
Curse words are usually abrupt and harsh sounding, which is less than ideal for recall. (© Andrii Zastrozhnov - stock.adobe.com)
‘Gruhious’ Was Harder to Remember Than ‘Clisious’ — Researchers Have Theories As To Why
In A Nutshell
- The study tested made-up words, not actual profanity. Researchers created nonsense words using sound types that prior research has linked to curse words (harsh plosives like “k” and “g”) versus gentler sounds (flowing consonants like “l” and “s”).
- Words with harsher sounds were remembered less often. Participants recalled 53 percent of words built from softer sounds but only 36 percent of those constructed from harsh consonants and back vowels.
- The “prettiest” sounds didn’t get the highest ratings. Words designed to sound beautiful received similar appeal scores to words designed to sound ugly. Neutral words—built from mid-range sounds—were rated most appealing.
- The reason harsh sounds hurt memory remains unclear. The researchers offer possible explanations involving speech mechanics and negative sound associations, but the study did not test these mechanisms directly.
Linguists have noticed that many swear words share the same sharp, punchy sounds—hard consonants like “k,” “t,” and “g” that burst out of the mouth rather than flow. When a study examined those sound types (not actual profanity) to see whether they affect how well people remember made-up words, the results indicated the answer appears to be yes. However, the reasons remain uncertain.
Researchers at the University of Vienna created twelve nonsense words following English pronunciation rules and asked 100 native speakers to memorize them. Theresa Matzinger and David Košić constructed some words from soft, flowing sounds, such as the “l” and “s” and “m” that linguistic research has found to be underrepresented in curse words. Others were built from plosives and back vowels that prior studies have linked to profanity and negative meaning.
Published in PLOS One, Participants recalled 53 percent of words built from gentler sounds but only 36 percent of those constructed from harsher ones. The pattern held even though none of the test words were actual curse words or carried any meaning at all.
Why Swear Words Sound the Way They Do
Previous research has documented just how consistently harsh sounds cluster in offensive vocabulary. A 2023 study found that swear words across languages are far less likely to contain approximants, a category of consonants that includes sounds like “w” and “l.” Separately, research on English found that curse words contain proportionally more plosives (sounds made by blocking and then releasing airflow, like “k” or “t”) than gentler genres like carols and lullabies. Even invented profanity follows the pattern: when researchers asked people to rate made-up words, those containing plosives were judged as more taboo.
Linguist David Crystal observed back in the 1990s that words people consider beautiful tend to feature sounds that flow rather than stop. Ugly-sounding words, by contrast, contain more abrupt stops and back vowels like the “oo” in “doom.”
Matzinger and Košić used Crystal’s observations to engineer their test words. “Clisious,” “snelious,” and “sleemious” combined flowing consonants with front vowels. “Gruhious,” “dwougious,” and “twuhious” stacked harsh stops and back vowels. A third set fell between these extremes.
Soft Sounds Were Remembered More Often
When participants typed out the words they remembered, a hierarchy emerged. Words from the appealing condition were recalled most frequently. Neutral words came next. Words built from harsh sounds were remembered least often.
Emotional arousal typically boosts memory; we remember things that shock or disturb us. Yet the harsher-sounding words, despite their acoustic similarity to taboo vocabulary, were the least memorable in this experiment. The researchers note that while intensely negative experiences can trigger powerful memories, stimuli that are merely unpleasant without crossing a threshold of emotional intensity may simply fail to register. However, the study did not test this explanation directly.
At the individual level, a modest connection appeared between personal ratings and recall. Words that a given participant rated as more appealing were somewhat more likely to be remembered by that same person. Recalled words averaged 3.93 on a seven-point appeal scale, compared to 3.66 for forgotten words.
When ‘Beautiful’ Sounds Don’t Win Beauty Contests
The memory results aligned with predictions, but the appeal ratings did not. Crystal’s theory predicted that words packed with pleasant sounds would be rated most beautiful. Instead, the neutral words—those built from sounds in the middle of Crystal’s rankings—earned the highest appeal scores. Words designed to sound beautiful and words designed to sound ugly received similarly low ratings.
The researchers suspect this reflects a well-documented pattern in aesthetic psychology: highly familiar stimuli breed boredom while highly unfamiliar ones provoke discomfort. Because many of Crystal’s “beautiful” sounds are also among the most common in English, words built entirely from these sounds may have felt generic rather than appealing.
Participants’ ratings were consistent across two rounds of evaluation, indicating genuine preferences rather than random responses. People knew what they liked, even if their preferences didn’t match the classic theory.
Why were harsh-sounding words remembered less often? The study did not test mechanisms directly, but the researchers offer possibilities. One involves the mechanics of speech itself. Hard plosives require abrupt closure of the vocal tract followed by a burst of air. They interrupt the flow of speech rather than blending into it. Words built from such sounds might resist smooth encoding in memory.
Another possibility draws on sound symbolism—the widespread tendency for certain sounds to evoke consistent meanings across languages. Back vowels associate with darkness and heaviness. Voiceless stops link to sharpness and negativity. If these sounds trigger subtle negative associations, they might also reduce the attention devoted to processing them. Both explanations remain speculative.
From Lab Findings to Real-World Applications
If harsh sounds are associated with poorer memorability, the consequences could extend into commercial and educational territory. Brand names, product labels, and marketing slogans depend on sticking in consumers’ minds. Language learners might acquire vocabulary more efficiently if lessons emphasize words with favorable sound profiles.
Over longer timescales, memory advantages for pleasant-sounding words could shape how languages evolve. Words that resist memorization might fade from use while easily recalled alternatives spread. The sounds we find subtly pleasing may, over centuries, win the competition for space in human vocabularies.
The research adds to evidence that the building blocks of words carry weight we don’t consciously perceive. The harsh syllables that appear frequently in profanity also showed up in the least-remembered pseudowords. Whether this connection reflects a genuine cognitive mechanism or simply a statistical association, future studies will need to determine.
Paper Summary
Limitations
Several constraints affect how broadly these findings apply. The experiment presented words both visually and audibly, meaning visual memory likely influenced recall alongside sound-based memory. All pseudowords shared the “-ious” suffix, potentially confounding the effects of varied sounds. The study enrolled only native English speakers without controlling for regional dialect, so results may not generalize to other languages. Crystal’s original rankings from the 1990s were based on informal observations rather than controlled experiments, and the stimulus design inherited those limitations. Finally, memory tasks using written recall may involve different processes than purely auditory recognition tests.
Funding and Disclosures
This work was supported by a Disruptive Innovation Grant from the Austrian Academy of Sciences and the Austrian Science Fund (grant number: DI_2023-108_MATZINGER_BEALP) awarded to Theresa Matzinger. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. The authors declared no competing interests.
Publication Details
Authors: Theresa Matzinger and David Košić (joint first authors), Department of English and American Studies, University of Vienna; Vienna Cognitive Science Hub, University of Vienna
Journal: PLOS One, Volume 20, Issue 12 | Published: December 3, 2025 | DOI: https://doi.org/10.1371/journal.pone.0336597 | Editor: Li-Hsin Ning, National Taiwan Normal University | Preregistration and Data Availability: All hypotheses, study protocols, and analyses were preregistered on the Open Science Framework (https://osf.io/npe2g/overview). Data and R files are available at the same repository.







