You don't learn a word the first time you meet it

Here is a moment every learner knows. You look up a word — let's say the Italian raggiungere, to reach or to arrive at — and the meaning clicks. It makes sense. You even use it correctly in the sentence in front of you. You feel, briefly, that you have learned it. Then the word turns up again four days later, and there is nothing there. Not a wrong guess. A blank.

The instinct in that moment is to blame yourself: you weren't paying attention, your memory is poor, you should have drilled it harder. But the more honest reading is that nothing went wrong at all. A single meeting was never going to be enough, for anyone. The interesting question — the one learners rarely ask out loud — is how many meetings it actually takes.

Trying to count the uncountable

It sounds like the kind of thing that should have a clean answer, and for a long time the field offered rough rules of thumb. Paul Nation, whose work has shaped much of how vocabulary is taught, has long argued that a word needs to be met many times before it is securely known — and that the figure is larger than most learners assume. But the estimates that followed scattered widely. Depending on the study and on what counted as "knowing," researchers have put the number anywhere from around six encounters to more than twenty.

That spread is not sloppiness. It reflects a real difficulty: knowing a word is not a switch that flips. You can recognise its spelling before you can recall its meaning, recognise its meaning before you can produce it, and use it in one phrasing long before you know which other words it likes to keep company with. So "how many times does it take" depends entirely on which version of knowing you mean.

The cleanest experiment

The study that pinned this down most carefully was run by Stuart Webb and published in Applied Linguistics in 2007. Webb gave 121 Japanese learners of English passages to read in which target items appeared a controlled number of times — one, three, seven, or ten encounters. To keep prior knowledge from contaminating the results, the targets were invented words slotted into otherwise ordinary text, so every learner met them genuinely fresh.

Then he measured not one kind of knowledge but ten — spelling, grammatical behaviour, the associations a word carries, recognition of meaning, the ability to produce it, and more. The pattern was consistent and quietly demanding: with each step up in repetition, gains appeared on at least one aspect of word knowledge. Ten encounters produced sizeable gains across the board. There was no point in the data where the curve flattened and said enough. More meetings simply meant more of the word known.

What that study makes vivid is that the encounters are not redundant. The third meeting is not a wasted copy of the first. Each one is doing different work — one shores up the spelling, another fixes the meaning, another teaches the grammar — which is exactly why a word can feel "known" after two or three exposures and still collapse the moment you need to produce it cold.

The same shape, outside the lab

You might expect the controlled-passage result to dissolve in the mess of real reading. It doesn't. Ana Pellicer-Sánchez and Norbert Schmitt, writing in Reading in a Foreign Language in 2010, had advanced learners read an unmodified novel — Chinua Achebe's Things Fall Apart — and tracked how well they picked up unfamiliar words according to how often each one happened to appear.

The curve had the same upward tilt. A noticeable amount of learning showed up once a word had been met five to eight times; a substantial jump came between roughly ten and seventeen encounters. And the study was candid about the ceiling of pure reading: learners recognised the meaning of a fair share of the target words, but reaching the point of being able to recall a word, unprompted, was much rarer from reading alone. Recognition is cheaper than recall, and the gap between them is precisely the gap we have written about before — the distance between a word you know when you see it and a word you can summon when you need it.

A word is not learned on a single meeting and forgotten on a single lapse. It is built, encounter by encounter, and each meeting adds a different brick.

The number is not the real lesson

It would be easy to walk away with a target — "meet each word a dozen times" — and treat the rest as bookkeeping. But the more important finding sits one layer beneath the count. How the encounters are arranged turns out to matter at least as much as how many there are.

Two arrangements do most of the work. The first is spacing. A long line of research, stretching back to the nineteenth century and still being confirmed today, shows that encounters spread out over time produce far more durable memory than the same number crammed together — the spacing effect we traced in why words slip away. Ten meetings packed into one sitting are worth less than ten meetings scattered across two weeks, even though the total is identical.

The second is variety of context. In a 2019 study in Cognitive Science, Ashley Pagán and colleagues had adults learn words through reading and found that meeting a word in a range of different sentences, rather than the same one repeated, improved both recognition and the ability to match the word to its meaning. The reason is intuitive once you see it: when a word shows up across several different contexts, the things those contexts share — the word's stable, core meaning — get reinforced, while the incidental details of any one sentence fall away. The mind abstracts the meaning out of the variety. A single sentence, however many times you reread it, cannot do that work.

What this means for how you practise

Put the findings together and a practical shape emerges, one that runs against the way most of us instinctively study.

• Expect to forget the first few times. A word vanishing after one or two meetings is not failure; it is the normal early stage of a process that takes several more encounters to finish.
• Count on returns, not marathons. Coming back to a word on six different days will teach it better than meeting it six times in one evening. Distribution beats intensity.
• Seek the same word in different clothes. A word met in five different sentences is being learned more thoroughly than the same word reread five times, because variety is what lets the core meaning separate from the noise.

Designing for the curve

This is, in the end, the quiet logic behind the way vocabla is built. A word does not appear once and disappear. It comes back — on a later day, not the same minute — and each time it returns it arrives in a different exercise and a different sentence: translate it here, reach for it from its opposite there, recover it from context somewhere else. That is not variety for its own sake. It is the spacing effect and contextual diversity, the two things the research keeps pointing at, turned into the rhythm of practice itself.

So the next time a word slips away a few days after you were sure you had it, you can let go of the small guilt that comes with it. You did not fail to learn the word. You met it once. The learning was always going to take the rest of the meetings — and the most useful thing you can do is make sure they come, spaced out and in good company, until the word is finally, durably, yours.