Non-arbitrary attachment of sounds to object shapes
This picture is used as a test to demonstrate that people may not attach sounds to shapes arbitrarily. When given the names "kiki" and "bouba", many cultural and linguistic communities worldwide robustly tend to label the shape on the left "kiki" and the one on the right "bouba".
Thebouba–kiki effect (/ˈbuːbəˈkiːkiː/) ortakete–malumaphenomenon[1][2][3] is a non-arbitrarymental association between certain speech sounds and certain visual shapes. The most typical research finding is that people, when presented withnonsense words, tend to associate certain ones (likebouba andmaluma) with a rounded shape and other ones (likekiki andtakete) with a spiky shape. Its discovery dates back to the 1920s, when psychologists documented experimental participants as connecting nonsense words to shapes in consistent ways.
There is a strong general tendency towards the effect worldwide; it has been robustly confirmed across a majority of cultures and languages in which it has been researched,[4] for example including among English-speaking American university students,Tamil speakers in India, speakers of certain languages with no writing system, young children, infants, and (though to a much lesser degree) thecongenitally blind.[4] It has also been shown to occur with familiar names. The bouba–kiki effect is one form ofsound symbolism.[5]
This effect was first observed by GeorgianpsychologistDimitri Uznadze in a 1924 paper.[6][non-primary source needed] He conducted an experiment with 10 participants who were given a list with nonsense words, shown six drawings for five seconds each, then instructed to pick a name for the drawing from the list of given words. He describes the different "strategies" participants developed to match words to drawings and quotes their reasoning. He also describes situations where participants described very specific forms that they associated with a nonsense word, without reference to the shown drawings. He develops a theory of four factors that influence the way names for objects are decided.
In total, there were 42 words. For one particular drawing, 45% picked the same word. For three others, the percentages were 40%. Uznadze points out that this is significantly more overlap than one could expect, given the high number of possible words. He speculates that there must therefore be certain regularities "which the human soul follows in the process of name-giving".
German AmericanpsychologistWolfgang Köhler referred to Uznadze's experiment in a 1929 book[7] which showed two forms and asked readers which shape was called "takete" and which was called "maluma". Although he does not say so outright, Köhler implies that there is a strong preference to pair the jagged shape with "takete" and the rounded shape with "maluma".[8]
In 2001,V. S. Ramachandran and Edward Hubbard repeated Köhler's experiment, introducing the words "kiki" and "bouba", and asked American college undergraduates andTamil speakers in India, "Which of these shapes is bouba and which is kiki?" In both groups, 95% to 98% selected the curvy shape as "bouba" and the jagged one as "kiki", suggesting that the human brain somehow attaches abstract meanings to the shapes and sounds consistently.[9][failed verification –see discussion]
A research experiment was conducted in 2022 that found evidence supporting the idea that the bouba/kiki effect is across-cultural phenomenon. 917 participants speaking 25 different languages, with 10 different writing systems, maintain a higher than chance consistency in bouba/kiki identification, intuitively associating the "bouba" with a rounded shape and "kiki" with a sharp, pointed shape, regardless of their native language, though the effect is stronger in some languages than others. It also supports thatRoman orthography is a factor that could enhance the bouba/kiki effect. However, this biasing effect of orthography is rather weak since the participants that speak languages with Roman orthography are only marginally more likely to show the bouba/kiki effect.[clarification needed][4]
Daphne Maurer and colleagues showed that even children as young as 21⁄2 years old may show this preference.[10] More recent work by Ozge Ozturk and colleagues in 2013 showed that even 4-month-old infants have the same sound–shape mapping biases as adults and toddlers.[11] Infants are able to differentiate between congruent trials (pairing an angular shape with "kiki" or a curvy shape with "bubu") and incongruent trials (pairing a curvy shape with "kiki" or an angular shape with "bubu"). Infants looked longer at incongruent pairings than at congruent pairings. Infants' mapping was based on the combination ofconsonants andvowels in the words, and neither consonants nor vowels alone sufficed for mapping. These results suggest that some sound–shape mappings precedelanguage learning, and may in fact aid in language learning by establishing a basis for matching labels to referents and narrowing the hypothesis space for young infants. Adults in this study, like infants, used a combination of consonant and vowel information to match the labels they heard with the shapes they saw. However, this was not the only strategy that was available to them. Adults, unlike infants, were also able to use consonant information alone and vowel information alone to match the labels to the shapes, albeit less frequently than the consonant–vowel combination. When vowels and consonants were put in conflict, adults used consonants more often than vowels.
The effect has also been shown to emerge in other contexts, such as when words are paired with evaluative meanings (with "bouba" words associated with positive concepts and "kiki" words associated with negative concepts)[12] or when the words to be paired are existing first names, suggesting that some familiarity with the linguistic stimuli does not eliminate the effect. A study showed that individuals will pair names such as "Molly" with round silhouettes, and names such as "Kate" with sharp silhouettes. Moreover, individuals will associate different personality traits with either group of names (e.g., easygoingness with "round names"; determination with "sharp names"). This may hint at a role of abstract concepts in the effect.[13]
Other research suggests that this effect does not occur in all communities,[14] and it appears that the effect breaks if the sounds do not make licit words in the language.[15] The bouba–kiki effect seems to be dependent on a longsensitive period, with high visual capacities in childhood being necessary for its typical development. Although the congenitally blind have been reported to show a bouba–kiki effect, they show a much smaller one for touched shapes than sighted individuals do for visual shapes.[16][17]
A major 2021 study showed that certain languages, namely Mandarin Chinese, Turkish, Romanian, and Albanian, on average showed lower-than-50% matches for both associating bouba with roundedness and kiki with jaggedness. However, the authors consider their analysis conservative and not clear enough to confirm if these four definitively lacked the bouba–kiki phenomenon. For example, the phonetic structures of these languages or their participants' cultural associations with sound and shape could have led to the weaker correlations observed.[4] Further research is being conducted to further verify the correlation between low-effect languages and the bouba-kiki phenomenon.
In 2019, Nathan Peiffer-Smadja and Laurent Cohen published the first study usingfMRI to explore the bouba–kiki effect.[18] They found that prefrontal activation is stronger to mismatching (bouba with spiky shape) than to matching (bouba with round shape) stimuli. A subsequent study by Kelly McCormick and colleagues reported a similar pattern of greater activation for mismatched word-shape stimuli, but with most activity inparietal regions including theintraparietal sulcus andsupramarginal gyrus, regions known to play a role in sensory association and perceptual-motor processing.[19] Peiffer-Smadja and Cohen also found that sound-shape matching also influences activations in the auditory and visual cortices, suggesting an effect of matching at an early stage insensory processing.[18]
Ramachandran and Hubbard suggest that the kiki/bouba effect has implications for the evolution of language, because it suggests that the naming of objects is not completely arbitrary.[9]: 17 The rounded shape may most commonly be named "bouba" because the mouth makes a more rounded shape to produce that sound while a more taut, angular mouth shape is needed to make the sounds in "kiki".[20] Alternatively, the distinction may be betweencoronal ordorsal consonants like/k/ andlabial consonants like/b/,[21] or, as Fort and Schwartz suggest, the difference may be attributed to the noise a "bouba" shape makes when bounced (lower frequency and more continuous) in comparison to a spiked object.[22] Additionally, it was shown that it is not only different consonants (e.g., voiceless versus voiced) and different vowel qualities (e.g., /a/ versus /i/) that play a role in the effect, but also vowel quantity (long versus short vowels). In one study, participants rated words containing long vowels to refer to longer objects and short vowels to short objects, at least for languages that make avowel length distinction.[23] The presence of these "synesthesia-like mappings" suggest that this effect may be the neurological basis forsound symbolism, in which sounds are non-arbitrarily mapped to objects and events in the world.[citation needed] Research has also indicated that the effect may be a case ofideasthesia,[24] a phenomenon in which activations of concepts (inducers) evoke perception-like experiences (concurrents). The name comes from the Greekidea andaisthesis, meaning "sensing concepts" or "sensing ideas", and was introduced by Danko Nikolić.[25]
^Nielsen, Alan; Rendall, Drew (2011). "The sound of round: evaluating the sound-symbolic role of consonants in the classic Takete-Maluma phenomenon".Canadian Journal of Experimental Psychology.65 (2):115–124.doi:10.1037/a0022268.PMID21668094.
^abcdĆwiek, Aleksandra; Fuchs, Susanne; Draxler, Christoph; Asu, Eva Liina; Dediu, Dan; Hiovain, Katri; Kawahara, Shigeto; Koutalidis, Sofia; Krifka, Manfred; Lippus, Pärtel; Lupyan, Gary; Oh, Grace E.; Paul, Jing; Petrone, Caterina; Ridouane, Rachid; Reiter, Sabine; Schümchen, Nathalie; Szalontai, Ádám; Ünal-Logacev, Özlem; Zeller, Jochen; Perlman, Marcus; Winter, Bodo (2022)."Thebouba/Kiki effect is robust across cultures and writing systems".Philosophical Transactions of the Royal Society B: Biological Sciences.377 (1841).PMC8591387.PMID34775818. This paper examines cross-linguistic data supporting the bouba/kiki effect and explores its potential implications for understanding iconicity in language.
^Bross, Fabian (2018). "The Good, the Bad, the Bouba, and the Kiki. Cross-Modal Correspondences Between Evaluative Meanings, Speech-Sounds, and Object Shapes".14th conference "Phonetics & Phonology in the German-Speaking World". University of Vienna.doi:10.13140/RG.2.2.11463.14240.
^McCormick, Kelly; Kim, Jee Young; List, Sara; Nygaard, Lynne C. (2015)."Sound to Meaning Mappings in the Bouba-Kiki Effect"(PDF).Proceedings of the 37th Annual Conference of the Cognitive Science Society: Mind, Technology, and Society: Pasadena, California, 23–25 July 2015. Austin, TX: Cognitive Science Society. pp. 1565–1570.ISBN978-0-9911967-2-2. Archived fromthe original(PDF) on 9 February 2019. Retrieved8 February 2019.
Chen, S., & Maurer, D. (2022).The Bouba/Kiki Effect in Early Childhood: Evidence from Preschoolers. Developmental Science, 25(5), e13277. This developmental study provides insights into how young children perceive sound-symbolic shapes, indicating that the bouba/kiki effect may emerge in early childhood, potentially pointing to innate sound-shape associations that become refined with language exposure.
Allison Parshall, "Pain Language: The sound of 'ow' transcends borders",Scientific American, vol. 332, no. 2 (February 2025), pp. 16–18. "Manylanguages have aninterjection word for expressing pain. [Katarzyna Pisanskiet al., writing in theJournal of the Acoustical Society of America, have] found that pain interjections tend to contain thevowel sound 'ah' (written as [a] in theInternational Phonetic Alphabet) and letter combinations that incorporate it, such as 'ow' and 'ai.' These patterns may point back to the origins of human language itself." (p. 16.) "Researchers are continually discovering cases ofsymbolism, or soundiconicity, in which a word's intrinsic nature has some connection to its meaning. These cases run counter to decades oflinguistic theory, which had regarded language as fundamentally arbitrary... [Many wordsonomatopoeically imitate a sound. Also] there's the'bouba-kiki' effect, whereby people from varying cultures are more likely to associate the nonsense word 'bouba' with a rounded shape and 'kiki' with a spiked one.... [S]omehow we all have afeeling about this,' says Aleksandra Ćwiek... [She and her colleagues have] show[n] that people associate thetrilled 'R' sound with roughness and the 'L' sound with smoothness.Mark Dingemanse... in 2013 found [that] the conversational 'Huh?' and similar words in other languages may be universal." (p. 18.)
