University of Dundee
Sounds sweet, sounds bitter
Pathak, Abhishek; Calvert, Gemma A.
Published in:Food Quality and Preference
DOI:10.1016/j.foodqual.2020.103918
Publication date:2020
Licence:CC BY-NC-ND
Document VersionPeer reviewed version
Link to publication in Discovery Research Portal
Citation for published version (APA):Pathak, A., & Calvert, G. A. (2020). Sounds sweet, sounds bitter: How the presence of certain sounds in a brandname can alter expectations about the product’s taste. Food Quality and Preference, 83, 1-10. [103918].https://doi.org/10.1016/j.foodqual.2020.103918
General rightsCopyright and moral rights for the publications made accessible in Discovery Research Portal are retained by the authors and/or othercopyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated withthese rights.
• Users may download and print one copy of any publication from Discovery Research Portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain. • You may freely distribute the URL identifying the publication in the public portal.
Take down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.
Download date: 02. Feb. 2022
Sounds sweet, sounds bitter: How the presence of certain sounds in a brand name can
alter expectations about the product’s taste
Abhishek Pathak1 and Gemma Anne Calvert2
1 School of Business, University of Dundee, UK
2 Nanyang Business School, Nanyang Technological University, Singapore
Correspondence to: Correspondence concerning this article should be addressed to Dr. Abhishek
Pathak, School of Business, 4 Nethergate, University of Dundee, Scotland, UK (Tele:
+44(01382)884867; E-mail: [email protected])
Author'
s fina
l vers
ion
Author Accepted Manuscript version of Pathak, A & Calvert, GA 2020, 'Sounds sweet, sounds bitter: How the presence of certain sounds in a brand name can alter expectations about the product’s taste', Food Quality and Preference. Released under the terms of a CC-BY-NC-ND License https://creativecommons.org/licenses/by-nc-nd/3.0/
Abstract
This paper examines how certain speech sounds within a brand name can alter expectations
about the product’s taste. Across two studies we demonstrate that the presence of voiced (b, d, g,
z & v) vs. voiceless (p, t, k, s & f) obstruents (speech sounds produced when airflow is
obstructed in the oral cavity) in a chocolate’s brand name can alter its expected taste as bitter vs.
sweet. We propose this is because voiced obstruents are typically of low frequency (frequency
code hypothesis), contain harsh acoustic qualities and evoke negativity (due to aerodynamic
difficulties in their pronunciation). In a third study, we extended these findings to show, using
the Brand Personality Scale (BPS), that the presence of voiced (vs. voiceless) obstruents makes
brand names sound more masculine, rugged and tough and less honest, charming, glamorous,
wholesome, cheerful and sentimental. Research linking consonant sound symbolism (specifically
voiced obstruents) and product attributes is sparse. Most research in this field links vowels with
product attributes, sounds with shapes and music (or musical notes) with taste attributes. We
contribute by extending these findings to voicing (and associated harshness) and the sweet-bitter
continuum of the expected taste.
Keywords: Sound symbolism; voiced obstruents; voicing; brand names; brand personality
traits; taste attributes Author'
s fina
l vers
ion
Highlights
When asked to make a choice, respondents rate food products having voiced obstruents
(b, d, g, z & v) in their brand names as bitter (compared to ones with voiceless
obstruents)
When asked to make a choice, respondents rate food products having voiceless obstruents
(p, t, k, s & f) in their brand names as sweet (compared to ones with voiced obstruents)
Hypothetical brand names with voiced obstruents are perceived as more rugged and less
sophisticated than the brand names with voiceless obstruents
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
1
Sounds sweet, sounds bitter: How the presence of certain sounds in a brand name can
alter expectations about the product’s taste
1 Introduction
A brand’s name typically represents the first touchpoint between a brand and its
consumers, making its selection a vital component of a brand’s marketing strategy (Charmasson,
1988; Klink & Athaide, 2012). Indeed, there is evidence to suggest that up to 40% of a brand’s
success can be attributed to a well-chosen name (Alashban, Hayes, Zinkhan, & Balazs, 2002;
Zaltman & Wallendorf, 1979). Research has shown that during the process of selecting a product
consumers are more influenced by the brand name to a greater extent than other important
determinants (e.g., price and physical appearance) (Dawar & Parker, 1994). There is now a
considerable body of research demonstrating the relationship between brand name sounds and
perceived product attributes (e.g., slowness vs. fastness, heaviness vs. lightness, basic vs. luxury,
feminine vs. masculine). This suggests that consumers often expect some congruency between
the sound of a brand name and its associated product features. Furthermore, when these brand
evoked expectations are consistent with the actual product experience, consumers typically
appreciate them more (Friese, Wänke, & Plessner, 2006). In fact, the relationship between brand
names and product expectancy has been shown to be more salient when the brands (or brand
names) are unfamiliar (Fenko, Lotterman, & Galetzka, 2016).
All languages contain sounds which are perceived as harsh or soft (Aryani, Conrad,
Schmidtke, & Jacobs, 2018). In this paper, we provide evidence for a sound symbolic link
between harsh sounds and expectations of how a product will taste. More specifically, we show
that brand names containing harsh sounds lead to expectations that the associated product will
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
2
taste bitter whereas those with soft sounds generate taste expectations of a sweet tasting product.
We used voiced (vs. voiceless) obstruents, the speech sounds produced when airflow is
obstructed in the oral cavity, in hypothetical brand names to imbue them with a sense of
harshness. This paper builds on previous research in that here (i) we use voicing in consonants to
induce perceptual harshness [previous research has used ‘hard’ vs ‘soft’ consonants (e.g., /k/ vs
/l/)] with taste attributes, and (ii) other research has shown voicing (vs. non-voicing) to be
associated with roundness (vs. spikiness) and roundedness with sweetness (vs. spikiness with
bitterness). These findings suggest an association of voicing (vs. non-voicing) with sweetness
(vs. bitterness). We on the contrary show an opposite association of voicing (vs. non-voicing)
with bitterness (vs. sweetness).
1.1 Sound symbolism and taste
A vast body of research has now shown a systematic relationship between sounds and
their perceived meanings (e.g., Hinton, Nichols, & Ohala, 1994; Ramachandran & Hubbard,
2001; Sapir, 1929). This relationship has been shown in the context of names (e.g., Sidhu &
Pexman, 2015; Slepian & Galinsky, 2016), psycholinguistic elements (e.g., Nuckolls, 1999),
shapes (e.g., Ramachandran & Hubbard, 2001), product attributes (e.g., Klink, 2000), personality
traits (e.g., Klink & Athaide, 2012) and more specifically, related to the current paper, with taste
perceptions (see Spence, 2012 for a review), to name but a few.
Fónagy (1963) was the first to suggest a relationship between taste (on a sweet-bitter
continuum) and the sound of articulated front vs. back vowels. For example, amongst Hungarian
respondents, /i/ was found to be sweeter, and /u:/ as bitter. A similar relationship was shown by
Yorkston and Menon (2004) albeit in the context of the creaminess of ice-creams. Similarly,
Simner, Cuskley, and Kirby (2010) have linked vowel height and back-ness with tastes. In their
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
3
studies, higher vowels tended to signify sweetness while lower vowels connoted bitterness and
sourness. Similarly, back vowels were linked to sweetness while front ones indicated bitterness
and sourness. More recently, sharper speech sounds (e.g., takete, kiki) have been linked with
sharp objects (e.g., spiky shapes) and bitter tastes (vs. softer sounds, e.g. maluma, lula, with
rounded objects and sweet tastes) (Gallace, Boschin, & Spence, 2011). A similar cross-modal
relationship has been reported by Ngo, Misra, and Spence (2011). These researchers showed that
chocolates containing only 30% cocoa (i.e. sweeter chocolates) were more easily matched with
rounded speech sounds (e.g., maluma) whereas chocolates containing 70% cocoa (i.e. bitter
chocolates) were more closely matched with sharper speech sounds (e.g., takete). The association
between sounds (e.g., musical notes) and tastes has also been explored in depth (e.g., Bronner,
Frieler, Bruhn, Hirt, & Piper, 2012; Crisinel & Spence, 2009; Knoferle, Woods, Käppler, &
Spence, 2015; Knöferle & Spence, 2012), but the relationship between speech sounds and tastes
has not been as extensively researched as those between speech sounds and other product
attributes (e.g., fastness, heaviness, lightness) (e.g., Abel & Glinert, 2008; Klink, 2000; see
Spence, 2012 for a review). The current paper is an attempt to bridge this gap.
So why should voiced obstruents signify bitterness while voiceless ones connote
sweetness? We propose two mechanisms that may explain this phenomenon. The first is the
frequency code hypothesis (FCH) and the second is the aerodynamic difficulties associated with
the pronunciation of voiced obstruents. Voiced (vs voiceless) obstruents have been shown to
have lower (vs. higher) frequencies (Kingston & Diehl, 1995; Kingston, Diehl, Kirk &
Castleman, 2008) and low frequency sounds have been shown to be more closely associated with
warnings, threat and hostility (vs. pleasure, appeasement and friendliness for higher frequency
sounds) (Morton, 1977; Ohala, 1994). Similarly, low (vs. high) frequencies have been indirectly
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
4
(e.g., music) shown to be associated with bitter (vs. sweet) tastes (Simner et al., 2010; Knöferle
& Spence, 2012; Spence, 2012).
The second mechanism that may explain these obstruent-taste associations is the
significant aerodynamic difficulties associated with the pronunciation of voiced obstruents (see
Ohala, 1983 for a review). These difficulties have often been cited as the reason why voiced
obstruents are associated with negative meanings or perceptions (Kawahara, Shinohara, &
Uchimoto, 2008; Kubozono, 1999; Ohala, 1983). Voiced obstruents have also been shown to be
associated with perceived heaviness and largeness (e.g., large and dangerous animals)
(Kubozono, 1999; Saji, Akita, Imai, Kantartzis, & Kita, 2013). Other research suggests that,
when used in the names of fictitious characters, voiced obstruents evoke feelings of a bigger,
larger and more threatening character. For example, studies using Pokémon characters in
Japanese and English speakers show that a larger and more menacing character’s name is
considered more appropriate if it contains voiced obstruents (Kawahara & Moore, 2018; Shih,
Ackerman, Hermalin, Inkelas, & Kavitskaya, 2018).
In the current paper, we extend this relationship between negativity and harshness of
sounds associated with voiced obstruents to the taste dimension. We aim to show that when used
in the context of brand names (BNs) of chocolates, consumers consider names containing harsh
speech components as signifying that the chocolate is bitter. Chocolate was chosen as the
preferred product category as this has been successfully used in past research on cross-modal
associations, specifically on the sweet-bitter continuum (e.g., Ngo et al., 2011; Ngo & Spence,
2011; Spence & Gallace, 2011). Auth
or's f
inal v
ersion
VOICING AND TASTE ATTRIBUTES
5
1.2 Speech sounds
Consonants and vowels form the two building blocks for all languages. Vowels are
typically smoother, do not involve much movement of the articulatory muscles and are produced
by the movement of the air within the oral cavity with little obstruction. The production of
consonants, on the other hand, typically involves a higher degree of obstruction, friction and
constriction of the vocal tract (Carrol, 2004; Ladefoged, 1993). Speech sounds can be broadly
classified into obstruents and resonants. Obstruents are sounds made due to a partial (or full)
obstruction of the airflow within the oral cavity, whereas resonants are produced without any
obstruction to the airflow. Obstruents can further be divided into, stops (e.g., /p/, /b/), fricatives
(e.g., /f/, /v/), and affricates (e.g., /t∫/ as in Chin, /dʒ/ as in General). Within these, voiced
obstruents is a sub-class of sounds where an additional voicing accompanies the phonemic
sounds (e.g., /b/, /ɡ/, /z/). Voicing is the vibration of the vocal cords during the pronunciation of
some speech sounds. For example, feel the vibration of the vocal cords on Adam’s apple while
saying /b/ (voiced) vs /p/ (unvoiced) (Slepian & Galinsky, 2016)]. The accompanied voicing and
modulation of the vibrating vocal cords, is believed to make the voiced obstruents sound even
harsher compared to the smoother, unmodulated sound of the voiceless obstruents.
Voicing is a complicated process, involving complex articulatory movements and
aerodynamic control. For voicing to occur, two essential conditions have to be met. Firstly, there
should be an adequate amount of tension in the vocal cords, and secondly, air must be expelled
from the lungs into the vocal cords to vibrate them. While tensing the vocal cords is easy, the
process of vibrating them is not so. The entire oral cavity consists of two smaller spaces
demarcated by the glottis, the oral cavity (having an intra oral pressure of Po) and subglottal
cavity (having sub glottal pressure of Ps). The vocal cord vibration requires a differential of
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
6
pressure (approx. 2–3 cm H20 in normal speaking conditions) between the oral cavity and
subglottal cavity. To initiate voicing, the speaker has to actively lower the Po so that Po < Ps and
the air flows across the vocal cords to vibrate them (Ohala, 1983; Shinohara & Kawahara, 2010).
In the case of voiced obstruents, the process is further complicated due to the obstruction of the
airways, which makes the condition of making Po< Ps more difficult (as the intra oral pressure
tends to rise due to the obstruction) (Hayes & Stivers, 1996). In order to reduce the Po, the
speaker resorts to complex manoeuvers such as extending the oral cavity by lowering the larynx,
raising the velum or expanding the cheeks (Ohala, 1983; Shinohara & Kawahara, 2010).
Moreover, for the voicing to sustain, the speaker has to continuously maintain this pressure
differential across the cavities, otherwise the vocal cord vibration will cease and so in turn will
the voicing. This complex exercise of voicing, also called Aerodynamic Voicing Constraint
(Ohala, 2011), takes less than 15 ms (Chollet, Di Benedetto, Esposito, & Marinaro, 2012).
Specifically, based on voicing there can be three types of voiced (vs. voiceless)
obstruents: stops (voiced & voiceless; /p/ & /b/; /t/ & /d/; /k/ & /g/), fricatives (voiced &
voiceless; /f/ & /v/; /θ/& /ð/ as in Theatre vs. Then; /s/ & /z/ as in See vs. Zoo; /ʃ / & /ʒ/ as in
She vs. Measure) and affricates (/tʃ/ & /dʒ/). Out of these voiced obstruents, in the current paper
we excluded /θ/ & /ð/ and /ʃ/ & /ʒ/ as the differences in these speech sounds is purely auditory
and not visual (e.g., a hypothetical name Thodake, can be pronounced by a respondent in two
different ways, one with /θ/ as in Theatre and other with /ð/ as in /Then/ and in a visual stimuli,
these can be a confound1)
2 Methods (common to all the studies)
Twenty-five hypothetical brand names (HBNs) in a CV-CV-CV structure were created
using voiced (/b/, /d/, /g/, /z/, and /v/) vs. unvoiced (/p/, /t/, /k/, /s/ and /f/) obstruents (Appendix
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
7
1) and the HBNs were tested for similarity with the existing words or known brands. To reduce
the known sound symbolic associations linked to vowels, three different vowels were used
within the HBNs. All studies were designed on Inquisit 5 software (millisecond.com) and
participants were native English speakers, recruited from the USA through Amazon Mechanical
Turk (Paolacci & Chandler, 2014; also see Woods, Velasco, Levitan, Wan, & Spence, 2015 for a
recent review on the online experimental research). The Institutional Review Board of a large
South East Asian university approved the research and respondents who consented to participate,
were paid for their time and effort. Each respondent was allowed to participate only in one of the
studies. In all studies, participants were asked to guess the purpose of the experiment; none of the
participants could correctly guess the hypotheses, except for two participants in Study 1, whose
data was excluded. With N≈60, the power to detect a medium-sized effect (0.5) in the Wilcoxon
Signed Rank tests and paired t-tests (α = 0.05) was found to be, 1 – β = 0.96 (approx.) (using
G*Power 3.1.9.2; Faul, Erdfelder, Lang, & Buchner, 2007).
3 Study 1
3.1 Participants
A total of 61 participants between the ages of 24 to 70 years completed the study (2
participants guessed the hypothesis to some extent and were excluded from analysis); Mage =
40.58 yrs., SD = 12.83, Males = 37, Females = 22). All participants were English speakers (one
participant also knew the Thai language).
3.2 Procedure and design
Participants were told that a company was launching two new chocolates (one sweet and
one bitter) in an international, non-English speaking market and was looking for suitable brand
names for them. They were then shown fourteen HBN pairs (one with voiced obstruents and
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
8
another with voiceless obstruents), chosen randomly from the 25 HBNs created. Each HBN was
presented only once. Participants were asked to ‘drag and drop’ the HBNs to one of the product
categories (sweet and bitter) shown on top, which they felt was appropriate for the HBN shown.
Position mapping of the HBNs was counterbalanced within-subjects and those of the product
categories was counterbalanced between-subjects. Before the study, participants were
familiarized with the ‘drag and drop’ procedure in a few practice trials, where real brands were
used (e.g., Mars). Participants also signed an electronic consent and provided demographic
information before commencing the study.
3.3 Results
The results show that participants rated HBNs with voiced (vs. voiceless) obstruents as
more appropriate for a bitter (vs. sweet) chocolate. It was labelled as a congruent response for
our analysis, incongruent being voiced obstruents with sweet chocolate, and vice-versa; data was
segregated into total number of congruent vs. incongruent trials, Z = 25.10, p < 0.0001, r = 0.17.
To explore the individual differences across participants, a Wilcoxon Signed Rank test was
conducted which revealed significant differences between the proportions of congruent and
incongruent responses (M congruent = 58.72, SD = 20.72, M Incongruent = 41.28, SD = 20.72, Z = 2.94,
p = 0.003, r = 0.27). In the English language, the phoneme /g/ can have a different sound
depending on its usage e.g., a hard /g/ sound (as in Gate) or a soft /g/ sound (as in Gem).
Research suggests that often the phoneme /g/ is pronounced as hard whenever it precedes vowels
/a/, /o/ and /u/, and as a soft sound, whenever it precedes the vowels /e/, /i/ and /y/ (Emerson,
1997). In view of this confound, we reanalyzed the data after excluding the affected HBNs.
Specifically, seven such HBNs (e.g., Zadoge) were found. The affected trials were excluded and
the rest of the data was reanalyzed. The results showed that participants rated HBNs with voiced
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
9
(vs. voiceless) obstruents as more appropriate for a bitter (vs. sweet) chocolate. Here, we refer to
these as congruent (voiced with bitter) and incongruent (voiceless with sweet) combinations, Z =
22.08, p < 0.0001, r = 0.19. Across participants, a Wilcoxon Signed Rank test showed significant
differences between the proportions of congruent and incongruent responses (M congruent = 59.53,
SD = 22.32, M Incongruent = 40.49, SD = 22.32, Z = 3.10, p = 0.002, r = 0.29). The results of Study
1 provide evidence for a link between brand names (voiced vs. voiceless) and expected taste
attributes (bitter vs. sweet).
4 Study 2
Study 1 employed a forced-choice task. Forced-choice tasks have been used extensively
by scholars working in the field of sound symbolism. However, they have received some
criticism (Lockwood, Dingemanse, & Hagoort, 2016; Lockwood, Hagoort, & Dingemanse,
2016; Monaghan, Mattock, & Walker, 2012; Saji, Akita, Imai, Kantartzis, & Kita, 2013). Firstly,
the contrast between two given choices is often maximal (i.e. often the choices presented have a
maximum possible difference, e.g., very angular vs. very rounded or the two extremes of a
continuum). Secondly, forced-choice tasks are argued to be unrepresentative of natural language
learning or the way in which we use novel words (e.g., new brand names) (Lockwood,
Dingemanse, & Hagoort, 2016). To control for these potential confounds, in Study 2 we asked
participants to create novel brand names for sweet and bitter chocolates in a free choice task, and
explored the differences in the use of voiced (vs. voiceless) obstruents. In Study 2a, a few
participants used some known English words or random, unpronounceable letters (e.g., SZV) as
the brand names. Consequently the study was repeated (Study 2b) with clearer instructions.
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
10
4.1 Study 2a
4.1.1 Participants
A total of 60 participants between the ages of 24 to 68 years completed the study (Mage =
42.4 yrs., SD = 11.68, Males = 38, Females = 22). All participants were English speakers (seven
participants could speak a foreign language; five knew Spanish, one Arabic and one Chinese).
4.1.2 Procedure and design
Participants were told that a company was launching two new chocolate brands (one sweet
and one bitter) in an international non-English speaking market, and was looking for suitable
brand names for both its products. Participants were then asked to create three pairs of HBNs,
from a given set of letters continually displayed at the top of the computer screen (consonants:
/p/, /b/, /k/, /g/, /t/, /d/, /f/, /v/, /s/, /z/) and vowels: (/a/, /e/, /i/, /o/, /u/). Participants were told to
create BNs, such that the name itself can suggest to the consumer whether the chocolate is sweet
or bitter. Participants then typed in the HBNs in six text boxes in pairs (e.g., HBN Sweet 1 and
Bitter 1) and could proceed further only when they had provided all six HBNs.
4.1.3 Results
The number of voiced (vs. voiceless) obstruents used for creating the HBNs for sweet
(vs. bitter) chocolates were segregated. Only the number of voiced (vs. voiceless) obstruents
were included in the analysis (vowels and consonants (e.g., /r/) other than obstruents were
excluded). Wilcoxon Signed Rank tests revealed that participants chose a significantly higher
number of voiced obstruents in the BNs for bitter (vs. sweet) chocolates, Z = 2.26, p = 0.024, r =
0.21. No difference was observed in the use of voiceless obstruents, Z = 0.83, p = 0.41.
However, the results were not entirely clear-cut as some participants used known English words
as BNs (e.g., ‘delightfully sweet’, ‘better bitter’) which might be argued, does not constitute the
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
11
testing of our hypothesis due to the semantic connotations these words might have.
Consequently, we decided to repeat the study with stricter instructions.
In Study 2b, participants were asked to create brand names with a few additional
instructions, 1) BNs should be pronounceable (i.e., no random placement of letters e.g., SZV), 2)
BNs should contain at least three consonants, 3) BNs should not be an English word or a known
brand name.
4.2 Study 2b
4.2.1 Participants
A total of 60 participants between the ages of 25 to 75 years completed the study (Mage =
42.49 yrs., SD = 12.57, Males = 29, Females = 30; one participant completed the study twice and
his/her data was deleted). All participants were English speakers (one also knew Russian and one
Turkish). The rest of the instructions and experimental design was similar to Study 2a.
4.2.2 Results
As in Study 2a, only the number of voiced (vs. voiceless) obstruents used for creating the
HBNs were included in the analysis. Wilcoxon Signed Rank tests revealed a significantly higher
use of voiced obstruents in the BNs for bitter (vs. sweet) chocolates, Z = 3.39, p = 0.001, r =
0.31. No difference was observed in the use of voiceless obstruents, Z = 1.48, p = 0.14. Study 2
confirms our findings and demonstrates that even in a free choice task, respondents tend to use a
higher number of voiced (vs. voiceless) obstruents for creating names for bitter (vs. sweet)
chocolates. The results also suggest that while voiced obstruents are more associated with the
expectation of bitterness, the same may not be true for the link between voiceless obstruents and
sweetness (as no difference was found in the usage of voiceless obstruents across sweet vs. bitter
chocolate categories).
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
12
5 Study 3
It is important to note that multiple meanings can be associated with a brand name,
depending on the context and goals of a consumer (Motoki et al., 2020). For example, it is likely
that a luxurious sounding brand name can evoke a sense of rarity and uniqueness simultaneously
(Pathak, Calvert, & Lim, 2017; Pathak, Calvert, & Velasco, 2017). In the F&B sector, research
on sound symbolism has mostly linked brand names with gustatory and other sensory attributes
or shapes (e.g. flavor, color, sounds, roundness, spikiness), but not with brand personality traits.
As with any category, a brand name used in connection with a food product does not only signify
the gustatory attributes of the product. While the taste of a food product is important for a brand
in the F&B space, other, non-gustatory brand attributes are also considered in the naming process
(e.g., sincerity and honesty). For a brand across any category, sound symbolic congruency
between the brand name and the multitude of traits that it evokes is important. Indeed, no
company would wish for their brand name to connote sweetness and dishonesty at the same time.
Given this, we feel that it is appropriate to explore the link between sound symbolic taste
attributes with the other brand personality dimensions. To our knowledge, there has been only
one study (e.g., Klink & Athaide, 2012) that has linked brand personality traits with sound
symbolic attributes (though not related to taste per se). Therefore, in Study 3, we aimed to extend
this research further, and explored the link between sound symbolic attributes of taste (sweetness
vs. bitterness) with well-known brand personality traits.
To date, sound symbolic associations related to brand names have been found for a large
pool of product traits (e.g., slow, fast, luxury, heavy, sharp, round, feminine). The same BN has
been shown to evoke multiple simultaneous connotations. For example, apart from being sweet
or bitter, a brand name can also be perceived as round or spiky. Many previous studies have
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
13
linked sounds with shapes (e.g., Kovic, Plunkett, & Westermann, 2010; Ramachandran &
Hubbard, 2001). This body of research suggests that voiced consonants are generally considered
rounded, and voiceless consonants spiky (D’Onofrio, 2014; McCormick, Kim, List, & Nygaard,
2015). These results have been demonstrated using a mix of consonants and not exclusively with
obstruents alone, as in the current paper. In order to explain our results, we also explored the
association of voiced (vs. voiceless) obstruents with roundedness (vs. spikiness), but our results
were inconclusive. We did not find a link between voiced obstruents and spikiness or vice versa
(61 participants were asked to match the HBNs with either rounded or spiky shapes [similar to
D’Onofrio’s (2014) and McCormick et al.’s, (2015) study], Voiced with roundedness = 0.54, Voiced
with spikiness = 0.46, SD = 0.22, t (60) = 1.58, p = 0.12; Voiceless with roundedness = 0.53, Voiceless with
spikiness = 0.47, SD = 0.20, t (60) = 1.30, p = 0.20).
Research suggests that consumers create an image of a brand based on the sound of its
name, even if it is hypothetical2 (Klink & Athaide, 2012). Moreover, many of the perceived
attributes of a brand name are sound-symbolically congruent with each other, for example,
spikiness, harshness, bitterness and angularity are all linked attributes of a brand. Similarly, all
brands are known to have a personality with many dimensions (Aaker, 1997; Eisend &
Stokburger-Sauer, 2013). In Study 3, we aimed to explore the other linked sound symbolic
dimensions of the brand personality. For example, it is likely that a bitter/harsh sounding food
brand name is also perceived as more masculine (vs. feminine), rugged (vs. soft), tough (vs.
soft), dishonest (vs. honest), less cheerful (vs. less cheerful) and less friendly (vs. more friendly).
Klink and Athaide (2012) have used the Brand Personality Scale (BPS) to explore the
sound symbolic associations of HBNs and in Study 3, we use a similar method to explore
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
14
specifically, three BPS factors (ruggedness, sophistication and sincerity), eight BPS facets
contained within these factors (outdoorsy, tough, charming, upper-class, down-to-earth, honest,
wholesome and cheerful) and twenty-two BPS traits contained within these facets (masculine,
outdoorsy, western, rugged, tough, charming, feminine, glamorous, good looking, smooth,
upper-class, down-to-earth, family oriented, small town, honest, sincere, real, wholesome,
original, cheerful, sentimental and friendly).
5.1 Participants
A total of 61 participants between the ages of 22 to 67 years completed the study, M age =
38.59 yrs., SD = 10.62, Males = 30, Females = 31. All participants were native English speakers
(one participant knew Western Armenian and another Spanish).
5.2 Procedure and design
Participants were told that all brands have a personality (e.g., BMW is upper-class and
Nike is sporty) and in this study we were interested in finding out the personality traits of two
chocolate brands, judged purely on the basis of their names. Participants were told that they
would be shown two new brand names for chocolates. They were asked to read the name aloud
and to rate them on different personality trait dimensions displayed on the computer screen [e.g.,
‘How tough do you think this brand is?’ on a Likert scale from 1 (not at all) to 11 (very much)].
There were ten practice trials to familiarize them with the procedure, in which car brands were
used (e.g., Ford, Mercedes). Participants then rated only two HBNs (one each of voiced and
voiceless, e.g., Satoke vs. Zadoge) for all the 22 BPS traits. Presentation of the HBNs was
random and counterbalanced between participants, with a break in between. The HBN being
rated was continuously shown on top of the screen, while the question on the BPS trait kept on
changing.
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
15
5.3 Results
Ratings of the BPS traits for the HBN voiced vs. HBN voiceless were compared using paired
sample t-tests; BPS trait ratings were also collapsed into eight BPS facets and three BPS factors
to compare (see Table 1). The means of the BPS factors and facets were calculated as the
averages of the means of the associated BPS traits with them. For example, the BPS facet of
outdoorsy (M Voiced = 5.34) was computed as the average of the BPS traits of masculine (M Voiced
= 6.34), outdoorsy (M Voiced = 5.62) and western (M Voiced = 4.07) (see Table 1 for the mappings
of BPS traits on to the BPS facets and factors). Out of the three BPS factors, voiced (vs.
voiceless) HBNs were found to be significantly different in their perceived ruggedness and
sophistication, but not on the sincerity factor (Table 1 & Figure 1). In a view to control the Type
1 errors, three univariate repeated measures ANOVA were conducted, with the Likert ratings of
the HBNs (voiced vs. voiceless) as the independent variables and the three BPS factors
(ruggedness, sophistication and sincerity) as the dependent variables. The results revealed that
ruggedness and sophistication factors (ruggedness, F (1, 60) = 9.76, p = 0.003, ηp2 = 0.14;
sophistication, F (1, 60) = 4.91, p = 0.03, ηp2 = 0.08) were significantly different in the voiced vs.
voiceless HBNs, whereas no difference was observed in the sincerity factor (F (1, 60) = 1.99, p =
0.16, ηp2 = 0.03). The results are reported below for each BPS factor.
Insert Table 1 & Figure 1 about here
Ruggedness HBNs with voiced obstruents were found to be significantly more rugged
compared to their voiceless counterparts (Ruggedness Voiced = 5.64, SD = 2.28; Ruggedness
Voiceless = 4.68, SD = 1.90; t (60) = 3.12, p = 0.003, d = 0.40). Within the rugged dimension too,
both toughness (Tough Voiced = 6.90, SD = 2.76; Tough Voiceless = 4.73, SD = 2.28; t (60) = 3.33, p
= 0.002, d = 0.43) and outdoorsy (Outdoorsy Voiced = 5.34, SD = 2.17; Outdoorsy Voiceless = 4.66,
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
16
SD = 1.90; t (60) = 2.38, p =0.02, d = 0.30) facets were found to be significantly higher for the
voiced obstruents than for the voiceless.
Sophistication Similarly, for the sophistication dimension, HBNs with voiceless
obstruents were found to be more sophisticated than their voiced counterparts (Sophistication
Voiced = 5.48, SD = 2.02; Sophistication Voiceless = 6.22, SD = 2.05; t (60) = 2.22, p = 0.03, d =
0.28). Within the sophistication dimension, it was the charming BPS facet which was found to be
different between the HBNs (Charming Voiced = 5.30, SD = 2.03; Charming Voiceless = 6.25, SD =
2.16; t (60) = 2.76, p = 0.008, d = 0.35), but not the upper-class facet (Upper-class Voiced = 5.65,
SD = 2.31; Upper-class Voiceless = 6.18, SD = 2.22; t (60) = 1.39, p = 0.17).
Sincerity No differences were found overall in the BPS factor of sincerity, but within the
sincerity factor, there were differences found in the cheerfulness facet (i.e., cheerful, sentimental
and wholesome BPS traits) (Cheerful Voiced = 5.74, SD = 2.04; Cheerful Voiceless = 6.39, SD = 2.01;
t (60) = 2.08, p = 0.042) and the wholesome facet (only the wholesomeness BPS trait was found
to be different, but not the originality trait, i.e., both type of HBNs were found to be equally
original), Wholesome Voiced = 5.77, SD = 2.16; Wholesome Voiceless = 6.47, SD = 2.33; t (60) =
2.78, p = 0.007; Sentimental Voiced = 5.03, SD = 2.18; Sentimental Voiceless = 5.77, SD = 2.32; t (60)
= 2.10, p = 0.04), no differences were found between honest and down-to-earth BPS facets.
The results of Study 3 suggest that the HBNs with voiced obstruents are perceived to be
more masculine, rugged and tough, but at the same time, less charming, glamorous, honest,
wholesome, cheerful and sentimental than their voiceless counterparts.
6 General discussion
Taste is a multi-sensory experience (Reinoso Carvalho et al., 2015) and not only the
appearance of food, but even cues unrelated to food e.g., lighting (Cho et al., 2015) or color of
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
17
the plate can influence taste expectations and perceptions (Piqueras-Fiszman, Alcaide, Roura, &
Spence, 2012). Topolinski and Boecker (2016) showed that a simple imitation of inwards (i.e.
imitating deglutition) vs. outwards (i.e. imitating regurgitation) muscle movement involved in
spoken words (e.g., brand names) could increase the palatability of the associated food product.
Similarly, sound symbolic associations of non-words (i.e. inwards and outwards) have been
linked with approach vs. avoidance states (Topolinski, Maschmann, Pecher, & Winkielman,
2014) and even with likeability of persons and products (Topolinski, Boecker, Erle, Bakhtiari, &
Pecher, 2017; Topolinski et al., 2014). Even the fluency of a brand name (e.g., ease vs. difficulty
of pronunciation) (Janiszewski & Meyvis, 2001; Topolinski, Likowski, Weyers, & Strack, 2009)
has been shown to affect consumers attitudes (e.g., likability, willingness to pay) towards brands.
The purpose of the current paper was to examine the sound symbolic associations between
voiced (vs. voiceless) obstruents and product taste expectations.
Across three studies, we show that consumers associate HBNs containing voiced
obstruents with a bitter tasting product, and voiceless obstruents with a sweet tasting product. We
demonstrate the results by using a forced-choice task and a free-choice task. By using the BPS
traits, we additionally show the link between other associated (or similar) sound symbolic
dimensions of the voiced obstruents (i.e., higher toughness, ruggedness and masculinity and a
lack of charming, glamorous, honest, wholesome, cheerful and sentimental traits, which were
instead found to be more closely associated with voiceless obstruents). We contribute to the
literature by providing empirical evidence to connect voiced (vs. voiceless) obstruents with a
sensory dimension (i.e. taste) and at the same time integrating the widely followed BPS
framework with the research on sound symbolism. The only other paper which integrates these
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
18
two theories is Klink and Adelaide (2012), in which they link front and back vowels with BPS
traits.
We propose two reasons3 behind our findings: the frequency code hypothesis and
aerodynamic difficulties associated with the utterances of voiced obstruents. In the animal
kingdom, lower frequencies are used to express hostility and threat, and higher frequencies for
friendliness and non-threatening behavior (Morton, 1977). Because of this evolutionary link
between sound harshness (vs. softness) and stressfulness (vs. pleasantness), scientists believe that
across all languages, swear (and cuss, threatening) words often use more voiced obstruents
(Nielsen & Rendall, 2011; Nielsen & Rendall, 2013; Van Lancker & Cummings, 1999). Humans
too associate low (vs. high) frequency sounds with aggressive (vs. submissive) behavior (Dahl,
2010; Krishna, 2012; Pogacar et al., 2018). Similarly, the pronunciation of voiced obstruent
requires extreme aerodynamic control which likely explains why these are rare in many
languages, and some languages do not use them at all (e.g., Cantonese, Hawaiian, Zuñi, Ainu,
Korean, Mandarin) (Major & Faudree, 1996; Ohala, 2011; Żygis, Fuchs, & Koenig,
2012).Voiced obstruents have also been shown to be associated with negative meanings (e.g.,
concepts of dirtiness, sneakiness and negative images in listeners on hearing the hypothetical
words containing voiced phonemes) (Hamano, 1998; Hinton et al., 1994; Kawahara et al., 2008).
Since the sound symbolic association of voiced obstruents is with harshness and negative
images/perception, the present research extends these findings to a sweet-bitter continuum (here
note that the voiced phonemes are of low frequency and are perceived as harsh; Kingston et al.,
2008; Slepian & Galinsky, 2016).
A large body of research now shows a strong association between high (vs. low) pitches4
with sweet and sour (vs. bitter) tastes (Crisinel & Spence, 2009; also see Knöferle & Spence,
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
19
2012 for a review). A similar association of tastes has been shown with certain musical
instruments; for example, pleasant/sweet tastes with piano (or high pitches) and unpleasant/bitter
tastes with the brass instruments (or low pitches) (Crisinel & Spence, 2010). It has also been
shown that even the perception of taste can be altered by changing the music
being played in the background. For example, the same toffee can taste slightly more bitter (vs.
sweet) with low (vs. high) pitched sounds being played (Crisinel et al., 2012). It is worthwhile to
note here that low (vs. high) pitch associations between sounds and tastes may be explained by
the linkages of oro-facial responses with unpleasant (vs. pleasant) tastes (Spence, 2012).
According to this explanation, babies of different mammals (e.g., humans, monkeys, rats)
protrude their tongues outwards (vs. inwards) in response to pleasant (e.g. sweet) vs. unpleasant
tastes (e.g. bitter) (Steiner, Glaser, Hawilo, & Berridge, 2001). Since humans imitate
characteristic facial expressions for disgust, distaste and unpalatable taste (e.g., bitter) (Forestell
& Mennella, 2017), in this account (i.e., due the learned associations between specific gustatory
responses and associated tongue position and vice versa), one would expect utterances with the
same or matching tongue positions to have similar meanings (Knöferle & Spence, 2012). Since
gustatory centers (e.g., insular cortex) and regions processing phonology (e.g., Broca’s area) in
the brain are said to lie in close proximity to each other (Colizoli, Murre & Rouw, 2013; Ward,
Simner, & Auyeung, 2005), this might explain these linguistic-sensory pairings.
In contrast to the large body of literature on the cross-modal association between tastes
and sounds, there is scant evidence specifically connecting consonants with taste associations.
Most research demonstrates the connection of tastes with either sounds or musical notes (e.g.,
Crisinel & Spence, 2010, 2011; Ngo & Spence, 2011), but not with consonants per se. This is
surprising given the fact that sound symbolic associations were initially popularized by scientists
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
20
linking psycholinguistic properties of words with sensory attributes (see Lockwood &
Dingemanse, 2015 for a review; Ramachandran & Hubbard, 2001) and that consonants
substantially outnumber vowels across most languages (Maddieson & Disner, 1984; Nespor et
al., 2003).
We believe that the perception of voiced obstruents/consonants, in the context of brand-
naming and product attributes, is an under-researched area (Guevremont & Grohmann, 2015)
and that the current research contributes to our understanding of how consumers associate BNs
with expectation of taste attributes. An obvious extension of this research is to investigate
whether our results generalize to a sweet-sour continuum. We hypothesize that voiced obstruents
would be more preferred in brand names for sour products, compared to voiceless for the sweet
ones.
We argue that sound symbolism should be considered carefully by marketing managers
during the process of product and brand naming, as we now know that consumers intuitively
expect a congruency of sensory expectations. For example, the brand name for a hiking boot
should better be rugged; Klink and Athaide (2012) and that of a skin conditioner should be soft
and gentle. A similar congruency is desired between brand names of food products and their
taste expectations (Crisinel, Jones, & Spence, 2012) as incongruent combinations may affect the
subsequent product ratings. For example, in the context of medications for cancer /p/, /t/, /k/, /f/
and /s/ (mainly voiceless obstruents), are over represented whereas /b/, /d/, /g/, /v/ and /z/
(mainly voiced obstruents) are underrepresented (Abel & Glinert, 2008). Most researchers (e.g.,
Spence, 2014) attribute this to the fastness vs. slowness of these phonemes. Since we now
suggest that voiced obstruents can also signify bitterness, is it likely that these inherent ‘bitter’
associations cue manufacturers sound symbolically to avoid these phonemes. This is perhaps
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
21
especially pertinent when naming drugs designed to combat a disease that in any case induces
public fear (e.g., cancer).
We acknowledge a few potential limitations of this paper, which generates some
suggestions for future research. Firstly, we used only tri-syllabic names (e.g., Sa-to-ke); and
research suggests that even the syllabic length of a word can affects its perceived attributes (e.g.,
luxury; Pathak, Velasco, Petit, & Calvert, 2019). It is likely that the length of a name itself holds
meaning (in terms of taste) for consumers, and this has not been tested in the current paper.
Secondly, the word structure used to create the HBNs in the current paper was a simple CV-CV
(Consonant-Vowel) structure. It is unlikely that the results will be different for HBNs with
complex linguistic structures (e.g., the word ‘trap’ has a more complex CCVC structure), but the
current paper has not tested this. Similarly, HBNs with even more complex consonant clusters
(e.g., /Zk/) can be created and tested. Thirdly, we have used an equal number of consonants in
HBNs and have not tested the increase (vs. decrease) in the perceived sweetness (vs. bitterness)
of a brand name with a varying number of voiced (vs. voiceless) obstruents. It might be the case
that the perceived sweetness (vs. bitterness) associated with an HBN vary with the number of
obstruents used (e.g., one vs. two vs. three voiced obstruents in an HBN). Future studies could
use HBNs with one, two vs. three voiced obstruents to examine the effect of the number of
voiced obstruents present in an HBN on the perceived sweetness (or bitterness). Lastly, in this
paper we have only chosen to study the sound symbolic attributes associated with brand names
of chocolates. Further research is required to determine whether our results extend to other
product categories (e.g., sweet vs. bitter beer/jams).
Sensory perceptions play an important role in a consumer’s interaction with brands and
research has consistently shown that colors, sounds, textures, tastes and smells can often affect a
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
22
product’s perception. However, when it comes to expressing the multitude of sensory feelings
through speech, one encounters linguistic limitations (Caballero & Paradis, 2015). For example,
how can one communicate different types of smells in speech, or how can one talk about flavors
which go beyond traditional sweet, bitter, salt and sour axes? The findings of this paper will help
brand managers to communicate some of these complicated sensory experiences through novel
names and psycholinguistic elements.
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
23
References
Aaker, J. L. (1997). Dimensions of brand personality. Journal of Marketing Research, 34(3),
347-356.
Abel, G. A., & Glinert, L. H. (2008). Chemotherapy as language: Sound symbolism in cancer
medication names. Social Science & Medicine, 66(8), 1863-1869.
Alashban, A. A., Hayes, L. A., Zinkhan, G. M., & Balazs, A. L. (2002). International brand-
name standardization/adaptation: Antecedents and consequences. Journal of
International Marketing, 10(3), 22-48.
Aryani, A., Conrad, M., Schmidtke, D., & Jacobs, A. (2018). Why 'piss' is ruder than 'pee'? The
role of sound in affective meaning making. PloS One, 13(6), e0198430.
Bronner, K., Frieler, H., Bruhn, H., Hirt, R., & Piper. D. (2012). What is the sound of citrus?
Research on the correspondences between the perception of sound and flavour. In
Proceedings of the 12th International Conference of Music Perception and Cognition
(ICMPC) and the 8th Triennial Conference of the European Society for the Cognitive
Sciences of Music (ESCOM) (pp. 142-148). Thessaloniki, Greece.
Caballero, R., & Paradis, C. (2015). Making sense of sensory perceptions across languages and
cultures. Functions of Language, 22(1), 1-19.
Carroll, D. W. (2004). Psychology of language. Belmont, California, USA: Thomson Wadsworth
Charmasson, H. (1988). The name is the game: How to name a company or product. Homewood,
Illinois, USA: Dow Jones-Irwin .
Cho, S., Han, A., Taylor, M. H., Huck, A. C., Mishler, A. M., Mattal, K. L., . . . Seo, H.-S.
(2015). Blue lighting decreases the amount of food consumed in men, but not in women.
Appetite, 85, 111-117.
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
24
Chollet, G., Di Benedetto, M. G., Esposito, A., & Marinaro, M. (2012). Speech Processing,
Recognition and Artificial Neural Network. Paper presnted at the Proceedings of the 3rd
International School on Neural Nets “Eduardo R. Caianiello”. Salerno, Italy.
Colizoli, O., Murre, J. M., & Rouw, R. (2013). A taste for words and sounds: A case of lexical-
gustatory and sound-gustatory synesthesia. Frontiers in Psychology, 4, 775.
Crisinel, A.-S., Cosser, S., King, S., Jones, R., Petrie, J., & Spence, C. (2012). A bittersweet
symphony: Systematically modulating the taste of food by changing the sonic properties
of the soundtrack playing in the background. Food Quality and Preference, 24(1), 201-
204.
Crisinel, A.-S., Jones, S., & Spence, C. (2012). ‘The sweet taste of maluma’: Crossmodal
associations between tastes and words. Chemosensory Perception, 5(3-4), 266-273.
Crisinel, A.-S., & Spence, C. (2009). Implicit association between basic tastes and pitch.
Neuroscience Letters, 464(1), 39-42.
Crisinel, A.-S., & Spence, C. (2010). As bitter as a trombone: Synesthetic correspondences in
non-synesthetes between tastes and flavors and musical instruments and notes. Attention,
Perception & Psychophysics,72, 1994-2002
Dawar, N., & Parker, P. (1994). Marketing universals: Consumers’ use of brand name, price,
physical appearance and retailer reputation as signals of product quality. Journal of
Marketing, 58(2), 81-95.
Dahl, D. W. (2010). Understanding the role of spokesperson voice in broadcast advertising. In A.
Krishna (Ed.), Sensory marketing: Research on the sensuality of products (pp. 169-182).
New York, USA: Routledge.
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
25
D’Onofrio, A. (2014). Phonetic detail and dimensionality in sound-shape correspondences:
Refining the bouba-kiki paradigm. Language and Speech, 57(3), 367-393.
Eisend, M., & Stokburger-Sauer, N. E. (2013). Brand personality: A meta-analytic review of
antecedents and consequences. Marketing Letters, 24(3), 205-216.
Emerson, R. H. (1997). English spelling and its relation to sound. American Speech, 72(3), 260-
288.
Faul, F., Erdfelder, E., Lang, A. G., & Buchner, A. (2007). G* Power 3: A flexible statistical
power analysis program for the social, behavioral, and biomedical sciences. Behavior
Research Methods, 39(2), 175-191.
Fenko, A., Lotterman, H., & Galetzka, M. (2016). What’s in a name? The effects of sound
symbolism and package shape on consumer responses to food products. Food Quality
and Preference, 51, 100-108.
Fónagy, I. (1963). Die Metaphern in der Phonetik [The metaphors in phonetics]. The
Hague:Mouton.
Forestell, C. A., & Mennella, J. A. (2017). The relationship between infant facial expressions and
food acceptance. Current Nutrition Reports, 6(2), 141-147.
Friese, M., Wänke, M., & Plessner, H. (2006). Implicit consumer preferences and their influence
on product choice. Psychology & Marketing, 23(9), 727-740.
Gallace, A., Boschin, E., & Spence, C. (2011). On the taste of “Bouba” and “Kiki”: An
exploration of word–food associations in neurologically normal participants. Cognitive
Neuroscience, 2(1), 34-46.
Guevremont, A., & Grohmann, B. (2015). Consonants in brand names influence brand gender
perceptions. European Journal of Marketing, 49(1/2), 101-122.
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
26
Hagtvedt, H., & Brasel, S. A. (2016). Cross-modal communication: Sound frequency influences
consumer responses to color lightness. Journal of Marketing Research, 53(4), 551-562.
Hamano, S. (1998). The sound-symbolic system of Japanese. Stanford, California, USA: CSLI
Publications
Hinton, L., Nichols, J., & Ohala, J. (Eds.) (1994). Sound symbolism. Cambridge, UK: Cambridge
University Press.
Janiszewski, C., & Meyvis, T. (2001). Effects of brand logo complexity, repetition, and spacing
on processing fluency and judgment. Journal of Consumer Research, 28(1), 18-32.
Kawahara, S., & Moore, J. (2018). Exploring sound symbolic knowledge of English speakers
using Pokémon character names. Ms. Keio University and Sophia University.
Kawahara, S., Shinohara, K., & Uchimoto, Y. (2008). A positional effect in sound symbolism:
An experimental study. Proceedings of the Japan Cognitive Linguistics Association, (vol.
8, pp. 417-427). Tokyo, Japan.
Kingston, J., & Diehl, R. L. (1995). Intermediate properties in the perception of distinctive
feature values. Papers in Laboratory Phonology, 4, 7-27.
Kingston, J., Diehl, R. L., Kirk, C. J., & Castleman, W. A. (2008). On the internal perceptual
structure of distinctive features: The [voice] contrast. Journal of Phonetics, 36(1), 28-54.
Klink, R. R. (2000). Creating brand names with meaning: The use of sound symbolism.
Marketing Letters, 11(1), 5-20.
Klink, R. R., & Athaide, G. A. (2012). Creating brand personality with brand names. Marketing
Letters, 23(1), 109-117.
Kovic, V., Plunkett, K., & Westermann, G. (2010). The shape of words in the brain. Cognition,
114(1), 19-28.
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
27
Ramachandran, V. S., & Hubbard, E. M. (2001). Synaesthesia -A window into perception,
thought and language. Journal of Consciousness Studies, 8(12), 3-34.
Knoeferle, K. M., Woods, A., Käppler, F., & Spence, C. (2015). That sounds sweet: Using cross‐
modal correspondences to communicate gustatory attributes. Psychology & Marketing,
32(1), 107-120.
Knöferle, K. M., & Spence, C. (2012). Crossmodal correspondences between sounds and tastes.
Psychonomic Bulletin & Review, 19, 992-1006.
Krishna, A. (2012). An integrative review of sensory marketing: Engaging the senses to affect
perception, judgment and behavior. Journal of Consumer Psychology, 22(3), 332-351.
Kubozono, H. (1999). Japanese phonetics. Tokyo, Japan: Iwanami Shoten.
Lockwood, G., & Dingemanse, M. (2015). Iconicity in the lab: A review of behavioral,
developmental, and neuroimaging research into sound-symbolism. Frontiers in
Psychology, 6, 1246.
Lockwood, G., Dingemanse, M., & Hagoort, P. (2016). Sound-symbolism boosts novel word
learning. Journal of Experimental Psychology: Learning, Memory, and Cognition, 42(8),
1274-1281.
Lockwood, G., Hagoort, P., & Dingemanse, M. (2016). How iconicity helps people learn new
words: Neural correlates and individual differences in sound-symbolic bootstrapping.
Collabra, 2(1), 1-15.
McCormick, K., Kim, J. Y., List, S. & Nygaard, L. C. (2015). Sound to meaning mappings in the
bouba-kiki effect. Proceedings of the 37th Annual Meeting of Cognitive Science Society,
(pp. 1565-1570), Pasadena, CA, USA.
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
28
Monaghan, P., Mattock, K., & Walker, P. (2012). The role of sound symbolism in language
learning. Journal of Experimental Psychology: Learning, Memory, and Cognition, 38(5),
1152-1164.
Motoki, K., Saito, T., Park, J., Velasco, C., Spence, C., & Sugiura, M. (2020). Tasting names:
Systematic investigations of taste-speech sounds associations. Food Quality and
Preference, 80, 103801.
Saji, N., Akita, K., Imai, M., Kantartzis, K., & Kita, S. (2013). Cross-linguistically shared and
language-specific sound symbolism for motion: An exploratory data mining approach.
Proceedings of the 35th Annual Conf. of the Cognitive Science Society, (vol. 31, pp.
1253-1259), Berlin, Germany.
Major, R. C., & Faudree, M. C. (1996). Markedness universals and the acquisition of voicing
contrasts by Korean speakers of English. Studies in Second Language Acquisition, 18(1),
69-90.
Morton, E. S. (1977). On the occurrence and significance of motivation-structural rules in some
bird and mammal sounds. The American Naturalist, 111(981), 855-869.
Ngo, M. K., Misra, R., & Spence, C. (2011). Assessing the shapes and speech sounds that people
associate with chocolate samples varying in cocoa content. Food Quality and Preference,
22(6), 567-572.
Ngo, M. K., & Spence, C. (2011). Assessing the shapes and speech sounds that consumers
associate with different kinds of chocolate. Journal of Sensory Studies, 26(6), 421-428.
Nielsen, A., & Rendall, D. (2011). The sound of round: Evaluating the sound-symbolic role of
consonants in the classic Takete-Maluma phenomenon. Canadian Journal of
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
29
Experimental Psychology/Revue canadienne de psychologie expérimentale, 65(2), 115-
124.
Nielsen, A. K., & Rendall, D. (2013). Parsing the role of consonants versus vowels in the classic
Takete-Maluma phenomenon. Canadian Journal of Experimental Psychology/Revue
canadienne de psychologie expérimentale, 67(2), 153-163.
Nuckolls, J. B. (1999). The case for sound symbolism. Annual Review of Anthropology, 28(1),
225-252.
Obrist, M., Comber, R., Subramanian, S., Piqueras-Fiszman, B., Velasco, C., & Spence, C.
(2014). Temporal, affective, and embodied characteristics of taste experiences. In
Proceedings of the 32nd annual ACM conference on Human factors in Computing
Systems (pp. 2853–2862). New York, USA: ACM Press.
Ohala, J. J. (1983). The origin of sound patterns in vocal tract constraints. In P. F. MacNeilage
(Ed.), The production of speech (pp. 189-216). New York, NY, USA: Springer.
Ohala, J. J. (1994). The frequency code underlies the sound-symbolic use of voice pitch. In
Hinton, L., Nichols, J., & Ohala, J.J. (Eds.). Sound symbolism (pp. 325-347). Cambridge,
UK: Cambridge University Press.
Ohala, J. J. (2011). Accommodation to the aerodynamic voicing constraint and its phonological
relevance. Paper presented at the Proceedings of the 17th International Congress of
Phonetic Sciences (pp. 64-67), Hong Kong, China.
Pathak, A., Calvert, G. A., & Lim, E. A. (2017). How the linguistic characteristics of a brand
name can affect its luxury appeal. International Journal of Market Research, 59(5), 567-
600.
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
30
Pathak, A., Calvert, G., & Velasco, C. (2017). Evaluating the impact of early-and late-acquired
phonemes on the luxury appeal of brand names. Journal of Brand Management, 24(6),
522-545.
Pathak, A., Velasco, C., Petit, O., & Calvert, G. A. (2019). Going to great lengths in the pursuit
of luxury: How longer brand names can enhance the luxury perception of a brand.
Psychology & Marketing, 36(10), 951-963.
Pathak, A., Calvert, G. A. & Lim, L. (Submitted). Harsh voices, sound branding: How voiced
consonants in a brand’s name can alter its perceived attributes. Psychology & Marketing
Paolacci, G., & Chandler, J. (2014). Inside the Turk: Understanding Mechanical Turk as a
participant pool. Current Directions in Psychological Science, 23(3), 184-188.
Piqueras-Fiszman, B., Alcaide, J., Roura, E., & Spence, C. (2012). Is it the plate or is it the food?
Assessing the influence of the color (black or white) and shape of the plate on the
perception of the food placed on it. Food Quality and Preference, 24(1), 205-208.
Pogacar, R., Kouril, M., Carpenter, T. P., & Kellaris, J. J. (2018). Implicit and explicit
preferences for brand name sounds. Marketing Letters, 29(2), 241-259.
Ramachandran, V. S., & Hubbard, E. M. (2001). Synaesthesia-A window into perception,
thought and language. Journal of Consciousness Studies, 8(12), 3-34.
Reinoso Carvalho, F., Van Ee, R., Rychtarikova, M., Touhafi, A., Steenhaut, K., Persoone, D., . .
. Leman, M. (2015). Does music influence the multisensory tasting experience? Journal
of Sensory Studies, 30(5), 404-412.
Saji, N., Akita, K., Imai, M., Kantartzis, K., & Kita, S. (2013). Cross-linguistically shared and
language-specific sound symbolism for motion: An exploratory data mining approach.
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
31
Paper presented at the Proceedings of the Annual Meeting of the Cognitive Science
Society (pp. 1253-1259), Berlin, Germany.
Sapir, E. (1929). A study in phonetic symbolism. Journal of Experimental Psychology, 12(3),
225-239.
Shih, S. S., Ackerman, J., Hermalin, N., Inkelas, S., & Kavitskaya, D. (2018). Pokémonikers: A
study of sound symbolism and Pokémon names. Proceedings of the Linguistic Society of
America (vol 3(42), pp. 1-6). Salt Lake City, Utah, USA.
Shinohara, K., & Kawahara, S. (2010). A cross-linguistic study of sound symbolism: The images
of size. Paper presented at the Proceedings of the 36th Annual Meeting of the Berkeley
Linguistics Society (pp. 396-410). Berkeley, California, USA.
Sidhu, D. M., & Pexman, P. M. (2015). What’s in a name? Sound symbolism and gender in first
names. PloS One, 10(5), e0126809.
Simner, J., Cuskley, C., & Kirby, S. (2010). What sound does that taste? Cross-modal mappings
across gustation and audition. Perception, 39(4), 553-569.
Slepian, M. L., & Galinsky, A. D. (2016). The voiced pronunciation of initial phonemes predicts
the gender of names. Journal of Personality and Social Psychology, 110(4), 509-527.
Spence, C. (2012). Managing sensory expectations concerning products and brands: Capitalizing
on the potential of sound and shape symbolism. Journal of Consumer Psychology, 22(1),
37-54.
Spence, C. (2014). Assessing the influence of shape and sound symbolism on the consumer’s
response to chocolate. New Food, 17(2), 59-62.
Spence, C., & Gallace, A. (2011). Tasting shapes and words. Food Quality and Preference,
22(3), 290-295.
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
32
Steiner, J. E., Glaser, D., Hawilo, M. E., & Berridge, K. C. (2001). Comparative expression of
hedonic impact: Affective reactions to taste by human infants and other primates.
Neuroscience & Biobehavioral Reviews, 25(1), 53-74.
Topolinski, S., & Boecker, L. (2016). Mouth-watering words: Articulatory inductions of eating-
like mouth movements increase perceived food palatability. Appetite, 99, 112-120.
Topolinski, S., Boecker, L., Erle, T. M., Bakhtiari, G., & Pecher, D. (2017). Matching between
oral inward–outward movements of object names and oral movements associated with
denoted objects. Cognition and Emotion, 31(1), 3-18.
Topolinski, S., Likowski, K. U., Weyers, P., & Strack, F. (2009). The face of fluency: Semantic
coherence automatically elicits a specific pattern of facial muscle reactions. Cognition
and Emotion, 23(2), 260-271.
Topolinski, S., Maschmann, I. T., Pecher, D., & Winkielman, P. (2014). Oral approach–
avoidance: Affective consequences of muscular articulation dynamics. Journal of
Personality and Social Psychology, 106(6), 885-896.
Van Lancker, D., & Cummings, J. L. (1999). Expletives: Neurolinguistic and neurobehavioral
perspectives on swearing. Brain Research Reviews, 31(1), 83-104.
Ward, J., Simner, J., & Auyeung, V. (2005). A comparison of lexical-gustatory and grapheme-
colour synaesthesia. Cognitive Neuropsychology, 22(1), 28-41.
Woods, A. T., Velasco, C., Levitan, C. A., Wan, X., & Spence, C. (2015). Conducting perception
research over the internet: A tutorial review. PeerJ, 3, e1058.
Yorkston, E., & Menon, G. (2004). A sound idea: Phonetic effects of brand names on consumer
judgments. Journal of Consumer Research, 31(1), 43-51.
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
33
Yoon, C., Gutchess, A. H., Feinberg, F., & Polk, T. A. (2006). A functional magnetic resonance
imaging study of neural dissociations between brand and person judgments. Journal of
Consumer Research, 33(1), 31-40.
Zaltman, G., & Wallendorf, M. (1979). Consumer behavior: Basic findings and management
implications. New York, USA: John Wiley and Sons.
Żygis, M., Fuchs, S., & Koenig, L. L. (2012). Phonetic explanations for the infrequency of
voiced sibilant affricates across languages. Laboratory Phonology, 3(2), 299-336.
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
34
Footnotes
1 Words with /g/ also can be pronounced in two different ways, with a hard g (as in Gate)
sound vs. a soft /g/ (as in Gem). This was an oversight in the design of our Study 1, which was
kindly pointed out by an anonymous reviewer; we now report an additional analysis excluding
the /g/.
2 Recent studies also point out that our brain processes the personality of a brand vs. that of
a person differently, and not in a way that marketing scholars have proposed for long (Yoon,
Gutchess, Feinberg, & Polk, 2006).
3 Another probable explanation can be the temporality of tastes, as described by Obrist et
al. (2014). These authors argue that some tastes (e.g. sour) are short-lived and explosive in nature
whereas others (e.g. sweet) are lingering, pleasant and slow. Participants describe these tastes in
a similar fashion (e.g. instant vs. slow). Similarly, voiced (vs. voiceless) obstruents can be
perceived as harsh and more intense (vs. smooth and less intense) and may explain their linkages
with bitterness (vs. sweetness).
4 Low (vs. high) pitch and low (vs. high) frequency sounds hold the same meaning. Pitch
denotes the perceptual dimension of the frequency of sounds (Hagtvedt & Brasel, 2016).
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
35
Appendix 1
HBNs used in the current paper.
HBN with voiceless obstruents HBN with voiced obstruents
Satoke Zadoge
Sokita Zogida
Sipetu Zibedu
Sukapi Zugabi
Sekuto Zegudo
Kaseti Gazedi
Ketiso Gedizo
Kipotu Gibodu
Kopusa Gobuza
Kutape Gudabe
Tupasi Dubazi
Tosepa Dozeba
Tekopu Degobu
Tisuko Dizugo
Tapise Dabize
Pokasi Bogazi
Pesoka Bezoga
Pituse Biduze
Pukeso Bugezo
Patisu Badizu
Fakise Vagize
Fetoki Vedogi
Fipatu Vibadu
Fokuta Voguda Source: Pathak, Calvert & Lim (Submitted)
Author'
s fina
l vers
ion
VOICING AND TASTE ATTRIBUTES
36
Appendix 2
A sample of brand names created by participants in Studies 2a and 2b.
Study 2a Study 2b
Sweet chocolates Bitter chocolates Sweet chocolates Bitter chocolates
BN1 BN2 BN3 BN1 BN2 BN3 BN1 BN2 BN3 BN1 BN2 BN3
nice milk sweet dark tart cocoa zetes govst kakest bedits stafe pitsid
kis good favs bit sorz betz kizzley bapple grivvely pizzley taffe buddle
pasuite pati tavi zasoote pelo tavo tuoi peiou givs dost zoks gisk
debita apedis vobisa ekgazi biveto patozi svent minity bahbahtee bivat dagoosh aklin
paive pugie vea gotuk vizat odi perfektionz blizzed skrumptionz grateens vorzatsan panzkats
set svet seet zip bepz beed kavt fost sekt vakt kovt pegt
swwecho swechoc swchoco bttlate bittatt bittlat pagio gestavi tebapou fibaet zubogsi vistopug
delightfully
sweet
ohh so
sweet!
my
sweet
choloate
cringe
delight
bitter
blast
better
bitter
shugarkic popsikulee kandees bavet bisot bizuk
svet zeap kavo gobk pogi dapu dibs steet fibet peets zouts viteek
kast dak tee dae kiou busie zupod sdap koged tadgo vopas bokeg BN= Brand name
Author'
s fina
l vers
ion
* p<.05; **p=.098; *** p=.056; Error bars represent the SE of means
Figure 1. Study 3: Brand Personality Scale (BPS) factors, facets and traits of the HBNs.
4
5
6
7
Ruggedness* Sophistication* Sincerity
Rat
ings
BPS factors
2
3
4
5
6
7
8
9
Rat
ings
BPS traits
VoicedVoiceless
2
3
4
5
6
7
8
Rat
ings
BPS facets
Author'
s fina
l vers
ion
(BPS factors: HBNs containing voiced obstruents are perceived as more rugged whereas HBNs containing voiceless obstruents are perceived as more
sophisticated.
BPS facets: HBNs containing voiced obstruents are perceived as more outdoorsy and tough, whereas HBNs containing voiceless obstruents are perceived as
more charming, cheerful and wholesome.
BPS traits: HBNs containing voiced obstruents are perceived as more masculine, rugged and tough, whereas HBNs containing voiceless obstruents are perceived
as more charming, feminine, glamourous, honest, wholesome, cheerful and sentimental)
Author'
s fina
l vers
ion
Tables
Table 1. Results of Study 3.
BPS traits Voiced
obstruents
SD Voiceless
obstruents
SD BPS facets Voiced
obstruents
SD Voiceless
obstruents
SD BPS factors Voiced
obstruents
SD Voiceless
obstruents
SD
Masculine 6.34 3.07 4.89 2.63 Outdoorsy 5.34 2.17 4.66 1.90 Ruggedness 5.64 2.28 4.69 1.90
Outdoorsy 5.62 2.90 4.97 2.73
Western 4.07 2.59 4.11 2.56
Rugged 6.02 2.92 4.57 2.45 Tough 6.09 2.76 4.73 2.28
Tough 6.16 2.87 4.89 2.54
Upper-class 5.85 2.66 6.11 2.57 Upper class 5.65 2.31 6.18 2.22 Sophistication 5.48 2.02 6.22 2.05
Glamorous 4.89 2.71 5.87 2.66
Good looking 6.21 2.38 6.56 2.50
Charming 5.46 2.48 6.54 2.40 Charming 5.30 2.04 6.25 2.16
Feminine 4.51 2.80 5.84 2.91
Smooth 5.93 2.44 6.38 2.57
Down-to-earth 6.20 2.34 5.92 2.34 Down-to-
earth
5.63 1.89 5.63 2.02 Sincerity 6.13 1.56 6.39 1.71
Family oriented 5.64 2.32 6.11 2.37
Small town 5.05 2.70 4.87 2.56
Honest 6.46 2.05 6.92 2.14 Honest 6.52 1.83 6.59 1.91
Sincere 6.54 2.20 6.59 1.95
Real 6.57 2.46 6.26 2.60
Wholesome 5.77 2.16 6.48 2.33 Wholesome 6.89 1.81 7.22 1.92
Original 8.00 2.20 7.97 1.97
Cheerful 5.75 2.29 6.56 2.39 Cheerful 5.74 2.05 6.39 2.01
Sentimental 5.03 2.18 5.77 2.32
Friendly 6.44 2.37 6.85 2.28
Figures in bold show significantly different results Author'
s fina
l vers
ion
Ruggedness (BPS factor) = Average means of the BPS facets (outdoorsy and tough)
Sophistication (BPS factor) = Average means of the BPS facets (upper-class and charming)
Sincerity (BPS factor) = Average means of the BPS facets (down-to-earth, honest, wholesome and cheerful)
Outdoorsy (BPS facet) = Average means of the BPS traits (masculine, outdoorsy and western)
Tough (BPS facet) = Average means of the BPS traits (rugged and tough)
Upper-class (BPS facet) = Average means of the BPS traits (upper-class, glamorous and good looking)
Charming (BPS facet) = Average means of the BPS traits (charming, feminine and smooth)
Down-to-earth (BPS facet) = Average means of the BPS traits (down-to-earth, family oriented and small town)
Honest (BPS facet) = Average means of the BPS traits (honest, sincere and real)
Wholesome (BPS facet) = Average means of the BPS traits (wholesome and original)
Cheerful (BPS facet) = Average means of the BPS traits (cheerful, sentential and friendly)
Author'
s fina
l vers
ion
Top Related