear in a lab and vanish by the following week. It survived more than a year of ordinary forgetting.

Why the body helps the mind

Here is where the story stops being a collection of behavioural results and becomes something stranger. When researchers put gesture-learners into a brain scanner, the words carried a physical signature. In fMRI work by Macedonia and colleagues, hearing a word that had been learned with a gesture lit up the premotor cortex — a motor-planning region — even though the person was lying perfectly still, doing nothing but listening. The memory of the word had a movement folded inside it. And the more strongly that motor region activated at test, the better the word was remembered.

The tidy objection is that this is just correlation: perhaps the motor flicker is a harmless side-effect, riding along beside the real memory rather than doing any work. But a 2021 study by Brian Mathias and colleagues went at that objection directly. Using transcranial magnetic stimulation to briefly disrupt the motor cortex, they found that interfering with that region impaired people's memory for words they had learned with gestures. Knock on the motor system, and the gesture-learned words wobble. That is about as close as this kind of research gets to saying the body isn't decorating the memory — it is holding part of it up.

This fits a broader idea that has been gaining ground for decades: that meaning is not stored in some abstract, disembodied dictionary in the head, but is grounded in the sensory and motor systems we used to experience the world in the first place. The word throw may partly be, in the brain, a faint echo of throwing.

Where it stops

It would be a poor summary that oversold this, so a few honest limits. First, as with saying words aloud, much of the classic evidence comes from mixed lists — some words gestured, some not — and part of the benefit is the gestured words standing out against their stiller neighbours. Enact absolutely everything, robotically, and you dilute the very distinctiveness you were buying.

Second, the effect is cleanest for words you can physically depict — actions, objects, sizes, directions. Abstract words are harder to throw a shape around, though newer work (including studies from Macedonia's own group) suggests a well-chosen gesture can help there too, just less dramatically. Don't expect to mime your way into the vocabulary of philosophy.

Third, and most important: gesture is an encoding move. It makes the moment you meet a word count for more; it does not repeal the way memories fade. A word acted out once and then never revisited is still exposed to the ordinary forgetting curve. The body gets the word in deeper. Coming back to it later is what keeps it there.

What to do with this

None of this asks for special equipment or extra study time. It asks you to stop treating your body as furniture while you learn.

• Act out the actable words. Verbs and concrete nouns are the low-hanging fruit. For to open, open something in the air; for mountain, draw its shape with your hands. Make the gesture yourself — watching someone else do it is far weaker than doing it.
• Make the movement match the meaning. A random wave does nothing. The gesture works only when it delivers the same idea your voice is delivering, through a second channel.
• Gesture as you recall, not only as you learn. Re-performing the movement while you reach for the word from memory folds enactment into retrieval practice — two of the most dependable findings in the field, working the same beat.
• Pair it with the sentence, not the bare word. Acting out a word inside a real sentence hands the memory its grammar and its neighbours as well as its movement — more handles still.
• Don't mime everything. Save the gestures for the words that are actually hard, and let the effort fall where the difficulty is. Enacting the whole list flat drains the distinctiveness that makes any single gesture worth it.

Where vocabla fits

You can't reach through a screen and shake the reader's hands, so vocabla doesn't pretend to teach gesture directly. What it does do is refuse to let you learn words the most passive way possible. You don't sit and re-read lists; you type the missing word into a real sentence — itself a small, deliberate motor act rather than a glance — and the dialogues you unlock give the words back to you as language in use rather than entries on a card. Words arrive attached to meaning, in context, with something for you to do.

The gesture is the part only you can add, and it costs nothing but a little self-consciousness. When a new word comes up and it's the kind you could act out, act it out — throw the imaginary ball, lift the invisible weight, trace the mountain in the air. You will feel slightly ridiculous for a second, and then, a year later, you will still know the word. Of all the trades in language learning, that is one of the better ones.