Rhyme-as-Reason Effect: Why a Saying That Rhymes Sounds More True

A statement that rhymes is easier to process than one that does not: the second half is partly predicted by the first, the words arrive with a beat, and the whole thing is simpler to hold in mind and to repeat. That ease, which psychologists call processing fluency, is a feeling, and feelings get attributed to whatever is in front of you. When the thing in front of you is a claim, the ease of taking it in feels like the ease of a claim that is correct. Rolf Reber and Norbert Schwarz showed in 1999 that even the visual fluency of a statement, printed in a high-contrast color rather than a hard-to-read one, raised judgments of its truth.

Matthew McGlone and Jessica Tofighbakhsh called the version that runs on rhyme the Keats heuristic, after the poet's line that beauty is truth. Their 1999 paper in Poetics reported that people judged unfamiliar rhyming aphorisms to be more accurate descriptions of human behavior than paraphrases with the same meaning, even though they rated the two forms as equally comprehensible. Their 2000 paper in Psychological Science repeated the accuracy result and added a condition in which participants were cautioned to judge the content and not the poetry. In that condition the effect was attenuated, which is the signature of a fluency effect: the feeling is still there, but it is no longer attributed to truth.

Fluency is a family of cues rather than one. Adam Alter and Daniel Oppenheimer's 2009 review lists repetition, legibility, pronounceability, rhyme, and conceptual familiarity among the sources, and argues that they all feed one metacognitive signal that people read as truth, liking, confidence or familiarity depending on the question being asked. Rhyme is a strong member of the family because it adds predictability and memorability at once; a rhyming line is also easier to recall later, and a line that comes to mind easily feels true for that reason too.

The effect does not require anyone to believe that rhyming makes things true. Few people would endorse rhyme as evidence if asked, and the participants in the 2000 studies rated the rhymes higher anyway, until they were told to watch for it. That is what makes the effect worth knowing: it operates on people who would reject it as a principle, and it is cheap to exploit.

Matthew McGlone and Jessica Tofighbakhsh published "Birds of a feather flock conjointly (?): Rhyme as reason in aphorisms" in Psychological Science in 2000. The title is its own demonstration: the familiar rhyme has been broken by swapping one word, and the result sounds like a weaker claim though it says the same thing. The materials were unfamiliar aphorisms, chosen so that participants could not rate them from memory of having heard them before. Each existed in two forms: the original, which rhymed, and a modified version in which one word was replaced by a synonym that broke the rhyme while preserving the meaning. Their example pair was "What sobriety conceals, alcohol reveals" and "What sobriety conceals, alcohol unmasks."

Participants read the aphorisms and judged how accurately each described human behavior. Any one participant saw each aphorism in only one of its forms, so the comparison between rhyming and non-rhyming versions was made across people for the same saying, not by showing anyone a pair side by side. The rhyming originals were judged more accurate than the modified versions. In a further condition, participants were cautioned beforehand to base their judgments on the claim being made and not on the way it was phrased, and the advantage of the rhyming versions was attenuated.

The authors interpreted the result as a fluency effect. Rhyme, like repetition, makes a statement easier to process, and the ease is misattributed to conviction about the statement's truth. The warning worked because it gave participants a different explanation for the feeling, which is the standard way a fluency effect is switched off.

The demo on this page differs from the experiment in one respect that matters. Because the engine compares one visitor's own ratings, it cannot show you the same saying in both forms without revealing the trick, so the six plain items are paraphrases of six other rhyming proverbs that are not shown. Your two sets therefore differ in subject as well as sound, and a gap between them is consistent with the effect rather than a measurement of it; the between-groups comparison in the 2000 paper is the evidence that rhyme itself is the cause.

The rows below are illustrations of where sound stands in for argument, not findings from the studies cited on this page.

The 2000 study found its own counter: participants who were warned to separate poetry from content showed a smaller effect. The habits below are versions of that warning.

• Paraphrase before you judge. If a claim rhymes, say it again in words that do not, and rate the plain version. The original study's modified aphorisms are exactly this move.
• Notice the feeling and name its cause. Fluency is misattributed when it has no other explanation; once you have said "this is easy to take in because it rhymes," the ease has somewhere else to go.
• Ask what the saying would need to be true. A rhyme compresses a claim about the world into a line; the line has no room for the conditions under which the claim holds.
• Be most careful when a line is designed. Slogans, taglines and catchphrases are written by people who know the effect, and the better the line sounds, the less likely it is that the sound is an accident.
• Apply the same discount to your own writing. A rhyming heading will be believed a little more than it deserves, which is a reason to use one only when the claim can bear it.

The rhyme-as-reason effect is one member of the fluency family, and its relatives are easy to mistake for it.

• It is not the illusory truth effect, in which a statement is judged truer because it has been encountered before. That is fluency from repetition; rhyme produces fluency on the first hearing, with no prior exposure.
• It is not the mere exposure effect, which is about liking rather than truth: things seen before are liked more. Rhyme can raise liking too, but the effect on this page is on judgments of accuracy.
• It is not belief bias, where an argument is judged valid because its conclusion is already believed. Rhyme-as-reason works on unfamiliar claims the hearer has no prior view of, which is why the 2000 study used obscure aphorisms.
• It is not authority bias. A rhyming line carries no credentials; it persuades by its form alone.