In praise of the wrong answer

Here is a word from Basque: sagarra. Before you read another line, take a real guess at what it means. Not a shrug — a genuine attempt, out loud if you can. Fruit? A place? Something you'd say to a friend?

You almost certainly got it wrong. (It means apple.) And the instinct, when a word is that unknowable, is to skip the theatre of guessing and go straight to being told — why waste a breath on an answer you can't possibly have? It feels not just useless but faintly counterproductive, like practising a mistake. The surprising thing, and the subject of this piece, is that the wasted breath wasn't wasted. That doomed little guess just made the correct answer easier to keep.

A phenomenon that shouldn't work

Psychologists call it the pretesting effect, and its close cousin errorful generation. The claim is deliberately provocative: being tested on material before you have studied it — and failing that test — leaves you learning the material better than if you had spent the same time simply studying it.

The cleanest early demonstration came from Nate Kornell, Matthew Hays and Robert Bjork in 2009, in the Journal of Experimental Psychology: Learning, Memory, and Cognition. In several experiments they showed people a cue word and asked them to guess a weakly related partner word — a target they had no way of knowing. The guesses were, as designed, almost always wrong. A comparison group skipped the guessing and simply studied the cue–target pairs for longer. On the final test, the guessers won. In all of their experiments, the group that had failed first remembered more than the group that had only studied. The paper's title says it without flinching: "Unsuccessful retrieval attempts enhance subsequent learning."

That the effect survives even when the guess is unknowable is the part that unsettles people. This isn't partial knowledge quietly paying off. It is the act of reaching for an answer — and coming back empty — that does the work.

The version that matters to language learners

Word-association pairs are a laboratory convenience. What you actually want to know is whether this helps with real vocabulary — the unfamiliar word and its meaning, the foreign term and its translation. It does, and there is a study aimed squarely at exactly that.

In 2014, Rosalind Potts and David Shanks ran four experiments, published in the Journal of Experimental Psychology: General, on precisely this material: definitions for obscure English words, and translations for foreign vocabulary. People learned each item in one of three ways. Some read the word and its meaning together, the ordinary way. Some chose from a set of options and were then corrected. And some were made to generate a guess from scratch before being shown the right answer.

The guess-first group came out ahead — remembering the correct definitions and translations better than the people who had simply read them, and better than the people who had picked from a menu. Generating an answer and getting it wrong, then seeing the correction, beat being handed the answer cleanly. For anyone who has ever felt that guessing at a word they couldn't know was a slightly embarrassing waste of time, this is a quiet vindication.

The wrong answer isn't the opposite of learning the right one. Reaching for it is part of how the right one arrives.

Why a doomed guess helps

The tidy explanation would be effort — guessing is more work than reading, and effort is good for memory. But that isn't quite it, and researchers have spent a decade arguing about what is. Potts and Shanks weighed two families of explanation. One puts the benefit at retrieval: a word you arrived at through an act of generation ends up wrapped in more potential cues and a more distinctive context, so it's easier to find again later. The other puts it at encoding: the moment you strain to produce an answer, you rouse a whole neighbourhood of related knowledge, and when the correct word finally lands, it lands in soil already turned over and waiting for it.

Either way, the shape is the same one that keeps recurring in this journal. A word you struggle toward is a word with more handles on it — the same currency of distinctiveness that makes mnemonics work, and the same reason that reaching for a word from memory teaches more than re-reading it. Pretesting simply moves that reaching to the very beginning, before you technically know anything at all.

The condition that makes or breaks it

None of this licenses guessing into the void. The wrong answer is only useful if the right one arrives right behind it. A guess with no feedback isn't a pretest — it's just a mistake you've now rehearsed, and rehearsed errors are sticky in exactly the way you don't want. Feedback is the hinge the whole effect turns on.

Janet Metcalfe made this the centre of her 2017 review "Learning from Errors," in the Annual Review of Psychology. Surveying decades of work, she argues that a culture of error-avoidance — never let the learner be wrong — is quietly counterproductive, provided corrective feedback follows the error. She also notes a lovely wrinkle called the hypercorrection effect: the errors people commit with the most confidence tend to be the ones they go on to correct most reliably. Being sure and being wrong, it turns out, is a surprisingly good state to learn from — the jolt of the correction does the teaching.

It isn't only word lists

You might suspect this is a trick that works on isolated pairs and nowhere else. It reaches further than that. In a 2009 paper in the Journal of Experimental Psychology: Applied, Lindsey Richland, Nate Kornell and Liche Kao had people read a passage about human vision. One group was asked questions about ideas in the passage before reading it — questions they were in no position to answer — while another group simply got extra time to study. The pretested group remembered the passage's contents better afterwards. Failing a test on something you haven't read yet primes you to absorb it when you do. The doomed guess sharpens your attention on precisely the thing you're about to be told.

Where the effect thins out

Two honest cautions, because the effect is real but not a magic switch. First, it leans on the guess being a meaningful attempt with something to grab onto. When a translation is wholly arbitrary — a sound with no path from cue to answer — the advantage of guessing shrinks, and plain study can hold its own. This is quietly good news for a learner: it means guessing pays off most when you can actually reason toward the word, from a cognate hunch, from the shape of a word you already half-know, or from the sentence around it. The context that makes a guess possible is the same context that makes the guess worth making.

Second, pretesting is an encoding advantage — it makes a first encounter count for more. It does not repeal forgetting. A word you guessed at brilliantly on Monday and never met again will still drift away like any other. Guessing makes the meeting richer; it doesn't make the meeting your last. As ever, the real power comes from stacking a good first encounter on top of spaced later ones.

What to do with this

The instruction is almost rude in its simplicity: stop peeking. When a new word appears, resist the pull to read the answer first.

• Guess before you reveal. Whatever you're studying — a card, a blank in a sentence, a word in the wild — commit to an answer before you check it. The commitment is the active ingredient; a shrug doesn't count.
• Then check, every time. The correction is not optional. Guess, look, and let the gap between the two register. That gap is where the word settles in.
• Reason toward the guess. Lean on cognates, on word roots, on the sentence around the blank. A guess with a thread of logic behind it teaches more than a coin-flip.
• Let yourself be confidently wrong. The bolder the guess, the sharper the correction lands. Being wrong out loud is not a failure of study — it's a form of it.

Where vocabla fits

A fill-in-the-blank is, quietly, a pretest. When vocabla drops a word out of a real sentence and asks you to supply it, you're committing an answer before you're shown the right one — the exact move the research keeps rewarding. Sometimes you'll know it; often, early on, you'll reach and miss. Either way the correct word arrives a half-second later, landing in ground your guess has already prepared. And because reviews are spaced to bring each word back just as it starts to fade, the rich first encounter isn't left to stand on its own.

So the next time a word you couldn't possibly know swims up on the screen, don't wait to be told. Take the guess. Get it wrong with conviction. Then look — and notice how much more firmly the right answer holds, precisely because you dared to be wrong about it first.