Sweet and sour: Music and taste associations

Abstract

Purpose
– This paper aims to extend current understanding concerning the cross-modal correspondences between sounds and tastes by introducing new research tools and experimental data to study associations and their reflections between music and taste.

Design/methodology/approach
– The experiment design addresses the multidisciplinary approach by using cultural, chemical and statistical analysis methods.

Findings
– The paper provides further evidence that exposure to the “sweet” or “sour” musical pieces influences people’s food-related thinking processes and behaviors. It also demonstrates that sweet or sour elements in the music may reflect to actual sweetness (as measured by sugar content) and sourness (as measured by organic acid content) of foods developed in association with music carrying similar taste characteristics.

Research limitations/implications
– The findings should be replicated and expanded using larger consumer samples and wider repertoires of “taste music” and dependent variables. Also, the level of experimental control should be improved; e.g., the “sweet” and “sour” music were produced using different instruments, which may have an influence to the results.

Practical implications
– Ambient “taste music” that is congruent with the basic flavors of the dishes can be played in restaurants to highlight guests’ sensory experience.

Social implications
– By carefully considering the symbolic meanings of the music used in different social situations, it is possible to create multimodal experiences and even subconscious expectations in people’ minds.

Originality/value
– Cross-modal associations are made between the tastes and music. This can influence on perception of food and provide new ways to build multimodal gastronomic experiences.

Full text

Nutrition & Food Science
Sweet and sour: music and taste associations
Maija Kontukoski Harri Luomala Bruno Mesz Mariano Sigman Marcos Trevisan Minna Rotola-Pukkila
Anu Inkeri Hopia
Article information:
To cite this document:
Maija Kontukoski Harri Luomala Bruno Mesz Mariano Sigman Marcos Trevisan Minna Rotola-Pukkila
Anu Inkeri Hopia , (2015),"Sweet and sour: music and taste associations", Nutrition & Food Science,
Vol. 45 Iss 3 pp. 357 - 376
Permanent link to this document:
http://dx.doi.org/10.1108/NFS-01-2015-0005
Downloaded on: 11 May 2015, At: 00:00 (PT)
References: this document contains references to 34 other documents.
To copy this document: permissions@emeraldinsight.com
The fulltext of this document has been downloaded 37 times since 2015*
Access to this document was granted through an Emerald subscription provided by 226783 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald
for Authors service information about how to choose which publication to write for and submission
guidelines are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as
well as providing an extensive range of online products and additional customer resources and
services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the
Committee on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for
digital archive preservation.
*Related content and download information correct at time of
download.
Downloaded by UNIVERSITY OF TAMPERE At 00:00 11 May 2015 (PT)

Sweet and sour: music and
taste associations
Maija Kontukoski
Functional Foods Forum, University of Turku, Turku, Finland
Harri Luomala
Department of Marketing, University of Vaasa, Vaasa, Finland
Bruno Mesz
Ciencia y Tecnologia, Universidad Nacional de Quilmes,
Buenos Aires, Argentina
Mariano Sigman
Laboratorio de Neurociencia Integrativa, Depto. de Física, FCEN,
Universidad de Buenos Aires, Buenos Aires, Argentina
Marcos Trevisan
Department of Physics, Universidad de Buenos Aires,
Buenos Aires, Argentina, and
Minna Rotola-Pukkila and Anu Inkeri Hopia
Functional Foods Forum, University of Turku, Turku, Finland
Abstract
Purpose – This paper aims to extend current understanding concerning the cross-modal
correspondences between sounds and tastes by introducing new research tools and experimental data
to study associations and their reections between music and taste.
Design/methodology/approach – The experiment design addresses the multidisciplinary
approach by using cultural, chemical and statistical analysis methods.
Findings – The paper provides further evidence that exposure to the “sweet” or “sour” musical pieces
inuences people’s food-related thinking processes and behaviors. It also demonstrates that sweet or sour
elements in the music may reect to actual sweetness (as measured by sugar content) and sourness (as
measured by organic acid content) of foods developed in association with music carrying similar taste
characteristics.
Research limitations/implications – The ndings should be replicated and expanded using larger
consumer samples and wider repertoires of “taste music” and dependent variables. Also, the level of
experimental control should be improved; e.g., the “sweet” and “sour” music were produced using
different instruments, which may have an inuence to the results.
Practical implications – Ambient “taste music” that is congruent with the basic avors of the dishes
can be played in restaurants to highlight guests’ sensory experience.
Social implications – By carefully considering the symbolic meanings of the music used in different
social situations, it is possible to create multimodal experiences and even subconscious expectations in
people’ minds.
Originality/value – Cross-modal associations are made between the tastes and music. This can
inuence on perception of food and provide new ways to build multimodal gastronomic experiences.
Keywords Music, Taste, Cross-modal associations, Cross-modality, Sourness, Sweetness
Paper type Research paper
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/0034-6659.htm
Music
and taste
associations
357
Received 19 January 2015
Revised 22 January 2015
Accepted 22 January 2015
Nutrition & Food Science
Vol. 45 No. 3, 2015
pp. 357-376
© Emerald Group Publishing Limited
0034-6659
DOI 10.1108/NFS-01-2015-0005
Downloaded by UNIVERSITY OF TAMPERE At 00:00 11 May 2015 (PT)

Introduction
Our enjoyment of food is multisensory in nature. All ve senses have their specic role:
the visual appearance and the odor of the dish create expectations about its taste and,
during eating, a combination of taste, smell, touch and hearing stimuli integrate into an
overall multisensory perception of food. Senses do not operate in isolation, but synergize
with one another. Interaction between odor and taste (King et al., 2006), sight and taste
(Koch and Koch, 2003;Spence et al., 2010) and many other interactions among different
senses are thoroughly reviewed, for example, by Ernst and Bülthoff (2004). Since the
1960s, the interaction between auditory and taste perception has attracted academic
attention (Holt-Hansen, 1968). This has driven a diverse eld of related research (Beeli
et al., 2005;Hänggi et al., 2008;Crisinel and Spence, 2009,2010a,2010b;Simner et al.,
2010;Mesz et al., 2011).
There is a relationship between auditory cues and the perception of taste and the
avor of foods (Knöferle and Spence, 2012). When people discuss about food or music,
they often have a tendency to use the same adjectives to describe them. Adjectives such
as sweet, dry, light, soft and crisp are used to describe the qualities of both food and
music. People make cross-modal associations between the tastes and sounds (Crisinel
and Spence, 2009,2010a,2010b;Simner et al., 2010;Mesz et al., 2011;Bronner et al., 2012).
For example, in the study of Crisinel and Spence (2010a), participants associated sweet
and sour taste solutions’ tastes to high-pitched sounds, bitter taste to the low-pitched
sounds and salty tastes to medium-pitched sounds.
Taste words can elicit consistent musical patterns (Mesz et al., 2011). Musicians were
asked to improvise about the theme of basic taste words (e.g. “sweet”, “salty”, “bitter”
and “sour”). The results showed that, for example, the taste word “sweet” elicited
musical patterns that were consonant, slow and soft, while “sour” elicited musical
patterns that were high-pitched and dissonant. “Bitter” improvisations were low-
pitched and legato. However, as Knöferle and Spence (2012) point out, the informative
value of the ndings might be limited because of the use of taste words rather than
actual taste solutions. In turn, Crisinel et al. (2012) were rst to demonstrate that a
“bitter” background soundtrack caused food samples (such as bittersweet toffee) to be
rated as more bitter than during a “sweet” background soundtrack.
Here, we investigated whether the “sweet” and “sour” musical patterns of four
different musical pieces have an effect on participants’ taste associations. This study
builds on Mesz et al.’s (2011) ndings on consistent taste – musical patterns. In addition
to verbal and descriptive associations, this experiment is extended to study how music,
which contains sweet or sour musical patterns, may inuence how we prepare and cook
food. This was done by measuring, quantitatively, sugar and acid contents of drinks,
which were prepared based on musical associations.
Methods and materials
Selection of tastes and food items: a pre-test
A pre-test was conducted to identify food items for the eld study that represents
primarily one of the taste dimensions (i.e. sweet, sour, bitter or salty) for typical
consumers in South Ostrobothnia, Finland. Eight participants, recruited from a local
food club, were asked to taste and mark 28 foods into a map having four boxes, each
representing one of the four tastes: sweet, sour, bitter and salty. The participants tended
to mix, especially the tastes sour and bitter, in test foods. This confusion between bitter
NFS
45,3
358
Downloaded by UNIVERSITY OF TAMPERE At 00:00 11 May 2015 (PT)

versus sour is rather common and exists in other studies as well (Laaksonen et al., 2013).
For example, for the tastes of raspberry, bilberry, orange, grapefruit, red currant and
tomato, half of the participants mapped them primarily as sour and the other half as
bitter. Thus, sweet and sour tastes, being easy to identify and distinguish, were chosen
for the actual experiment.
The pre-test results were used when selecting foods for the actual experiment. The
aim was to select the same amount of sweet and sour ingredients with similar color
(yellow) and structure (clear). Selected ingredients for the experiment were: mango juice
(100 per cent of pre-test participants classied this juice as sweet), orange juice (50 per
cent of pre-test participants, sour; 38 per cent, bitter; and 13 per cent, sweet), grapefruit
juice (50 per cent, sour; 50 per cent, bitter), lemon juice (100 per cent, sour), pineapple
juice (100 per cent, sweet) and liquid honey (100 per cent, sweet).
Selection of “sweet” and “sour” musical pieces: a pre-test
A pre-test was conducted to identify the musical pieces that most strongly
associated with sweet or sour. In total, 11 students, from Sibelius Academy’s
Doctoral Programme for Music and Performance Arts, listened to 16 musical pieces
(each lasting 30 seconds), which were selected based on Mesz et al.’s (2011) research.
These represented either “sweet” (musical elements of long duration, low
dissonance, low articulation and low loudness) or “sour” (musical elements of high
pitch, long duration and high dissonance) music. After hearing each of the musical
samples, the pre-testers were asked to make associations between music and taste.
By using a seven-point Likert scale, they mapped the sweetness, sourness, saltiness
and bitterness of the musical pieces. As a result, the following four musical pieces
were chosen for the experiment:
(1) Trois Gymnopédies, No.2 Lent et triste by Erik Satie, composed in 1888, piano
music (www.youtube.com/watch?v⫽1loSL7CjE_w).
(2) Davidsbündlertänze, Op. 6, No. 18, Nicht schnell, C major, Eusebius by Robert
Schumann, composed in 1837, piano music (www.youtube.com/watch?v⫽jIHs
NlwD6jQ).
(3) Superscriptio by Brian Ferneyhough, composed in 1981, ute music (www.
youtube.com/watch?v⫽dYnYimo8z2Q).
(4) Musical sample by Mesz et al. (2012), produced in 2012, fragments of
transformed Argentinian tangos (http://youtu.be/lfxdpNMYmlo).
Satie’s and Schmann’s samples represented “sweet” and Ferneyhough’s and Mesz’s
samples for “sour” music. The following musical parameters of these samples were
analyzed:
• roughness, also called sensory dissonance, which quantied the beating
frequencies;
• brightness was the proportion of high-frequency spectral energy of the sound;
• attack time was the time to reach maximum sound intensity of the note; and
• mean frequency in Hz.
359
Music
and taste
associations
Downloaded by UNIVERSITY OF TAMPERE At 00:00 11 May 2015 (PT)

The results of this analysis can be seen in Table I. The proles of “sweet” and “sour”
samples had distinctively different roughness and brightness, being higher in
Ferneyhough’s and Mesz’s samples than in Schumann’s and Satie’s musical pieces.
Mesz et al. (2011) describe taste words “sweet” and “sour” with a list of qualitative
values for musical parameters: “sweet” musical elements have long duration, low
dissonance, low articulation and loudness (soft), and “sour” musical elements have high
pitch, long duration and high dissonance. “Sweet” musical parameters are found from
the Satie’s and Schumann’s samples. “Sour” musical parameters are found from
Ferneyhough’s and Mesz’s samples.
Satie’s and Schumann’s pieces were performed by a pianist, but Ferneyhough’s
music was performed by a piccoloist. Mesz’s piece was algorithmically generated using
short fragments of tangos, which were transformed in pitch to be in the high piccolo and
clarinet register, speeded up to achieve a medium duration equal to that measured in the
Mesz et al. (2012) experiments for the sour taste and contrapuntally disposed to have
high chordal dissonance.
Description of the eld experiment: experimental design, data collection procedures
and preparing the drink sample
Science fair visitors, aged 16 or older, were recruited to participate in the eld
experiment. Four independent groups of consumers, without extensive musical or
culinary training, were recruited for the experiment that took place in connection with a
public science fair in Western Finland, September, 2012. The fair visitors were
approached and asked whether they wanted to participate in a study involving music
and food. Those consenting were led in small groups to a pre-arranged experimental
room. The rooms contained a table with four to ve seats that were secluded from each
other by light obstructions. Study participants sat down at a table, which had a
questionnaire (text-side down), a pencil, plastic mugs, spoons, wipers and six unlabeled
containers containing the ingredients for mixing the drink. No rewards were given to the
study participants.
The experiment started by playing the pre-tested music to the study participants.
Each of the four groups listened to only one of the four, 30-second musical pieces: Group
A listened to Satie; Group B, Schumann; Group C, Ferneyhough and Group D, Mesz. For
every group, the musical piece was replayed throughout the total time (circa 20 minutes)
of the experiment. The music loudness was not played too loud or quietly, but at an
ambient level.
The questionnaire had four sections. First, the gender and age were queried. Second,
the food-related associations triggered by the music were captured. Third, the study
participants were asked to indicate their taste associations congruent with the music on
a seven-point scale, which included six pre-selected food pairs in the following order:
wheat – oat, chocolate – lingonberry, lemon – banana, chicken – beef, meringue –
Table I.
The musical
parameters of
selected “sweet” and
“sour” musical
samples
Music Roughness Brightness Attacktime Mean frequency
Satie n⫽18 6.1 0.2 0.1 842.8
Schumann n⫽12 13.4 0.1 0.1 293.8
Ferneyhough n⫽20 29.2 0.9 0.1 1252.6
Mesz n⫽15 814.8 0.6 0.1 1085.7
NFS
45,3
360
Downloaded by UNIVERSITY OF TAMPERE At 00:00 11 May 2015 (PT)

sourdough crisp rye bread and red currant – pineapple. Fourth, the study participants
were requested to mix a drink congruent with the music by using ingredients available
on the table. They were asked to write down the ingredients they used for it. The details
of these measurements are disclosed below.
In total, 66 participants agreed to take part in the study. These participants were
divided into four groups. Group A, Satie (n⫽18), consisted of nine female and nine male
participants, and the mean age was 33 years. The demographics, in order, for Group B,
Schumann (n⫽12); Group C, Ferneyhough (n⫽20); and Group D, Mesz (n⫽15) were
the following: ve females and seven males with a mean age of 49; 14 females and ve
males, mean age 23; and 11 females and four males, mean age 42, respectively. There
was no difference in gender distribution among the groups; however, age was unevenly
distributed and there was a signicant difference between the groups: Satie and
Schumann (p⫽0.008), Schumann and Ferneyhough (p⬍0.001) and Ferneyhough and
Mesz (p⬍0.001).
Measurement and analysis of dependent variables
Food-related associations. Study respondents were asked to listen to music and think
about what kind of food-related ideas and associations it brings to their minds. An
empty box was provided for submitting their answers, anonymously. The Latent
Semantic Analysis, with word groups referring to either food names or to the
composition, texture, temperature and quality of food (Deerwester et al., 1990;Diuk et al.,
2012) was conducted. The reference document corpus was the TASA corpus, with 300
dimensions. The word-by-word semantic similarity of each of the word group with
“sweet” and “sour” was computed. This gave two series of similarity values (one series
for “sweet” and another for “sour”). Values vary between ⫺1 (semantically dissimilar to
“sweet” or “sour”) and 1 (semantically similar to “sweet” or “sour”). The customary
one-way ANOVA and post hoc Tukey’s t-tests were conducted to detect differences
between the two similarity values in different groups.
Taste associations. Associations were determined by four food word-pairs: banana –
lemon, chocolate – lingonberry, meringue – crisp rye bread pineapple – red currant. On
the basis of the pre-test for tasting, banana, chocolate, meringue and pineapple
represented sweetness, and lemon, lingonberry, crisp rye bread and red currant
represented sourness. Two ller/control food word-pairs were also included: wheat –
oats and chicken meat – beef meat. These items are generally considered to neutral. The
food word-pairs were shown to the study respondents, and they were asked to indicate
on a seven-point scale, which one of the food word-pairs describes precisely the taste of
the music they were hearing. For example, in the food word-pair, banana – lemon,
extreme lemon had value 1 and extreme banana had value 7. From the given values,
one-way ANOVA and post hoc Tukey’s honest signicant difference (HSD) tests were
conducted to detect the effects of “sweet” and “sour” music on consumers’ taste
associations.
Preparing the drink congruent with the music. The study participants were asked to
listen to the music and prepare a drink congruent with it. Each participant listened to
only one musical piece. First, they were asked to list the ingredients (selecting from
mango juice, orange juice, grapefruit juice, lemon juice, pineapple juice and liquid honey)
they used for mixing their drink, and then, they were asked to prepare a sample drink.
Samples of 20 ml were collected for chemical analysis.
361
Music
and taste
associations
Downloaded by UNIVERSITY OF TAMPERE At 00:00 11 May 2015 (PT)

Chemical analysis. The sugar and acid concentrations were chemically analyzed from
the drink samples. A sample of 50 ml was collected from each participant, and samples
were stored at 20°C until analyzed for their sugar and organic acid contents. Sugar and
acid contents of the juices were analyzed by a high-performance liquid chromatographic
(HPLC) method based on Phenomenex applications ID 5504 and 5505, and the analysis
method for acids was based on Phenomenex application ID 14270. The HPLC system,
Waters, consisted of a Waters in-line degasser AF, 515 HPLC pump, 717plus
autosampler, column oven, 2410 refractive index detector (for sugars) and 2487 dual
␭
absorbance detector (for organic acids). The Empower 3 software was used for data
acquisition and analysis.
The eluent used for organic acid analyses was 20 mM phosphate buffer, pH 2.9.
Potassium phosphate (monobasic ⱖ99 per cent) was purchased from Sigma-Aldrich
(Chemie GmbH, Steinheim) and orthophosphoric acid (85 per cent) used for adjustment
of pH from Merck (Darmstadt, Germany). The eluent used for sugar analysis was
deionized water, ltered through Phenex Nylon 0.45
␮
m lter membrane (Phenomenex).
Sucrose was purchased from Merck (Darmstadt, Germany), D(⫺)-fructose and citric
acid were obtained from Sigma-Aldrich (Steinheim, Germany), D(⫹)-glucose was
purchased from J.T. Baker (Denventer, Holland) and L(-)-malic acid from Acros
Organics (Geel, Belgium). Each juice sample was analyzed in triplicate. Quantication
was carried out with an external standard method. The concentrations of sugars and
acids were expressed as g/l.
A one-way ANOVA and post hoc Tukey’s HSD tests were conducted to detect the
effects of “sweet” and “sour” music’s impact on sugar and acid contents of the drinks.
Results
Food-related associations
Food-related word associations, generated in the four experimental groups, revealed a
consistent trend (Figure 1-4). The higher-ranked words, in response to the sweet
melodies, were chocolate, tasty and for the sour melodies fruits, sour. To quantify this
observation, we measured the projection of this distribution of words to the sweet and
the sour dimensions of semantic space using latent semantic analysis (Table II). In the
groups where the “sweet” music (Satie and Schumann) was played, the connotations of
food-related associations were signicantly closer to the taste word “sweet” than to the
taste word “sour”. In contrast, in the groups that had “sour” music (Ferneyhough and
Mesz) played, such signicant differences were not detected. Second, the key
connotations of food-related associations produced in the “sour” music groups were
signicantly closer to the taste word “sour” than those produced in the “sweet” music
groups, except in the case of the Schumann versus Ferneyhough groups. However, the
key connotations of food-related associations produced in the “sweet” music groups
were not signicantly closer to the taste word “sweet” than those produced in the “sour”
music groups. This experiment can be viewed as a mirror reection of Mesz et al. (2011).
In their original work, Mesz et al. showed that taste words elicit consistent musical
formations. Our analysis demonstrated that specic musical formations formed by the
musical dimensions of different tastes evoked a broad set of words, which consistently
projected to semantic taste dimensions.
NFS
45,3
362
Downloaded by UNIVERSITY OF TAMPERE At 00:00 11 May 2015 (PT)

Figure 1.
The frequencies of
food-related
associations in
different
experimental groups:
“sweet” music (Satie),
n⫽18
363
Music
and taste
associations
Downloaded by UNIVERSITY OF TAMPERE At 00:00 11 May 2015 (PT)

Taste associations
The results shown above indicate that “sweet and sour melodies” produce structured
projections of free word associations. In the next experiment, we investigated whether
they produce a consistent set of associations within a controlled and limited set of words,
which correspond to sweet, sour or neutral associations.
Figure 2.
The frequencies of
food-related
associations in
different
experimental group:
“sweet” music
(Schumann), n⫽12
NFS
45,3
364
Downloaded by UNIVERSITY OF TAMPERE At 00:00 11 May 2015 (PT)

As described earlier, six food word-pairs (of which, two were ller/control items) were
chosen to measure the taste associations triggered by the music. The sweet food words
included chocolate, banana, meringue and pineapple, and the sour food words were
lingonberry, lemon, crisp rye bread (the literal translation from Finnish is sour rusk) and
red currant. The ller/control food word-pairs encompassed wheat versus oats and
chicken versus beef. The presumption was that the “sweet” and “sour” music would
have an effect on study participants’ taste associations, as reected by the variation in
the food word selections.
In the cases of food word-pairs, chocolate – lingonberry and lemon – banana, the results
(Table III) suggest that the “sweet” music (Satie and Schumann) elicited sweeter taste
associations (stronger choice preference for chocolate and banana), while the “sour” music
(Ferneyhough and Mesz) yielded greater sour taste associations (stronger choice preferences
for lingonberry and lemon). In effect, the same pattern of results was obtained for the food
word-pair meringue – crisp rye bread (with the exception that Schumann’s sweet music did
not produce different results as Ferneyhough’s and Mesz’s sour music).
Figure 3.
The frequencies of
food-related
associations in
different
experimental group:
“sour” music
(Ferneyhough),
n⫽20
365
Music
and taste
associations
Downloaded by UNIVERSITY OF TAMPERE At 00:00 11 May 2015 (PT)

Only in the case of the red currant – pineapple food word-pair, the results were not in
accordance with our expectations – no signicant differences in the taste associations
between the “sweet” and “sour” music groups emerged. The null results were also
received in the analysis of the ller/control food word-pair, wheat – oats. A surprising
difference surfaced when the responses to the second ller/control food word-pair items
(chicken – beef) were compared between the groups when listening to Schumann’s
“sweet” and Ferneyhough’s “sour” music. The former seemed to have taste associations
Figure 4.
The frequencies of
food-related
associations in
different
experimental group:
“sour” music (Mesz),
n⫽15
NFS
45,3
366
Downloaded by UNIVERSITY OF TAMPERE At 00:00 11 May 2015 (PT)

Table II.
The semantic
similarity values
(mean and standard
deviation std) of
food-related word
associations with the
taste names “sweet”
and “sour” values of
latent semantic
analysis vary
between ⫺1
(semantically
dissimilar to “sweet”
or “sour”) and 1
(semantically similar
to “sweet” or “sour”)
Music/word similarities “Sweet” (Mean) “Sweet” (Std) “Sour” (Mean) “Sour” (Std) t-test “sweet” vs “sour”
Satie n⫽18 0.3 0.15 0.21 0.13 pⴝ0.00001
Schumann n⫽12 0.32 0.16 0.23 0.15 pⴝ0.01
Ferneyhough n⫽20 0.34 0.13 0.30 0.20 p⫽0.2
Mesz n⫽15 0.35 0.19 0.33 0.27 p⫽0.6
Comparative similarity to “sweet” (t-test) Satie n⫽18 Schumann n⫽12 Ferneyhough n⫽20 Mesz n⫽15
Satie n⫽18 p⫽0.5397 p⫽0.1424 p⫽0.0990
Schumann n⫽12 p⫽0.5352 p⫽0.4324
Ferneyhough n⫽20 p⫽0.7569
Mesz n⫽15
Comparative similarity to “sour” (t-test) Satie n⫽18 Schumann n⫽12 Ferneyhough n⫽20 Mesz n⫽15
Satie n⫽18 p⫽0.3025 pⴝ0.0007 pⴝ0.0002
Schumann n⫽12 p⫽0.06 pⴝ0.03
Ferneyhough n⫽20 p⫽0.49
Mesz n⫽15
Notes: The signicance (p-value) of the differences between the groups (statistically signicant differences marked bolded) was determined by ANOVA
and the Tukey’s tpost hoc tests
367
Music
and taste
associations
Downloaded by UNIVERSITY OF TAMPERE At 00:00 11 May 2015 (PT)

Table III.
Taste associations
evoked by the
“sweet” (Satie and
Schumann) and
“sour” (Ferneyhough
and Mesz) music
p-value and music
Chocolate 1
Lingonberry 7
Lemon 1
Banana 7
Chicken 1
Beef 7
Meringue 1 Crisp
rye bread 7
Red currant 1
Pineapple 7
Wheat 1
Oats 7
p-value p⫽0.000 p⫽0.000
1
p⫽0.009
1
p⫽0.003 p⫽0.159 p⫽0.732
Satie n⫽18 2.22 ⫾1.48
A
4.72 ⫾1.56
A
4.06 ⫾2.04
AB
2.61 ⫾1.42
A
3.56 ⫾1.34 4.22 ⫾1.48
Schumann n⫽12 3.00 ⫾1.71
A
3.75 ⫾1.96
A
5.33 ⫾1.23
A
3.25 ⫾1.82
AB
3.75 ⫾1.66 3.83 ⫾1.53
Ferneyhough n⫽20 5.55 ⫾1.43
B
2.00 ⫾1.45
B
2.95 ⫾1.23
B
4.45 ⫾1.61
B
3.80 ⫾1.91 3.90 ⫾1.21
Mesz n⫽15 6.29 ⫾0.73
B
1.67 ⫾0.82
B
3.53 ⫾1.51
AB
4.47 ⫾2.00
B
2.60 ⫾1.59 3.37 ⫾1.63
Notes: The signicance (p-value) of the differences between the music groups (marked with letters A and B) was determined by ANOVA and the Tukey
HSD post hoc tests; the mean values and standard deviations are reported in rows (e.g., 2.22 ⫾1.48); 1variances were not homogenous; either square root
or logarithmic transformations were used
NFS
45,3
368
Downloaded by UNIVERSITY OF TAMPERE At 00:00 11 May 2015 (PT)

related to beef whereas the latter did the same in relation to chicken. This nding is
discussed later in the concluding section of the article. Overall, the ndings concerning
the taste associations convey additional evidence for this study’s key hypothesis.
Ingredients chosen for a drink
Participants were asked to prepare a drink congruent with the music they heard and
write down what ingredients they used for it. They could freely select the ingredients
from the choice set of ve juices and liquid honey (refer to the description of the pre-test
in the Methods and Materials-section). This procedure had a resemblance to the
color-matching experiments by Helmholtz, where participants had to adjust the mix of
three different wavelengths of light to match a given color (Trichromatic Theory of
Color Vision, 2013).
As summarized in Table IV, the total use of ingredients with the higher sugar content
(honey, mango juice and pineapple juice) dominated when listening to “sweet” music,
but more sour ingredients (lemon and grapefruit juice) were used when listening to
“sour” music. For example, the total use of liquid honey in the “sweet” and “sour” music
experiment was 109 and 44 ml while the use of lemon juice was 34 and 160 ml,
respectively.
Samples were taken from each of the drink for the chemical analysis. Based on total
consumption and reported sugar and acid concentrations of the ingredients (Table IV),
the sugar and acid concentrations were predicted to differ between the drinks mixed
when study participants were exposed to “sweet” (Satie and Schumann) versus “sour”
(Ferneyhough and Mesz) music. The results are shown in Table V.
The average total sugar content of the drinks prepared in the room with the “sweet”
music (Satie and Schumann) were higher (125.2 and 121.2 g/l, respectively) than in the
drinks mixed while hearing the “sour” music of Ferneyhough and Mesz (101.2 and
93.6 g/l, respectively). There was a signicant difference between the total sugar
contents of the drinks mixed while listening to the music of Satie versus Mesz
(p⫽0.011), but the difference between the total sugar contents of the drinks prepared
while listening to the music of Satie versus Ferneyhough (p⫽0.051) was not signicant.
The total acid contents were higher in the drinks prepared in the rooms with
Ferneyhough’s and Mesz’s “sour” music being played (11.8 and 11.9 g/l, respectively)
than in the drinks mixed in the rooms lled with Satie’s and Schumann’s “sweet” music
(8.8 and 8.4 g/l, respectively). A signicant difference in the total acid contents of the
Table IV.
The total
consumption of
ingredients with
“sweet” and
“sour” music
Ingredient
Total consumption (ml)/
“Sweet” music
Total consumption (ml)/
“Sour” music
Total sugar
content (%)
Total organic acid
content (%)
Honey 109 44 80 0
Mango juice 1,350 450 10 ⬍0.1
Pineapple juice 1,000 650 30 0.6
Lemon juice 34 160 1.6 4.4
Grapefruit juice 750 1,750 9 0.3
Orange juice 900 650 10 0.6
Notes: Sugar and organic acid contents (%) of ingredients as reported by Fineli® – Finnish Food
Composition Database (Fineli) (2011)
369
Music
and taste
associations
Downloaded by UNIVERSITY OF TAMPERE At 00:00 11 May 2015 (PT)

Table V.
The sugar and acid
contents (g/l) of the
drinks prepared in
the “sweet” (Satie
and Schumann) and
“sour” (Ferneyhough
and Mesz) music
groups
p-value and music Sucrose Glucose Fructose
Total Malic Citric Total Sugar: acid
Sugars acid acid Acids Ratio
p-value p⫽0.000 p⫽0.096 p⫽0.086 p⫽0.007 p⫽0.058 p⫽0.000
1
p⫽0.000 p⫽0.000
1
Satie n⫽18 47.1 ⫾12.3
A
37.4 ⫾9.1 39.4 ⫾11.2 123.9 ⫾26.2
A
1.4 ⫾0.7 7.2 ⫾1.3
A
8.6 ⫾1.6
A
15.4 ⫾6.1
A
Schumann n⫽12 39.0 ⫾9.8
AB
39.9 ⫾12.2 43.1 ⫾14.9 121.9 ⫾34.9
AB
0.9 ⫾0.4 7.4 ⫾1.9
A
8.3 ⫾2.1
A
15.7 ⫾5.9
A
Ferneyhough n⫽20 35.4 ⫾9.7
B
32.0 ⫾11.6 33.9 ⫾13.2 101.2 ⫾29.1
AB
1.2 ⫾0.5 10.6 ⫾3.4
B
11.8 ⫾3.4
B
9.3 ⫾3.4
B
Mesz n⫽15 29.4 ⫾7.3
B
31.7 ⫾9.3 32.4 ⫾9.9 93.6 ⫾23.5
B
1.0 ⫾0.4 10.9 ⫾2.3
B
11.9 ⫾2.1
B
8.3 ⫾3.2
B
Notes: The signicance (p-value) of the differences between the groups (marked with letters A and B) was determined by ANOVA and the Tukey HSD
post hoc test; the mean values and standard deviations (g/l) are reported in rows (e.g., 47.1 ⫾12.3); 1variances were not homogenous; either square root or
logarithmic transformations were used
NFS
45,3
370
Downloaded by UNIVERSITY OF TAMPERE At 00:00 11 May 2015 (PT)

drinks prepared in the rooms with the music of Satie versus Ferneyhough (p⫽0.002)
was found. A similar signicant difference was observed between the “sweet” music of
Satie and the “sour” music of Mesz (p⫽0.003), between the “sweet” music of Schumann
and the “sour” music of Ferneyhough (p⫽0.004) and between the “sweet” music of
Schumann and the “sour” music of Mesz (p⫽0.005). In turn, the differences between the
“sweet” musical pieces of Satie and Schumann and between the “sour” musical pieces of
Ferneyhough and Mesz were non-signicant.
The results concerning the sugar:acid ratios mirrored those of total acid contents (see
above). To conclude, the results from the chemical analysis are supportive of the notion
that the exposure to the “sweet” and “sour” music indeed affected study participants’
choices of ingredients for a drink they were requested to mix. “Sweet” music led to more
sugary drinks and “sour” music to drinks with higher acidity.
Discussion and conclusions
The results indicate that exposure to the “sweet” musical piece and to the “sour” musical
piece inuences food-related thinking processes and even choice behaviors. This study
provides further evidence concerning the quality of associations made between music
and food of similar taste parameters. We found that the participants, when listening to
music which contained “sweet” and “sour” taste elements, made word associations
connected to these elements. These associations inuenced their selection of ingredients
for a drink preparation. It also impacted the sweetness and sourness of the drink that
they prepared.
The food-related word association analysis clearly indicated that individuals make
consistent associations between the music and food that can be characterized as
possessing either sweet or sour taste parameters. The same, freely associated words
were chosen repeatedly by different people for a given music (Figure 1-4). The most
often repeated words were chocolate (for Satie) and tasty (for Schumann), and other
highly repeated words for the “sweet” music les revealed connections with pleasant
tastes, softness and warm temperatures. These words were dessert, soft, velvety,
creamy, enjoyable, wine, ne, warm and sweet. In contrast, the most often repeated
words associated to “sour” music les were fruits (for Ferneyhough) and sour (for Mesz),
and other frequent words related to relatively more unpleasant taste sensations were
hardness and coldness, pungent, hard, icy, cold, strong, sharp, spicy, burning, spicy and
bitter. The terms berries and lemon appear often both for sweet and sour music
selections.
The qualitative analysis of the word-pairs showed that when listening to Satie’s
piece, rich in “sweet” musical elements, the participants preferred sweet food terms such
as chocolate, banana and meringue over the less sweet options. However, this effect was
not as pronounced in the case of Schumann’s music, which is similarly rich in “sweet”
elements. Ferneyhough’s piece, carrying several “sour” musical elements, encouraged
the participants to gravitate toward sour food terms, such as lingonberry and lemon. In
the case of Mesz’s “sour” music, this effect was even stronger.
We showed that sweet or sour elements in the music may even contribute to the
actual degree of sweetness (as measured by the sugar content) and sourness (as
measured by the organic acid content) of drinks. The participants created and prepared
drinks with similar characteristics while being exposed to this kind of music. Drinks
prepared, while listening to “sweet” music, had higher sugar content as compared to
371
Music
and taste
associations
Downloaded by UNIVERSITY OF TAMPERE At 00:00 11 May 2015 (PT)

when prepared under the inuence of “sour” music. The drinks prepared under the
inuence of “sour” music resulted in their higher organic acid content. The effect of
specic musical patterns on consumers’ food-related associations and behavioral
choices are not documented. Consequently, presenting the rst empirical evidence for it
contributes to an emerging research that addresses cross-modal interactions, in the
realm of consumption and cooking of food. Moreover, the incorporation of them as
dependent variables instead of taste perception extends the current academic
knowledge.
These novel ndings, produced by qualitative and quantitative methods, are
consistent with three recent discoveries. First, certain phonetic features (e.g. brand
names) map to specic tastes (Spence, 2012). Second, changing the sonic properties of a
non-musical background soundtrack can modulate consumers taste perceptions of food
products (Crisinel et al., 2012). Third, musical pieces such as Carl Orff’s Carmina Burana
and Michael Brook’s Slow Breakdown can lead consumers to rate the taste of the same
wine in the former case, as more “powerful and heavy”, whereas in the latter case, as
more “mellow and soft” (North, 2012).
In the search of the conceptual explanation for the effects observed in this study,
North’s (2012) elaboration on the cognitive priming theory and symbolic meanings of
music is a viable candidate. These attributes make the basic assumption that the same
terms or concepts are used to characterize both musical and taste sensations (Knöferle
and Spence, 2012). In effect, symbolic priming occurs when the presentation of musical
stimulus triggers interpretation or symbolic associations, which, in turn, primes a
characteristic that happens to be perceived at the same time (Spence and Deroy, 2013,
p. 138). Applying these ideas here suggests that the “sweet” symbolic connotations of
Satie’s and Schumann’s music and the “sour” symbolic connotations of Ferneyhough’s
and Mesz’s music prime mental content that is connected with the concepts of
“sweetness” and “sourness”. This activation results in the increased likelihood of
corresponding food-related thoughts and behaviors due to unconscious expectations
(Spence, 2012). Thus, after being asked to produce food-related associations while
hearing “sweet/sour” music, predominantly sweet/sour ones will come to people’s
minds. This favored sweet/sour ingredient addition when asked to mix a drink.
The research conducted concerning the priming effects of brands, in the eld of
consumer behavior, offers direct evidence for this account. The symbolic meaning of
creativity is associated with companies such as the Apple
©
brand, while the symbolic
meaning of smartness is associated with the IBM
©
brand, among American consumers
(Fitzsimons et al., 2008). This study reveals that the Apple
©
-primed consumers behave
more creatively than the IBM
©
-primed consumers or controls. Thus, this theoretical
rationale should be tested in the future in the research context that combines various
“taste music” and food-related dependent variables (e.g. associations, expectations,
sensory perception, emotion, food choices and even cooking). For example, the
food-related expectations and emotions generated after hearing “sweet”, “sour”, “bitter”
and “salty” music could be compared (cf. Sester et al., 2013) and their role in modifying
taste perceptions and other food-related processes should be analyzed.
It must be acknowledged, though, that in addition to symbolic priming, there are
other alternative accounts for understanding cross-modal associations between music
and taste: the amodality, mediation and transitivity hypotheses. According to the
amodality hypothesis, the coding of certain dimensions of sensory experiences (e.g.
NFS
45,3
372
Downloaded by UNIVERSITY OF TAMPERE At 00:00 11 May 2015 (PT)

magnitude or space) is common across modalities. The mediation hypothesis asserts
that some other indirect commonality (e.g. pleasantness or fastness) explains why all
sensory experiences that share this dimension correspond. Finally, the transitivity
hypothesis can be illustrated as follows. If a sensory feature A corresponds to a feature
B in another modality, and B, in turn, corresponds to a feature C in a third modality, our
brains will generate a cross-modal correspondence between A and C. (Deroy et al., 2013).
The validity of these explanatory formulations should be rigorously tested in future
studies.
The study also demonstrates the complexity of experiments where the connection
between music and taste is under examination. For example, the results concerning the
food word-pair red currant – pineapple were inconsistent with the expectations.
Although pineapple has high sugar content (30 per cent) and is often considered as
sweet, its relatively high organic acid content (0.6 per cent) may make tasters judge it as
a sour rather than as a sweet food. Second, an interesting theoretical implication springs
from the unexpected nding that Satie’s and Schumann’s “sweet” music evoked
stronger beef-related taste associations, while Ferneyhough’s and Mesz’s “sour” music
did the same with regard to chicken-related taste associations. Five participants of 12
associated Schumann’s piano music with “ne dining” and a restaurant milieu. Here, the
question is not necessarily about the cross-modal associations between the auditory and
gustatory stimuli, but about the activation (by the music) of inter-connected mental
constructs (the linkage between ne dining and beef meat eating is postulated to be
stronger than that of between ne dining and chicken meat eating) (Areni and Kim,
1993). Still, this interpretation is speculative and needs to be veried by further studies.
This points to the importance of understanding the symbolic meanings of cultural
objects, practices and places (Allen et al., 2008), especially when they are used as
experimental stimuli (this was controlled in the present study via a pre-test). The fact
that piano music is often played in ne restaurants is a plausible explanation for the
surprising effect referred to above. More generally, this implies that the symbolic
meanings of certain musical pieces (e.g. Beethoven’s fth symphony) are likely to be
differently understood in different socio-cultural contexts (e.g. in a German city versus
in a Tanzanian rural village) (Reyes, 2009). Thus, the cross-cultural perspective on the
cross-modal effects between music and taste perception provides another avenue for
future research.
Limitations of the study
The “sweet” and “sour” music were produced using different instruments. As different
instruments may exert distinct timbre effects on how consumers map sounds onto tastes
(Crisinel and Spence, 2010a), using piano versus ute music is a potential confounder.
Piano music is consistently associated with “sweetness” (Spence, 2012), and ute music
is associated with acidity (Spence et al., 2013). “Sweet” and “sour” music were selected
for the experiment because sweet and sour dimensions are easy to distinguish (Mesz
et al., 2011). Individual factors such as the pre-existing musical and the culinary
expertise, coupled with the usage of and attitude toward the juices and honey (the
ingredients that were available for mixing the drink) were not explicitly measured.
These factors could have moderated the effects detected, although the random
assignment of participants to the experimental groups should decrease the risk of such
a bias, here.
373
Music
and taste
associations
Downloaded by UNIVERSITY OF TAMPERE At 00:00 11 May 2015 (PT)

Due to a lengthy experimental session (ca. 20 min), commencing in a real life setting,
a between-subjects study design was, in practice, the only alternative. This design,
however, is suboptimal because it cannot produce conclusive evidence that it was the
musical elements (sweet vs sour) that affected the ingredients chosen for the drink and
not some other factors (e.g. certain idiosyncrasies in the sweet versus sour experimental
conditions). Thus, replication attempts using within-subjects designs are more apt.
Finally, there were no ller tasks in which the participants were asked to engage in
before they started to prepare the drinks. Moreover, no control questions hinting at the
presumed purpose of the study were posed to them. The questionnaire did not contain
any terms or words that were directly related to sweetness or sourness, and while the
participants were being recruited, they were given only a general idea of the study (e.g.
that is dealt with musical associations). Various sources of potential confounds must be
systematically controlled in future studies.
Practical implications
The key nding of this study also propels a few managerial implications. First, by
carefully considering the symbolic meanings of the music used in different social
situations, creating multimodal experiences and even subconscious expectations in
people’s minds is possible (Krishna, 2012). These subconscious expectations can,
through the assimilation processes (Sester et al., 2013), come to augment consumers’
actual taste experience, which occurs later. Second, ambient “taste music” that is
congruent with the basic avors of the dishes can be played in restaurants to highlight
guests’ sensory experiences. For instance, sweet desserts may be appreciated more
when “sweet taste music” is aired (Crisinel et al., 2012). Third, food marketers can
combine music or sounds with visual, haptic and olfactory stimuli to produce an optimal
mixture of taste emotions that increasingly fascinates sensory researchers at the
moment (Ng et al., 2013).
These ndings suggesting that specic musical elements have an inuence on how a
dish is prepared may have compelling practical implications. Indeed, they encourage
chefs and restaurant-keepers to develop new multimodal culinary experiences. As a
matter of fact, such culinary innovations already exist. One of the landmark example of
these are the “Sounds of the sea” dish served by Heston Blumenthal in the Fat Duck
restaurant in Bray, the UK. There, this dish is served while playing a soundtrack
specically designed for it to highlight the multimodal eating experience. Multimodal
cookbooks, such as those including music recommendations for each dish, are a novel
type of food literacy (Pelaccio, 2012). The signicance of this study lies in the notion that
a soundscape may inuence the design and creation of culinary dishes. This represents
a possibility to raise the culinary art to a new level.
References
Allen, M.W., Gupta, R. and Monnier, A. (2008), “The interactive effect of cultural symbols and
human values on taste evalution”, Journal of Consumer Research, Vol. 35 No. 2, pp. 294-307.
Areni, C.S. and Kim, D. (1993), “The inuence of background music on shopping behavior:
classical versus top-forty music in a wine store”, Advances in Consumer Research, Vol. 20
No. 1, pp. 336-340.
Beeli, G., Esslen, M. and Jänke, L. (2005), “Synaesthesia: when coloured sounds taste sweet”,
Nature, Vol. 434 No. 7029, p. 38.
NFS
45,3
374
Downloaded by UNIVERSITY OF TAMPERE At 00:00 11 May 2015 (PT)

Bronner, K., Frieler, K., Bruhn, H., Hirt, R. and Piper, D. (2012), “What is the sound of citrus?
Research on the correspondences between the perception of sound and avor”, available at:
http://audity-docs.s3.amazonaws.com/citrus_paper_icmpc12_201200514-5066ab939400d.
pdf (assessed 19 November 2013).
Crisinel, A.-S., Cosser, S., King, S., Jones, R., Petrie, J. and 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, Vol. 24 No. 1,
pp. 201-204.
Crisinel, A.-S. and Spence, C. (2009), “Implicit associations between basic tastes and pitch”,
Neuroscience Letters, Vol. 464 No. 1, pp. 39-42.
Crisinel, A.-S. and Spence, C. (2010a), “As bitter as a trombone: synesthetic correspondences in
non-synesthetes between tastes/avors and musical notes”, Attention, Perception &
Psychophysics, Vol. 72 No. 7, pp. 1994-2002.
Crisinel, A.-S. and Spence, C. (2010b), “A sweet sound? Food names reveal implicit associations
between taste and pitch”, Perception, Vol. 39 No. 3, pp. 417-425.
Deerwester, S., Dumais, S.T., Furnas, G.W., Landauer, T.K. and Harshman, R. (1990), “Indexing by
latent semantic analysis”, Journal of the American Society for Information Science, Vol. 41
No. 6, pp. 391-407.
Deroy, O., Crisinel, A.-S. and Spence, C. (2013), “Crossmodal correspondences between odors and
contingent features: odors, musical notes, and geometrical shapes”, Psychonomic Bulletin &
Review, Vol. 20 No. 5, pp. 878-896.
Diuk, C., Fernandez Slezak, D., Raskovsky, I., Sigman, M. and Cecchi, G.A. (2012), “A
quantitative philology of introspection”, Frontiers in Integrative Neuroscience, Vol. 6
No. 80, pp. 1-12.
Ernst, M.O. and Bülthoff, H.H. (2004), “Merging the senses into a robust percept”, Trends in
Cognitive Sciences, Vol. 8 No. 4, pp. 162-169.
Fineli® – Finnish Food Composition Database (2011), “National institute for health and welfare”,
available at: www.neli. (accessed 20 November 2013).
Fitzsimons, G.M., Chartrand, T. and Fitzsimons, G.J. (2008), “Automatic effects of brand exposure
on motivated behavior: how apple makes you ‘think different’”, Journal of Consumer
Research, Vol. 35 No. 1, pp. 21-35.
Hänggi, J., Beeli, G., Oechslin, M.S. and Jaencke, L. (2008), “The multiple synaesthete E.S.:
neuroanatomical basis of interval-taste and tone-colour synaesthesia”, NeuroImage, Vol. 43
No. 2, pp. 192-203.
Holt-Hansen, K. (1968), “Taste and pitch”, Perceptual and Motor Skills, Vol. 27 No. 1, pp. 59-68.
King, B.M., Arents, P., Bouter, N., Duineveld, C.A., Meyners, M., Schroff, S.I. and Soekhai, S.T.
(2006), “Sweetener/sweetness-induced changes in avor perception and avor release of
fruity and green character in beverages”, Journal of Agricultural and Food Chemistry,
Vol. 54 No. 7, pp. 2671-2677.
Knöferle, K. and Spence, C. (2012), “Crossmodal correspondences between sounds and tastes”,
Psychonomic Bulletin & Review, Vol. 19, pp. 992-1006.
Koch, C. and Koch, E.C. (2003), “Preconceptions of taste based on color”, The Journal of Psychology,
Vol. 137 No. 3, pp. 233-242.
Krishna, A. (2012), “An integrative review of sensory marketing: engaging the senses to affect
perception, judgment and behavior”, Journal of Consumer Psychology, Vol. 22 No. 3,
pp. 332-351.
375
Music
and taste
associations
Downloaded by UNIVERSITY OF TAMPERE At 00:00 11 May 2015 (PT)

Laaksonen, O., Ahola, J. and Sandell, M. (2013), “Explaining and predicting individually
experienced liking of berry fractions by the hTAS2R38 taste receptor genotype”, Appetite,
Vol. 61 No. 1, pp. 85-96.
Mesz, B., Sigman, M. and Trevisan, M.A. (2012), “A composition algorithm based on crossmodal
taste-music correspondences”, Frontiers in Human Neuroscience, Vol. 6 No. 71.
Mesz, B., Trevisan, M.A. and Sigman, M. (2011), “Taste of music”, Perception, Vol. 40 No. 2,
pp. 209-219.
Ng, M., Chaya, C. and Hort, J. (2013), “Beyond liking: comparing the measurement of emotional
response using Essense prole and consumer dened check-all-that-apply methodologies”,
Food Quality and Preference, Vol. 28 No. 1, pp. 193-205.
North, A.C. (2012), “The effect of background music on the taste of wine”, British Journal of
Psychology, Vol. 103 No. 3, pp. 293-301.
Pelaccio, Z. (2012), Eat with Your Hands, Ecco, New York, NY.
Reyes, A. (2009), “What do ethnomusicologists do? An old question for a new century”,
Ethnomusicology, Vol. 53 No. 1, pp. 1-17.
Sester, C., Dacremont, C., Deroy, O. and Valentin, D. (2013), “Investigating consumers’
representations of beers through a free association task: a comparison between packaging
and blind conditions”, Food Quality and Preference, Vol. 28 No. 2, pp. 475-483.
Simner, J., Cuskley, C. and Kirby, S. (2010), “What sound does that taste? Cross-modal mappings
across gustation and audition”, Perception, Vol. 39 No. 4, pp. 553-569.
Spence, C. (2012), “Managing sensory expectations concerning products and brands: capitalizing
on the potential of sound and shape symbolism”, Journal of Consumer Psychology, Vol. 22
No. 1, pp. 37-54.
Spence, C. and Deroy, O. (2013), “On why music changes what (we think) we taste”, i-Perception,
Vol. 4 No. 2, pp. 137-140.
Spence, C., Levitan, C.A., Shankar, M.U. and Zampini, M. (2010), “Does food color inuence taste
and avor perception in humans?”, Chemosensory Perception, Vol. 3 No. 1, pp. 68-84.
Spence, C., Richards, L., Kjellin, E., Huhnt, A.-M., Daskal, V., Scheybeler, A., Velasco, C. and
Deroy, O. (2013), “Looking for crossmodal correspondences between classical music and
ne wine”, Flavour, Vol. 2 No. 29.
Trichromatic Theory of Color Vision (2013), available at: http://psychology.about.com/od/
sensationandperception/f/trichrom.htm (accessed 18 November 2013).
Corresponding author
Anu Inkeri Hopia can be contacted at: anu.hopia@utu.
For instructions on how to order reprints of this article, please visit our website:
www.emeraldgrouppublishing.com/licensing/reprints.htm
Or contact us for further details: permissions@emeraldinsight.com
NFS
45,3
376
Downloaded by UNIVERSITY OF TAMPERE At 00:00 11 May 2015 (PT)