You're previewing the Bennett demo family. Numbers, schedule, and kids are sample data.Start your own
Loading page…
Hearing a statement a second time makes people rate it as more likely to be true. It has been reproduced for nearly fifty years, and the unsettling part is that it still works on people who can tell you the correct answer.
Pick something you are fairly sure is true but never studied. Some fact about sharks, or the Great Wall, or how much of your brain you use.
Now try to say where you got it.
Most people cannot. The fact is there, it feels solid, and the receipt is missing. That is not a failure of memory — memory is built that way. You keep the claim for years and throw the source away in about a week.
Which leaves a question with an uncomfortable answer: if you no longer know where a thing came from, what exactly is your sense that it's true made of?
In 1977 three researchers — Lynn Hasher, David Goldstein and Thomas Toppino — showed people sixty statements and asked them to rate how likely each was to be true.
Then they did it again two weeks later. And again two weeks after that.
The trick was in the list. Some statements appeared in all three sessions; others appeared only once. Nobody was told which was which, and none of the statements were things an ordinary person would know for certain either way.
The repeated statements climbed. Session after session, the same sentence was rated as more likely to be true — not because anything new had been learned about it, but because it had been seen before. The effect has been given a name since: the illusion of truth.
Repetition does not add evidence. It changes how true something feels, which is a completely different thing wearing the same clothes.
This is where nearly everybody goes, and it is the reason the effect is worth an episode rather than a footnote.
The obvious defence is knowledge. Repetition might shift you on some obscure claim about zebras, but surely not on something you actually know.
So in 2015 a team led by Lisa Fazio tested exactly that. They used statements whose correct answers the participants could be shown to hold — the sort of thing that, asked directly in a quiz, those same people got right.
Repetition raised the truth ratings anyway. People who could correctly name the largest ocean when asked outright still drifted towards a repeated statement naming a different one.
Knowing the answer did not switch the effect off. It sat on top of the knowledge rather than being filtered by it.
Someone can answer a question correctly when you ask them straight out, and yet rates a repeated statement contradicting that answer as more likely to be true. What does that tell you about how the effect works?
A sentence you have met before is easier to take in. It reads faster, the words come more smoothly, nothing snags. Researchers call that ease processing fluency, and your brain uses it as a rough signal.
That signal has to stand in for something, because most of what you are told cannot be checked. You have no way to personally verify how deep the ocean is or who invented the zip. So the mind reaches for a cheap substitute, and this feels easy and settled is the cheapest one available.
If that sounds familiar, it should. It is the same instrument being misread as in re-reading a chapter until it feels known — smooth going in, mistaken for solid. There the fluency counterfeits learning. Here it counterfeits truth.
Do not walk away thinking your head is broken. In ordinary life, things you hear repeatedly from different people usually are true — that the sea is salty, that the shop shuts at six, that ice is slippery. Repetition normally travels with truth, because true things get independently rediscovered by lots of people.
That is exactly what makes the failure mode so awkward. The rule is not stupid; it is a good rule being fed bad input. Repetition can be manufactured. One claim, repeated by one person a hundred times, or reposted by a hundred people who all read the same post, produces the identical feeling to a hundred people who each found it separately.
From the inside, those two situations are indistinguishable. That is the whole problem, and it is why the fix has to be a question rather than a feeling.
A made-up claim travels through a week. Press through and keep two counts in your head at once: how many times you have met it, and how many places it actually came from.
Three separate friends tell you the same surprising thing this week. What is the one question that decides whether that is worth anything?
Three things, in order of how much they are worth.
Ask the question at first contact. In one experiment, people who were asked to judge whether statements were true at the moment they first saw them did not show the later boost from repetition. Deciding what you think about a claim while it is still new appears to inoculate you against your own future familiarity with it. That is unusual — most of these effects cannot be undone from the inside — and it is worth spending ten seconds on.
Count sources, not mentions. The question is never "how many times have I heard this", which is a number about you. It is "how many places did it come from", which is a number about the world.
Learn the tell. A thing you are confident about but cannot trace — no book, no person, no lesson, no date — is not necessarily false. But confidence with no receipt attached is precisely the signature this effect leaves behind, and it is the one moment where you can catch it happening.
And one honest limit. This is not a defect you can fully switch off. You cannot un-see a sentence, and being told afterwards that something is untrue does not delete the familiarity it built — a corrected claim can go on feeling oddly plausible for years.
So the leverage is nearly all at the front end. Not in arguing with what already feels true, but in the ten seconds when a claim is still new to you and you can still ask, cheaply, whether you have any reason to believe it beyond having just read it.
Six things that could raise your confidence in a claim. Some are actual reasons to think it more likely true; the others only tell you that you have met it before. Sort each and read what it is really doing.
Tap an item, then tap where it belongs
Having read all this, should you start treating anything that feels familiar as probably untrue?
You’re previewing as a parent — nothing here is recorded.
Hasher, L., Goldstein, D., & Toppino, T. (1977). 'Frequency and the conference of referential validity.' Journal of Verbal Learning and Verbal Behavior, 16(1), 107-112. Participants rated the validity of 60 plausible statements in three sessions held two weeks apart; some statements repeated across sessions and some did not. Repeated statements received higher validity ratings over successive sessions. Fazio, L. K., Brashier, N. M., Payne, B. K., & Marsh, E. J. (2015). 'Knowledge does not protect against illusory truth.' Journal of Experimental Psychology: General, 144(5), 993-1002. Repetition increased perceived truth even for statements that contradicted knowledge the participants could demonstrate they possessed when tested directly. Dechêne, A., Stahl, C., Hansen, J., & Wänke, M. (2010). 'The truth about the truth: A meta-analytic review of the truth effect.' Personality and Social Psychology Review, 14(2), 238-257. The repetition-induced truth effect is robust across a large body of studies. Pennycook, G., Cannon, T. D., & Rand, D. G. (2018). 'Prior exposure increases perceived accuracy of fake news.' Journal of Experimental Psychology: General, 147(12), 1865-1880. A single prior exposure to a fabricated headline increased its later perceived accuracy. Brashier, N. M., Eliseev, E. D., & Marsh, E. J. (2020). 'An initial accuracy focus prevents illusory truth.' Cognition, 194, 104054. Participants prompted to evaluate accuracy at first exposure did not subsequently show the repetition-driven boost in perceived truth. McGlone, M. S., & Tofighbakhsh, J. (2000). 'Birds of a feather flock conjointly? Rhyme as reason in aphorisms.' Psychological Science, 11(5), 424-428. Rhyming aphorisms were rated as more accurate descriptions of human behaviour than non-rhyming versions expressing the same meaning — the processing-fluency finding behind the 'it rhymes' item in this episode. Processing fluency — the subjective ease with which information is handled — is the mechanism most commonly proposed for the effect. The same account appears in this channel's episode on remembering, where fluency is mistaken for learning rather than for truth. No effect sizes or percentages are quoted in this episode. Where a precise figure could not be verified against the original paper, the finding is described by its direction and design instead. The week-long chain of five encounters is an invented illustration, not a documented case. No political example is used anywhere in the episode, deliberately.