Describing product experience in different languages:

The role of sensory modalities

Anna Fenko*, Jacco J. Otten, Hendrik N.J. Schifferstein
Department of Industrial Design, Delft University of Technology, Landbergstraat 15, 2628 CE Delft, The Netherlands
1. Introduction
When people interact with products, they receive information through all the different senses. This sensory information
is processed in the brain and contributes to the overall product experience that people are aware of andcan describe verbally.
In describing their experience, people usually use adjectives that reect sensory properties (e.g., warm, solid, red, loud),
adjectives that refer to symbolic properties (e.g., elegant, expensive, modern), and affective evaluations (e.g., good, bad,
beautiful, ugly) of the product (see Hekkert and Schifferstein, 2008).
All descriptions of product experiences ultimately rely on sensory inputs derived from the product. Nonetheless,
information fromsome sensory modalities may be more important for describing certain product experiences than others. We
dene sensory importance as the relative contribution of each sensory modality to the description of a particular product
experience. Thedominant sensorymodalityis themodalitythat has thelargest effect onthespecicdescription. Inthis research
we wonder what the role of the senses is in the description of product experiences through different types of adjectives.
1.1. Factors affecting sensory dominance
A general and popular belief seems to be that vision is the dominant sensory modality in everyday experience. When
people are asked which sensory modality they would miss most if they lost it, the majority is likely to indicate vision (Fiore
and Kimle, 1997; Schifferstein, 2006). In addition, when people are asked to describe objects, they primarily use adjectives
that refer to the visual (60%) or tactual (32%) modalities (Stadtlander and Murdoch, 2000).
Journal of Pragmatics 42 (2010) 33143327
A R T I C L E I N F O
Article history:
Received 19 June 2009
Received in revised form 25 May 2010
Accepted 31 May 2010
Keywords:
Sensory descriptors
Literal meaning
Metaphorical meaning
Language differences
A B S T R A C T
People describe their product experiences using adjectives that can be divided into three
groups: sensory descriptors (e.g., hard, red, noisy); symbolic descriptors (e.g., interesting,
expensive, modern); and affective descriptors (e.g., pleasant, beautiful). All product
experiences rely on information from sensory modalities. We developed a questionnaire
approach to quantify the relative importance of the ve sensory modalities for various
descriptors of product experience. The approach was used in two studies that employed
Dutch and Russian participants. The results demonstrate that the importance of a sensory
modality differs for various groups of descriptors. Most symbolic descriptors demon-
strated strong visual dominance. Affective descriptors were equally related to all sensory
modalities. Sensory descriptors showed signicant cross-cultural differences. The latter
result can be explained by the different associations between literal and metaphorical
meanings of sensory adjectives in different languages.
2010 Elsevier B.V. All rights reserved.
* Corresponding author. Tel.: +31 15 278 2879; fax: +31 15 278 7179.
E-mail address: [email protected] (A. Fenko).
Contents lists available at ScienceDirect
Journal of Pragmatics
j our nal homepage: www. el sevi er . com/ l ocat e/ pr agma
0378-2166/$ see front matter 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.pragma.2010.05.010
However, vision is not the dominant modality in the interaction with all products. In a study using consumers self-
reported importances, participants reported that they found one of the other sensory modalities more important than vision
during product usage for about half of the products (Schifferstein, 2006). For example, for a computer mouse the tactual
characteristics were most important, for a vacuum cleaner the sound it made, for a cleaning product its smell, and for a soft
drink its taste. The relative importance of a modality may depend on various aspects, such as the availability of sources of
sensory stimulation, the degree of variation in sensory stimulation over various products, the usefulness of the sensory
information during functional use, the proportion of time a modality is used actively, and the role of the stimulation in
enjoying the product. Hence, the roles of the modalities depend not only on whether a certain type of sensory information is
present, but also on whether the information is perceived, how it is processed, and how people react to it emotionally.
Culture may play a role in determining the importance of a sensory modality. Culture has been shown to inuence a large
variety of behaviours across many diverse disciplines, including consumer behaviour (see Hofstede, 2001). Schifferstein
(2006) suggested that the visual systemis regarded dominant, because it plays a signicant role in many daily activities. The
importance of vision in Western societies may have increased over time due to the products that were created, such as books,
television, and computers that require major input from the visual modality. McLuhan (1961) believed that the nature of
media by which people communicate affects the ratio of their senses. For example, the alphabet stresses the sense of sight,
which in turn causes people to think in linear, objective terms. He argued that Europeans and North Americans live in the
visual mode, while for native Africans and other non-literate societies the auditory modality is dominant. However,
anthropological data suggest that signicant diversity exists in sensory dominance among non-literal peoples. Classen
(2005) describes three non-literal societies that have three different dominant sensory modalities: temperature is the most
important sensory property for the Tzotzil of Mexico, smell for the Ongee of the Little Andaman Island, and vision for the
Desana of Columbia.
1.2. Can language inuence product experience?
Language is one of the core components of any culture. It is central to communication and closely related to thought.
Much of human cultural heritage is encapsulated in semantic concepts packed into words (Levinson and Jaisson, 2006).
Through language individuals have access to the large accumulation of cultural ideas, practices and technology which
constitute a distinct cultural tradition (Gumperz and Levinson, 1996). The learning of natural languages is probably the most
complex cognitive task that humans routinely undertake and the major pressure for brain evolution in our species (Byrne
and Whiten, 1988).
Languages differ fundamentally in their semantic categoriesthe concepts built into their grammars and lexicons. It is
quite difcult to nd any exact cross-linguistic matches between linguistically coded concepts (Levinson and Meira, 2003).
To what extent are our ideas and concepts actually embodied in language? A proposal of linguistic relativity emphasizes a
distinctive role of language in interpreting experience and inuencing thought (Gumperz and Levinson, 1996). The idea that
thought is shaped by language is most commonly associated with the writings of Benjamin Lee Whorf. Whorf (1956)
impressed by linguistic diversity, proposed that the categories and distinctions of each language determine a way of
perceiving, analyzing, and acting in the world. In recent years, some studies have claimed evidence that language indeed has
an important inuence on thinking (e.g. Boroditsky, 2001; Bowerman, 1996; Davidoff et al., 1999; Gentner and Imai, 1997;
Levinson, 1996; Lucy, 1992), while others have reported evidence to the contrary (e.g. Heider, 1972; Malt et al., 1999; Li and
Gleitman, 2002).
Experimental evidence has reopened the debate about the extent to which language inuences cognitive processing of
sensory information. Extensive research has been done in domains such as the experience of space (Levinson, 1996; Li and
Gleitman, 2002; Majid et al., 2004), colour (Gilbert et al., 2006; Kay and Kempton, 1984; Robertson et al., 2000), number
(Gordon, 2004; Gelman and Gallistel, 2004; Pica et al., 2004), and time (Boroditsky, 2001; Chen, 2007; January and Kako,
2007; Nez and Sweetser, 2006).
As regards the time domain, speakers of different languages might talk about the future as if it lies ahead of us (in English),
behind us (in Aymara), or below us (in Mandarin Chinese). Behavioural studies suggest that speakers of languages that use
different spatiotemporal metaphors may think differently about time. Indeed, Mandarin speakers were faster to conrmthat
March comes earlier than April if they had just seen a vertical array of objects than if they had just seen a horizontal array.
The reverse was true for English speakers (Boroditsky, 2001; Nez and Sweetser, 2006).
Dramatic cross-linguistic differences have also been reported for the domain of space (Levinson, 1996). Whereas most
languages (e.g. English, Dutch) rely heavily on relative spatial terms to describe the locations of objects (e.g. left/right, front/
back), Tzeltal (a Mayan language) relies primarily on absolute reference (similar to the north/south system). In experiments
with spatial tasks Dutch and Tzeltal speakers saw an arrow pointing either to the right (north) or to the left (south). They
were then rotated 180% and were asked to identify the arrowlike the one they sawbefore. Dutch speakers overwhelmingly
chose the relative solution, while Tzeltal speakers overwhelmingly chose the absolute solution. Furthermore, speakers of
languages preferring absolute coordinates show more accurate skills when asked to indicate the direction of familiar
locations from an unfamiliar site (Levinson, 1996), suggesting that the differences may have everyday correlates.
Studies on occupational accidents in Finland give another example of everyday consequences of language differences.
Occupational accident rates are substantially lower in Sweden than in Finland, and also among the Swedish-speaking
minority within Finland, despite working in the same regions with similar laws and regulations (Salminen and Hiltunen,
A. Fenko et al. / Journal of Pragmatics 42 (2010) 33143327 3315
1995; Johansson and Strmnes, 1995). Researchers explained this difference by structural differences between Swedish and
Finnish. Swedish prepositions can be represented in terms of vector geometry in a three-dimensional space, whereas Finnish
cases can be represented in a two-dimensional space coupled with a third dimension of time (Strmnes, 1973, 1974). The
researchers suggest that the Finns organize the workplace in a way that favours the individual worker over the temporal
organization of the overall production process. Lack of attention to the overall temporal organization leads to frequent
disruptions and accidents (Johansson and Salminen, 1996; Johansson and Strmnes, 1995).
Studies have also shown that language characteristics (such as grammatical gender) can inuence product experience
(Lucy and Gaskins, 2001; Boroditsky, 2001; Boroditsky et al., 2003). For example, when German speakers were asked to
describe a key (a word masculine in German and feminine in Spanish), they were more likely to use words like hard, heavy,
jagged, metal, serrated, and useful, while Spanish speakers were more likely to say golden, intricate, little, lovely, shiny, and
tiny. On the other hand, to describe a bridge (a word feminine in German and masculine in Spanish), German speakers
referred to it as beautiful, elegant, fragile, peaceful, pretty, and slender, while Spanish speakers more often said big,
dangerous, long, strong, sturdy, and towering.
Schmitt et al. (1994) proposed that structural differences between the Chinese and English language affect mental
representations which, in turn, inuence consumer memory of verbal information. The authors showed that unaided brand
recall was differentially affected in Chinese and English when it was spoken compared to when it was written. Chinese brand
attitudes were primarily affected by the match between script associations and brand associations, but brand attitudes of
English names were primarily affected by the match between sound associations and brand associations (Pan and Schmitt,
1996). Furthermore, how much consumers like a brand name translation can depend on whether that name depicts
phonological or semantic characteristics of the original name (Pan and Schmitt, 1996).
Schmitt and Zhang (1998) and Zhang and Schmitt (1998) found that structural aspects of a language can affect one of the
most basic aspects of consumer behaviour: categorization of products. They demonstrated how classiers, a widespread
lexical and syntactical phenomenon in the Chinese, Japanese, and Thai languages, and almost non-existent in Indo-European
languages such as English, German and Spanish, affect the perceived similarity between objects, attribute accessibility, and
concept organization. They also demonstrated the impact of classier-based schemata on inferences about product features
and provided evidence of the effect of classiers on judgment and choice.
1.3. Linguistic descriptions of sensory characteristics
In the present study, we investigate language differences in descriptions of product experience. We assume that language
effects may depend on the type of adjectives used to describe product experience. Some aspects of product experience seem
to be uniquely unimodal: hue can only be experienced by sight, tickle can only be felt by touch, and pitch can only be
differentiated by audition. Nevertheless, our perceptual experience of the world is richly multimodal (Stein and Meredith,
1993). People are able to extract information derived from one sensory modality and use it in another. People can, for
example, knowa shape by touch and identify it correctly by sight. Furthermore, people are able to integrate the impressions
generated by different sensory modalities into a unied, rich percept. In addition, sensory terms that describe physical
properties of things (such as warm and cold) also describe some psychological qualities. Different languages such as
ancient Greek, Thai, Chinese and Hebrewpossess some morphemes that designate physicalpsychological pairings identical
with those found in English (Asch, 1955).
Human language partly operates through metaphors. Metaphors can structure peoples thoughts, govern their activities,
and enable their reasoning from the familiar to the unfamiliar (Lakoff and Johnson, 1980, 1999; Rein and Scho n, 1977).
Metaphors often refer to sensory phenomena. Good ideas are described as brilliant, pleasant dreams as sweet, important
topics as hot, and bright colours as loud. Gallup and Cameron (1992) argued that English and other European languages
mainly use visual terms as metaphors to describe or capture mental events (as illustrated by, reect on that, shed some
light, a clear explanation) by referring to human evolutionary heritage and the anatomical structure of the brain. Compared
with other sensory input, a disproportionate amount of brain tissue is devoted to the processing of visual information (Van
Essen et al., 1990).
Because people do not rely on each of the ve senses equally and use some sensory modalities more often than others, it
makes sense that certain sensory modalities have greater frequency in linguistic representation. Based on data from 53
languages from all parts of the world, Viberg has shown that the large amount of polysemy with respect to the sense
modalities for verbs is constrained by a hierarchy which gives the most prominent place to vision and the least prominent to
smell and taste (Viberg, 1984). The hierarchy can be presented as follows: sight > hearing > touch > smell, taste. According
to this hierarchy, a verb having a basic meaning belonging to a sense modality higher (to the left) in the hierarchy can get an
extended meaning that covers some (or all) of the sense modalities lower in the hierarchy (pp. 136137). A sense that holds
the top position has the highest number of polysemic patterns and also the highest number of lexemes at the level belowthe
basic one. Viberg concludes that the verbs of perception, especially see, have an extensive range of cognitive meanings, and
that know and see are covered by the same word in a number of languages. For example, in English it is acceptable to say
any of the following: (1) I listened to the voice to see if I could recognize it. (2) I felt his hand to see howhot it was. (3) I
tasted the soup to see whether it was ready. (4) I smelled the air to see how fresh it was. Miller and Johnson-Laird (1976)
also acknowledge the startling complexity of the verb see and link it to the conceptual complexity of visual processes.
However, none of these linguists considers the extension of the meaning of sensory verbs to be metaphorical.
A. Fenko et al. / Journal of Pragmatics 42 (2010) 33143327 3316
English verbs of perception and their complex polysemous structure and etymology have also been studied by Sweetser
(1990). UnlikeViberg, Sweetser argues that the connectionbetweenthe domainof external, physical sensationonthe one hand
and abstract, cognitive states, on the other hand, is metaphorical in nature. According to the cognitive theory of metaphor,
verbal metaphors, including conventional expressions based on metaphor, reect underlying conceptual metaphors in which
people conceptualize vague, abstract domains of knowledge (such as time, causation, ideas, and emotions) in terms of more
specic, familiar, and concrete knowledge (Gibbs, 1994). Human thinking is organized from simple to complex, and the basic
sensory experience (source domain) is used to understand more abstract phenomena (target domain). Sweetser suggests three
reasons for the prominent relationship between vision (as a source domain) and knowledge (as a target domain): (1) The
focusingabilityof thesenseof visionenablespeopletopickuponestimulus frommany, and, together withitsintentionalityand
directionality, differentiates it fromthe other senses. (2) Vision is our primary source of objective data about the world. Visual
features are the most marked in childrens early categorization, thus form an important source for concept formation. Vision
provides data from a distance, which through metaphoric transfer become a characteristic of objectivity in the intellectual
domain. (3) Vision is identical for different people who can take the same point of view.
Some authors argue that visual metaphors for mental processes are of relatively recent date. The salient position vision
occupies in conceptualization of the intellect is not shared by all cultures or even present in older stages in the Indo-
European culture (Ibarretxe-Antuano, 2008). Classen (1993) suggests that sight-based words such as bright, brilliant, and
lucid have been used to mean intelligent only since the Enlightenment period, perhaps in consequence of the general rise of
visualism at that time (p. 58). Many more terms of thought are tactile or kinaesthetic, such as apprehend, comprehend,
conceive, grasp, ruminate and understand. Thought is, or was, experienced primarily in terms of touch [. . .]. Knowing was
less like seeing than like holding (Classen, 1993:58).
Auditory verbs rarely serve as metaphors for thought or intelligence in English, perhaps because hearing is conceived of as
a passive sense, receiving information but not probing it. The main function of hearing is communication, and it is the major
path of interpersonal inuence. Therefore, hearing is associated with obedience and emotional receptivity (being deaf to
someones plea) rather than with intelligence (Sweetser, 1990). However, languages differ in the metaphorical use of
auditory verbs. For example, Evans and Wilkins (2000) showed that Australian languages recruit verbs of cognition like
think and know from hear, not from see. Suya Indians of Brazil use the same verb for listen, understand, and know
(Seeger, 1975), and in Russian the word (understand) has the same root as (listen carefully).
Describing the semantics of three Russian perception verbs, (perceive), (sense), and
(feel), Iordanskaja (1979) notes that Russian semantics reects the naive conceptionof smelling, feeling, and
taste as opposed to sight and hearing. The verb (perceive) can stand for any mental ability, including reason,
sight, and hearing, while the verbs (sense) and (feel) stand for olfaction, touch, taste, kinesthetic
and other internal senses. In addition, (feel) can be metaphorically used for intuition, instinct and the
unconscious mind. Optical or acoustic perception cannot be called or in Russian, which coincides
with psychological theories that consider sight and hearing to be the most sophisticated forms of perception, in contrast to
the other three senses.
A lot of perceptual (or sensory) adjectives, such as sharp, dull (touch), sweet, sour (taste), loud, and quiet (sound), often
extend their meaning from one basic, or prototypical, sense modality to one or more secondary modalities. From Ullmanns
(1957) work based on an investigation of 19th century poetry the following two hierarchies can be derived:
touch > taste > scent and touch > sound, sight. Out of a total of 2009 transfers, he found 1665 upward transfers (from
left to right) and only 344 downward transfers. The largest number of transactions occurred between touch and sound.
Ullmanns study mainly deals with literary metaphor, and many of the examples he provides are certainly not part of
established usage. Williams (1976) looked at historical changes of English adjectives referring to sensory experience in
English dictionaries. He found that such adjectives transferred according to a hierarchy that accords well with the one
presented by Ullmann. Williams noticed that metaphorical transfers, with relatively fewexceptions, go in one direction. That
is, a touch word may transfer to taste (hot) or directly to sound (sharp) or sight (soft). A taste word may transfer to smell
or to sound (sour smell, sweet music), and sight words may transfer to sound and vice versa (loud colour, clear sound).
Williams assumes the hierarchy to be biologically based. The transfers go from the physiologically least differentiating,
evolutionary and ontogenetically primitive sensory modalities to the most differentiating, most advanced.
It is puzzling that the hierarchies of adjectives and verbs differ substantially. While the verb to see can have many non-
visual meanings, the touch-related adjectives can transfer their meanings to the majority of other sensory domains. The
reason behind this discrepancy may be that transfer of meaning in verbs and adjectives refer to different areas of human
experience. Lehrer (1978) suggests that the transfer of meaning in adjectives is mainly based on the general experiential
dimensions of intensity and evaluation. In the early 1950s, Osgood asked participants to evaluate different concepts and
objects on a number of bipolar scales based on semantic opposites, such as goodbad, softhard, fastslow, cleandirty,
valuableworthless, fairunfair, and so on (Osgood et al., 1957). Osgood called them semantic differential scales, because
they differentiated personal attitudes based on a subjective understanding of the connotative meanings of words. Factor
analysis of the responses generally yielded three dimensions: Evaluation or valence (e.g. goodbad, pleasantunpleasant),
Activity (e.g. fastslow, stimulatingrelaxing, livelyquiet), and Potency (e.g. weakstrong, gentletough, persistent
accommodating). Evaluation (valence) and Activity (arousal) are often seen as the two primary dimensions for emotional
experiences (see, e.g., Feldman Barrett and Russell, 1998). According to Lehrer (1978), they may also be the most important
dimensions for the transfer of meaning in sensory adjectives.
A. Fenko et al. / Journal of Pragmatics 42 (2010) 33143327 3317
The sense of touchhas always been related to the eld of emotions. Expressions such as Imdeeply touched or touching
words are widely used in English. Already in 1921, Hans Kurath classied sense perception with respect to emotions and
stated how the kinaesthetic, the visceral, and the tactual perceptions have a relatively stronger tone than those of hearing
and especially of sight, the taste-smell perceptions taking a middle ground (p. 39). Kurath explained this transfer of
meaning fromsense perception to emotion on the basis of the similarity of feelings that both domains share. This connection
can also be explained by the etymology. Buck (1949) points out that the general word in West Germanic languages for feel
refers not only to tactile perception but also to emotions, even in the earliest periods of the languages.
Viberg (1984) argues that, unlike the hierarchy of sensory adjectives, which is probably based on the Evaluation and
Activity dimensions, the hierarchy of sensory verbs is based on another experiential dimension, related to the degree of
certainty. For example, the statement I saw that the building was huge implies more certainty than I heard that the
building was huge, which depends on the reliability of the source of information. The statement I feel that Mary has doubts
about her marriage is the least certain. The dimension of certainty is most relevant for verbs connected to sight, hearing, and
touch.
For the verbs connected to taste and smell the evaluative component seems dominant (to taste freedom, to smell
treason). There are relatively fewolfactory terms in English and most of themrefer to bad smells. Metaphorical meanings for
smell verbs include the detection of bad characteristics, as in Something about his testimony stinks (Caplan, 1973; Viberg,
1984; Sweetser, 1990) and to suspect, to guess, to sense something intuitively, as in She could smell money, power, victory
(Ibarretxe-Antuano, 1999). While the connotation of smelly is often negative, that of tasty is positive, which Classen
explains by the fact that people are confronted with foul smells more often than with foul tastes: We can choose our food,
but we cannot as readily close our noses to bad smells (p. 53). Taste guratively means judgment of what is beautiful and is,
therefore, characterized as a sense of aesthetic discrimination. This relation between taste and preferences is very common
cross-linguistically (Buck, 1949).
1.4. Present research
The present research aims to nd out which sensory modalities are dominant for different descriptors of product
experiences. We assume that for sensory descriptors of product experience the corresponding sensory modalities will be
dominant, such as taste for sweet, touch for hard, and audition for loud. If an adjective has both a literal and a gurative
meaning, we expect the link with the modality corresponding to the literal meaning to be stronger, because this is the
original meaning. As an original source is likely to have a stronger effect than a derivative, we expect the original modality to
have a stronger connection to the descriptor than any of the modalities that are involved in the gurative meanings.
Symbolic and affective descriptors of product experience are likely to be multisensory, but possibly one or more sensory
modalities play a more dominant role in assessing how interesting, modern, exciting or funny people nd a particular
product.
In Study 1 we try to nd experience descriptors that are related to all sensory modalities to the same degree and are not
biased towards any modality. According to our assumption, we expect to nd these among the symbolic and affective
descriptors. In Study 2 we look for adjectives that are mainly related to a single sensory modality. We expect to nd these
mostly among the sensory descriptors. In Study 2 we also test whether a difference exists between the sensory dependence
of the three types of product experience descriptors (sensory, symbolic and affective) for participants with different native
languages.
2. Study 1: selecting a sensory neutral experience
The aim of this study was to nd product experience descriptions that are not biased towards any of the modalities. We
assumed that concrete, sensory descriptors are likely to refer to a single sensory modality. An adjective that can be used in a
literal sense for one modality and in a metaphorical sense for other modalities (e.g., rough, heavy) is likely to be dominated
by the modality for which the descriptor has a literal meaning (i.e., touch). Therefore, we were looking for adjectives that
were abstract and had a meaning that could be applied to the same degree to all sensory modalities.
To develop a new index for evaluating aesthetic impressions in close relation to human modalities, Suzuki and Gyoba
(2001) introduced the sensory relevance coefcient. This coefcient indicates the extent to which pairs of adjectives are
related to sensory modalities. For each adjective pair, the coefcient reects the proportion of participants that indicated
that a particular modality was related most to this adjective. Examples of adjective pairs that were judged to be primarily
related to visual perception were beautifulugly (0.95), gaysober (0.99), and clearcloudy (0.86). Primarily tactual were
wetdry (0.91), softhard (0.96), coldwarm (0.91), and smoothrough (0.78). The only adjective pair that was mainly
auditory was livingquiet (0.83). No adjective pairs in their study were primarily olfactory or gustatory.
In a more recent study Gyoba et al. (2005) instructed participants to indicate the degree of sensory relevance of adjective
pairs on 6-point rating scales for 10 sensory modalities, including the perception of pain, warmness, coldness, equilibrium,
and kinaesthesia. In general, they obtained high ratings for vision for most pairs of adjectives. They found that adjective pairs
that usually span the Evaluation factor in Semantic Differential studies (Osgood et al., 1957), such as pleasantunpleasant
and goodbad tended to get high ratings for multiple modalities. Adjective pairs that span the Activity factor, such as quiet
noisy, staticdynamic, and calmturbulent received primarily high ratings for the auditory and kinaesthetic modalities.
A. Fenko et al. / Journal of Pragmatics 42 (2010) 33143327 3318
Adjective pairs that are related to the Potency factor, such as smoothrough, softhard, smallbig, and femininemasculine
were rated high for the tactile sense. These outcomes have been partly supported by brain activity measurements while the
participants rated stimuli on semantic differential scales (Suzuki et al., 2005). Activation patterns of the temporal and
parietal regions of the brain were signicantly related to the semantic polarities of Activity and Potency, but no changes in
brain activity were related to the ratings on the Evaluation scale.
To determine the extent to which adjective pairs are related to the different sensory modalities, we need to assess the
extent to which all modalities are related to the adjectives. In line with the studies performed by Gyoba, Suzuki, and
colleagues, we asked participants to rate the relevance of four sensory modalities for each semantic scale on a 5-point scale.
Because the research was performed in the context of the evaluation of durable consumer products (see Schifferstein et al.,
2010), the taste modality was not included in this study.
2.1. Participants
A convenience sample consisting of colleagues, friends and acquaintances was used, consisting of 57 participants (42
males and 15 females). All participants were native Dutch speakers. One female participant was removed from the sample,
because she was unable to smell anything. Ages varied between 20 and 60 years (mean 35.4).
2.2. Procedure
Participants received an email that invited them to go to a specic website. The introduction to the study pointed out
that the way in which people experience products depends on the information perceived through the senses. Judging
different aspects of the product experience, however, would not necessarily rely on each sensory modality to the same
degree. Participants then answered the question Suppose that you would have to rate a product on the following semantic
scale, to what extent would the sensory modalities play a role? on a 5-point scale, ranging fromnot at all onthe left side to
very large extent on the right side. A set of 39 possible descriptors was derived from previous studies that used the
Semantic Differential Method. In addition, 20 scales were derived from a set of unipolar product personality descriptors
(Govers, 2004) by creating items that ranged fromnot at all to very. All original items and their translations can be found
in Table 1.
Table 1
Mean sensory relevance ratings for bipolar and unipolar descriptive attributes on a 5-point scale.
English Dutch Version Modality
Vision Audition Touch Smell
Evaluation
Beautifulugly Mooilelijk A 5.0 3.7 2.9 2.2
B 4.7 3.4 2.7 2.6
Pleasantunpleasant Aangenaamonaangenaam A 4.5 4.3 4.5 4.5
Agreeabledisagreeable Prettigonprettig B 3.8 3.9 3.9 3.9
Goodbad Goedslecht B 3.6 3.4 3.2 2.8
Friendlyunfriendly Vriendelijkonvriendelijk B 3.9 3.7 2.9 2.1
Attractiveunattractive Aantrekkelijkonaantrekkelijk A 4.7 3.7 3.9 4.0
SeductiveRepulsive Verleidelijkafstotelijk A 4.7 3.7 3.9 4.4
Sympatheticunsympathetic Sympathiekonsympathiek B 3.8 3.8 2.5 2.4
Invitingrejecting Uitnodigendafwijzend A 4.3 3.9 3.2 3.1
Activity
Activerelaxed Actiefontspannen A 4.3 3.7 3.0 2.1
B 3.9 3.4 3.5 2.3
Stimulatingrelaxing Stimulerendontspannend B 3.4 3.8 3.5 2.9
Excitedcalm Opgewondenkalm B 4.1 4.3 3.4 2.6
Fastslow Snellangzaam B 4.2 3.6 2.8 1.5
Fussylethargic Druksloom A 4.5 3.8 2.1 1.4
Tensefusty Gespannenduf A 4.1 3.1 3.1 2.0
Flashydiscrete Flitsenddiscreet A 4.6 3.5 2.3 1.9
Conspicuousinconspicuous Opzichtigonopvallend B 4.6 3.4 2.0 2.7
Livelyquiet Levendigrustig A 4.5 4.5 3.1 2.5
Potency
Gentletough Schattigstoer A 4.8 3.7 3.2 2.2
B 4.4 3.4 3.3 2.3
Masculinefeminine Mannelijkvrouwelijk B 4.5 3.7 3.8 3.5
Strongweak Sterkzwak B 3.7 3.0 3.9 2.3
Carefulbrave Voorzichtigdapper A 4.3 3.0 2.6 1.9
Momentoushumble Gewichtignederig A 4.3 3.3 2.6 1.7
A. Fenko et al. / Journal of Pragmatics 42 (2010) 33143327 3319
To limit the number of questions the participants had to answer, two separate questionnaires were used,
each containing about half of the semantic items. To evaluate between-sample reliability, six items were identical in the
two versions. The order in which the modalities were presented was different for both surveys. For each semantic
item, the sensory relevance was rated for olfaction, touch, audition, and vision (version A) or audition, olfaction, vision,
and touch (version B), consecutively. Thirty-three (version A) or 32 (version B) items were rated. Twenty-nine
participants lled out version A, 28 lled out version B. The time needed to ll out the questionnaire was approximately
15 min.
2.3. Results
For each item, responses on the 5-point scales were subjected to repeated measures ANOVA with Modality as within-
subjects factor. Most items showed biases towards a sensory modality (see Table 1). In contrast to Gyoba et al. (2005), we
found that most items that referred to the Activity and Potency dimensions were most strongly related to the visual
modality. In accordance with Gyoba et al. (2005), most of the items spanning the Evaluation factor were related to multiple
modalities. Only four semantic scales did not show a signicant difference between sensory modalities: agreeable
disagreeable (p = 0.83), pleasantunpleasant (p = 0.55), stimulatingrelaxing (p = 0.05), and goodbad (p = 0.05). Therefore,
the Pleasantness of the product seems to be a sensory neutral product experience. This experience can be assessed by
averaging the responses on three items that appear to be sensory neutral (agreeabledisagreeable, pleasantunpleasant, and
goodbad).
Table 1 (Continued )
English Dutch Version Modality
Vision Audition Touch Smell
Impressivemeaningless Indrukwekkendnietszeggend A 4.5 3.9 2.8 2.4
Persistentaccommodating Vasthoudendmeegaand A 3.6 3.2 2.9 1.7
Other
Roughsoft Ruwzacht A 4.2 3.0 5.0 2.0
B 3.8 2.5 4.7 1.6
Freshmusty Frismuf A 3.4 1.9 2.6 5.0
B 3.0 2.0 2.6 4.7
Loudquiet Luidstil A 2.3 4.9 2.2 1.3
B 1.9 4.9 2.2 1.4
Matureyouthful Volwassenjeugdig B 4.1 3.4 2.5 2.1
Youngold Jongoud B 4.3 3.1 2.8 3.3
Interestingboring Interessantsaai B 4.3 3.9 3.0 2.5
Funnyserious Grappigserieus B 4.0 4.0 2.2 1.6
Safedangerous Veiliggevaarlijk B 4.3 3.9 3.5 3.2
Severemild Strengmild B 3.4 3.5 2.7 2.0
Expensivecheap Duurgoedkoop A 4.4 2.7 3.7 2.6
Valuableworthless Waardevolwaardeloos A 4.6 3.6 3.8 2.6
Comprehensibleincomprehensible Begrijpelijkonbegrijpelijk A 4.2 3.1 2.3 1.3
Predictableunpredictable Voorspelbaaronvoorspelbaar A 4.0 3.5 3.0 2.6
Moderntraditional Moderntraditioneel A 4.7 3.4 2.9 2.8
Personality
Dominant Dominant B 4.0 4.1 2.4 2.4
Cute Schattig B 4.3 3.2 3.1 2.1
Provocative Uitdagend B 4.3 2.9 2.7 2.2
Cheerful Vrolijk B 4.3 4.2 2.3 1.7
Pretty Leuk B 4.1 3.7 3.0 2.2
Childish Kinderachtig B 4.0 3.6 2.6 2.0
Interesting Interessant B 4.2 3.9 3.1 2.5
Silly Dom B 3.5 3.7 2.0 1.4
Relaxed Relaxed B 3.4 3.9 3.4 2.2
Idiosyncratic Eigenzinnig B 3.7 3.8 2.6 2.3
Lively Pittig A 2.9 1.8 4.1 2.2
Open Open A 4.4 3.1 2.9 1.6
Aloof Afstandelijk A 4.2 3.1 3.0 2.1
Honest Eerlijk A 3.8 3.6 2.7 2.1
Untidy Slordig A 4.7 2.8 2.8 1.8
Boring Saai A 4.5 3.9 2.7 2.3
Modest Bescheiden A 4.0 4.1 2.5 2.1
Easygoing Vlot A 4.5 3.8 2.3 2.0
Obtrusive Opdringerig A 4.3 4.3 3.0 3.2
Serious Serieus A 4.2 4.0 2.7 2.0
A. Fenko et al. / Journal of Pragmatics 42 (2010) 33143327 3320
To test if age or gender differences between respondents inuenced our results, we performed repeated measures ANOVA
on the A and B data sets, with Modality and Experience as within-subject factors and Age and Gender as between-subject
factors. All effects that included the two demographic factors were not signicant (all p > 0.20).
3. Study 2: modality importance for sensory product descriptions
Almost all product experiences have associations with more than one sensory modality. The colour of a dress can be
loud; the voice of a singer can be sweet, and so on. The primary aim of the second study was to determine which sensory
descriptors of product experience rely mainly on a single modality and which are primarily multisensory. We included
several symbolic and affective experiences in this study as control variables. Based on the results of Study 1, we assumed that
symbolic and affective experiences would be mostly multisensory.
Another aimof Study 2 was to look at language differences in modality dominance for three types of product descriptors:
sensory, symbolic, and affective. When the meaning of a product is expressed in words, it may be interpreted differently in
different languages. Most adjectives that describe product experiences have several meanings, and usually not all these
meanings can be translated adequately to another language. For example, the English word fresh has 16 different meanings
(Simpson and Weiner, 1989), which can be roughly divided into two groups: (1) new, recent, newly made, recently arrived,
retaining its original qualities, not deteriorated or changed by lapse of time; (2) pure, invigorating, refreshing (said especially
of air and water), not stale, musty, or vapid. In the Dutch language two different words are used to indicate these two
meanings in the case of food products (vers for the rst meaning and fris for the second meaning). When a text is translated
fromEnglish into Dutch, the translation of the word fresh is likely to have a more restricted meaning and fewer associations
in Dutch than in English. We assumed that some descriptors of product experience are culturally specic and bear linguistic
associations. To test this assumption, we conducted the study with two groups of respondents: native Dutch speakers and
native Russian speakers.
3.1. Participants
The sample consisted of 57 Dutch-speaking students of TU Delft (30 men and 27 women) and 55 Russian-speaking
students (27 men and 28 women) of the Higher School of Economics (Moscow, Russia). For the Dutch sample, ages ranged
from 21 to 34 years, mean age was 23.5 years. For the Russian sample, ages ranged from 17 to 36, mean age was 20.2 years.
3.2. Procedure
Respondents received paper questionnaire forms at a lecture and lled themin during the break. The questionnaire asked
respondents to indicate to what extent different sensory modalities contributed to the evaluation of 34 product properties.
The list included sensory descriptors: tactile (such as warm, sharp, and rough.); auditory (such as noisy, quiet, and loud);
visual (such as colourful, shiny, and clear); olfactory (such as fresh and stale), and gustatory (such as bitter and sweet);
affective descriptors (such as exciting, funny, and cute); and symbolic descriptors (such as complex, modern, and luxurious).
Respondents were asked to think of any product (a coffee maker, shoes, cheese, a tooth brush, shampoo, a camera, a chair, a
soft drink, a bag, etc.) and to answer the question: To what extent do the following senses contribute to your evaluation of a
product as. . .? They assessed the importance of 5 sensory modalities on 5-point scales from not important (1) to very
important (5). Because the second study was performed in the context of the evaluation of a set of products that included a
soft drink (see Fenko et al., 2009), taste was added to the list of modalities in this study.
There were two types of questionnaires which differed with respect to the order of the sensory modalities. The A-form
used the following sequence: audition, olfaction, touch, taste, and vision. The B-formused the sequence: touch, taste, vision,
olfaction, and audition. In the Dutch sample, 26 lled out the A-form, and 31 lled out the B-form. In the Russian sample, 27
respondents lled out the A-form, and 28 lled out the B-form. It took participants 1015 min to ll out the questionnaire.
3.3. Results
We performed repeated measures ANOVA with Descriptor and Modality as within-subjects factors and Language as
between-subjects factor. Themaineffects for Descriptor, ModalityandLanguageweresignicant: F(33, 3234) = 39.9; p < 0.001
for Descriptor; F(4, 392) = 332.8; p < 0.001 for Modality; and F(1, 98) = 4693.5; p < 0.001 for Language. The Modality Lan-
guage interaction was not signicant: F(4, 392) = 1.5; p > 0.2. The Modality Descriptor interaction [F(132, 12936) = 133.5;
p < 0.001] and the Descriptor Modality Language interaction [F(132, 12936) = 13.8, p < 0.001] were both signicant.
To test if age or gender differences between respondents inuenced results, we also performed repeated measures
ANOVA with Descriptor and Modality as within-subjects factors and Language, Age and Gender as between-subject factors.
The effects of both demographic factors and their interactions with other factors were not signicant (all p > 0.20).
Overall mean ratings of the 34 experiences showed that vision had the highest rating for 14 descriptors, touch for 10, taste
for 4, and audition and olfaction for 3 descriptors each. Symbolic descriptors tended to rate high on all modalities, although
all of them demonstrated visual dominance. As predicted, sensory descriptors demonstrated the highest ratings for the
corresponding sensory modalities (e.g., touch for warm, taste for bitter, audition for loud, olfaction for fresh).
A. Fenko et al. / Journal of Pragmatics 42 (2010) 33143327 3321
Because the three-way interaction was highly signicant, responses for each adjective were subjected to repeated
measures ANOVA with Modality as within-subjects factor and Language as between-subjects factor. The effect of Modality
was signicant for all descriptors (p < 0.01). For 24 out of 34 descriptors, the effect of the Modality Language interaction
was also signicant (see Table 2).
Table 2
Mean relevance ratings of modalities for 34 descriptors of product experiences.
Language Modality
English Dutch/Russian Vision Audition Touch Olfaction Taste
Sensory descriptors
Visual
Colourful
**
eurig 4.7 1.7 1.9 3.7 2.1
4.9 1.4 1.6 1.4 1.4
Clean
**
schoon 4.6 1.4 3.7 4.3 2.7
4.8 2.0 3.9 3.5 2.6
Shiny
**
glanzend 4.9 1.5 3.6 1.1 1.1
5.0 1.6 1.7 1.6 1.8
Clear helder 4.7 3.0 2.1 2.1 2.3
4.5 2.8 2.2 2.1 1.8
Conspicuous
*
opvallend 4.7 3.6 3.5 3.4 3.2
4.7 3.7 2.8 3.0 2.6
Beautiful mooi 4.9 3.2 3.4 2.5 2.1
4.8 2.9 2.9 2.3 2.0
Breakable
**
breekbaar 4.4 2.9 4.3 1.1 1.2
4.2 2.1 4.4 1.3 1.5
Quick
*
snel 4.7 3.9 2.4 1.5 1.3
4.8 3.3 2.4 1.6 1.4
Tactile
Warm
**
warm 3.3 1.9 4.6 2.1 2.6
2.3 1.3 4.6 2.0 3.5
Sharp
**
scherp 4.0 2.3 4.4 2.4 3.2
2.9 1.6 3.4 2.5 4.3
Hard
**
hard 3.8 3.0 4.7 1.3 1.9
3.3 1.6 4.8 1.3 2.4
Rough
**
ruw 4.1 2.2 4.8 1.4 1.8
4.1 3.1 4.3 1.9 2.5
Strong
*
sterk 4.2 2.5 4.4 2.3 2.5
3.9 2.0 4.5 1.3 2.1
Heavy
*
zwaar 4.1 2.5 4.6 1.4 1.2
4.0 2.6 4.3 1.4 1.8
Flexible exibel 4.0 2.2 4.6 1.2 1.2
4.1 2.0 4.7 1.3 1.8
Moist vochtig 3.9 1.7 4.7 2.5 2.8
3.8 1.8 4.7 2.9 3.4
Auditory
Loud
*
luid 2.7 4.9 1.6 1.2 1.2
2.1 4.8 1.6 1.4 1.3
Quiet
**
stil 2.6 4.9 2.1 1.2 1.2
4.1 4.3 3.1 2.4 1.8
Noisy lawaaierig 3.1 4.9 1.8 1.3 1.2
3.1 4.9 1.6 1.3 1.2
Olfactory
Fresh fris 3.9 1.8 2.7 4.4 4.1
4.1 1.9 3.3 4.6 4.5
Stale
**
muf 3.0 1.6 2.2 4.7 3.4
4.5 2.2 3.7 3.0 2.5
Gustatory
Bitter
**
bitter 2.2 1.2 1.5 3.2 4.8
1.8 1.3 1.4 2.4 4.9
Sweet
*
zoet 3.1 1.4 1.8 4.3 4.9
3.2 1.4 2.1 3.7 4.9
Sour zuur 2.4 1.1 1.3 3.8 4.9
2.8 1.2 1.6 3.5 4.9
Spicy
**
kruidig 2.6 1.1 1.6 4.2 4.7
2.2 1.5 1.6 4.6 3.1
Mild
**
mild 2.6 1.8 2.4 3.5 4.4
3.4 2.1 4.8 1.8 2.6
A. Fenko et al. / Journal of Pragmatics 42 (2010) 33143327 3322
Some of the cultural differences were big enough to affect the importance hierarchy of modalities for particular
descriptors. For example, spicy appeared to be mainly a gustatory experience for Dutch respondents (kruidig) and olfactory
for the Russian sample ( ). The dominant modality for mild in the Dutch group (mild) was also taste, but in the
Russian group ( ) touch was dominant, followed by vision. Olfaction was dominant for stale in the Dutch group (muf),
but for Russians stale ( ) was a visual experience. For Dutch respondents pure (puur) was a gustatory experience,
while for Russians ( ) it was visual.
These differences encouraged us to perform further analysis on the items. We looked at the response distribution
patterns of all the items to evaluate whether the responses were distributed normally within each language group. In the
Dutch sample 13 bi-modal distributions (7.6%) were observed by visual inspection: one for olfaction (for colourful), 4 for
audition (hard, pure, modern, and cute), 4 for touch (luxurious, pure, modern, and quick), and 4 for taste (sharp, luxurious,
clean, and conspicuous). In the Russian sample there were 31 bi-modal distributions (18.2%): 3 for vision (sharp, sweet,
and sour); 8 for audition (exciting, luxurious, rough, complex, pure, modern, cute, and clear), 9 for touch (exciting, sharp,
quiet, complex, pure, funny, modern, beautiful, and conspicuous), 5 for olfaction (bitter, luxurious, modern, cute, and
beautiful), and 6 for taste (warm, hard, luxurious, rough, complex, and clean). The bi-modal distributions suggest that
there are sub-groups of participants within each sample that differ in their opinion on modality importance for a
particular descriptor.
4. Discussion
4.1. Sensory relevance of adjectives in two languages
We set out to determine the importance of various sensory modalities for different types of descriptors of product
experience. As predicted, the affective descriptors (pleasantunpleasant, stimulatingrelaxing, and goodbad) relied
equally on all sensory modalities. The symbolic descriptors (complex, luxurious, modern, interesting) were also
multisensory, but reliedmainlyonthe visual modality. The latter result corresponds tondings of visual dominance bothin
the eld of product experience (Schifferstein, 2006) and in the area of sensory semantics (Viberg, 1984; Sweetser, 1990). As
we predicted, for many sensory descriptors the corresponding sensory modality was dominant (e.g., audition for noisy,
taste for bitter, touch for hard, vision for shiny). Nevertheless, we also found several sensory descriptors for which more
than one modality was important. For example, fresh is mostly an olfactory adjective, but it has also high importance
ratings for taste, vision, and touch in both the Dutch and the Russian sample. Most tactile adjectives (rough, heavy, moist,
warm, exible) also have high importance ratings for the visual modality. This agrees to Williams (1976) suggestion that
touch is the main source domain and vision is the main target domain for the metaphorical transfer of meaning in sensory
adjectives.
Signicant language differences were found for various sensory descriptions (such as colourful, rough, quiet, spicy, and
stale). The possible explanation for this result is that semantic connotations for sensory adjectives differ considerably
between the languages. For instance, scherp in Dutch and in Russian are bothequivalents to the English word sharp,
of which the literal meaning is tactile (having a keen edge or point). In English and Dutch it is also possible to characterize a
Table 2 (Continued )
Language Modality
English Dutch/Russian Vision Audition Touch Olfaction Taste
Pure
**
puur 3.9 2.4 2.6 3.6 4.2
3.9 2.8 2.4 1.7 1.8
Symbolic descriptors
Luxurious luxueus 4.7 2.8 3.7 3.1 2.9
4.9 2.9 3.3 3.0 2.9
Complex complex 4.9 2.9 3.4 1.9 2.1
4.6 3.1 3.2 2.1 2.6
Modern
**
modern 4.8 2.6 3.2 2.0 1.8
4.5 3.5 2.6 2.5 2.2
Interesting
*
interessant 4.7 3.5 3.9 3.1 3.0
4.6 4.0 3.6 3.2 3.2
Affective descriptors
Exciting
**
opwindend 4.3 2.9 3.8 3.3 3.2
3.8 3.1 3.1 3.4 2.6
Funny grappig 4.6 3.9 3.2 2.2 2.1
4.4 3.6 2.8 2.1 2.0
Cute
**
schattig 4.7 3.0 3.6 2.1 1.6
4.8 2.8 3.5 2.3 2.7
*
The effect of Modality Language interaction is signicant at 0.05 level.
**
The effect of Modality Language interaction is signicant at 0.01 level.
A. Fenko et al. / Journal of Pragmatics 42 (2010) 33143327 3323
high-pitch sound or pungent food as sharp. In Russian another word is used for characterizing a sharp sound ( ), so
audition has a low importance rating for the experience of sharp (1.6). On the other hand, gustatory associations with
sharpness are even stronger for the Russiansample than tactile associations (importance rating 4.3 for taste vs. 3.4 for touch).
This can be due to the fact that for the Russians associations were stronger with food than with tools like knives or scissors.
Vice versa, for Dutch respondents gustatory associations with pure (puur) were as strong as the visual associations
(importance rating 4.2 for taste vs. 3.9 for vision). Although the visual rating of pure ( ) was identical for the Russian
sample (3.9), taste had little importance for pure (1.8). The same was true for mild. The Dutch word mild showed the highest
importance rating for taste (4.4), but the Russian equivalent was mostly tactile (4.8). Apparantly, mild in Dutch is
similar to mild in English when applied to food: not sharp, strong, or hot in avour, not pungent, while in Russian is
more synonymous to soft and refers mostly to tactile properties.
When participants assess particular sensory modalities as important for judging a specic product attribute, they
probably imagine a product that has this attribute. These products may be considered as prototypical in a sense that they
rst come to mind in association with a certain attribute (Rosch, 1978; Mervis and Rosch, 1981). The fact that the
assessment of sensory modalities varies between languages may suggest that prototypical products may also vary cross-
linguistically. For example, knives andscissors may be prototypical sharp products for the Dutch, while spicy foods may be
prototypical for the Russians. Therefore, the language differences found in the present study may be related to the
categorization processes which were shown to differ between speakers of different languages (Schmitt and Zhang, 1998;
Zhang and Schmitt, 1998). Future research can evaluate whether prototypical products for specic adjectives differ
between languages.
The language differences we found in the present study might have an effect on how people experience products.
According to the linguistic relativity proposal (Whorf, 1956), speakers of different languages perceive and conceive the world
differently. Language may act as a lter through which people view reality in the process of perception, categorization and
the interpretation of information (Hunt and Agnoli, 1991). Differences in linguistic coding have been shown to correlate with
differences in non-linguistic conceptual coding (Levinson, 2003; Majid et al., 2004; Levinson and Wilkins, 2006), suggesting
that linguistic distinctions affect how we think (see also Lucy, 1992; Boroditsky, 2001). Lexical differences can impact
perception (Kay and Kempton, 1984; Davidoff et al., 1999) and cognition (Gordon, 2004). Further research is needed to
evaluate whether differences in product experiences exist between people speaking different languages, and whether any
such differences can be traced back to differences in sensory adjectives and sensory metaphors in different languages.
Some authors argue that each culture has its own mental frames (Hong et al., 2000), which are learned and used in
conjunction with that cultures language (Foucault, 1972). As a result, words in two different languages that may seemto be
exact translations of each other are likely to have different sets of culture-specic connotations (Kroll and De Groot, 1997),
reecting the differences in cultural frame content.
Because the interpretations of words vary among cultures and individuals, it is difcult to standardize verbal
communication with research participants from different cultures. Incompatibility of testing methods creates
methodological problems for interpreting cross-cultural differences. It is difcult to reach item equivalence, to make
sure that the instruments used in the research are similar, even with the most accurate translation and back translation
(Brislin, 1980).
In cross-cultural studies, it is extremely difcult to separate a cultural effect from a language effect, because people from
different cultures tend to speak different languages. This is why people who speak two languages (bilinguals) and people
who internalized the values, beliefs, and norms of two cultures (biculturals) have become the focus of research in
psycholinguistics, anthropology and consumer studies. Biculturals often report feeling like a different person when they
speak a different language (LaFromboise et al., 1993). This suggests that biculturals may possess two different culture-
specic mental frames for a single word in two different languages (translationequivalent words) (e.g., Hong et al., 2000).
Using bicultural bilingual participants can be a useful option in the future studies of cross-cultural and language differences
in product experience.
4.2. Cognitive and sensory modes of information processing
In this study we found more language differences for sensory descriptors of product experiences than for symbolic
descriptors. This result may be related to the differences between cognitive and sensory modes of information processing
(Hirschman, 1984; Pearson, 1970). These different modes initiate different mental and physical activities. Cognitive modes
of information processing refer to logical, rational, sequential thought processes and verbal modes of information processing
in contrast to holistic, gestalt, and visual modes of information processing (Childers et al., 1985). Cognitive experiences are
related to the need for being adaptively oriented to the environment and for achieving a sense of meaning, while sensory and
affective experiences are related to the need to achieve a feeling of satisfaction and to attain emotional goals (McGuire,
1976). We can assume that logical thinking and rationality is universal across cultures, while sensory experiences and
emotional goals will vary not only between social and cultural groups, but also between individuals. The latter suggestion
was supported by our nding that modality ratings for some sensory experiences demonstrated bi-normal distribution
patterns, even within a relatively uniform group of students of the same age and education level. Additional research is
needed into the associations people have with particular descriptions in various situations, in order to specify the exact
meaning of these adjectives for different products and different user groups.
A. Fenko et al. / Journal of Pragmatics 42 (2010) 33143327 3324
5. Conclusions
This research investigated the importance of sensory modalities for various descriptors of product experience. The results
showed that only for the affective descriptors of product experience (pleasantunpleasant, goodbad) all modalities were
about equally important. Symbolic descriptors (such as modern, expensive, or feminine) generally convey the social or
personal meaning of products. These descriptors tended to be multisensory, but most of them demonstrated visual
dominance. Sensory descriptors (such as colourful, loud, or soft) reect the perception of sensory information. In addition,
sensory meaning may transfer to other domains of experience. We found that sensory descriptors of product experience
showed signicant language differences. The latter result can be explained by the fact that metaphorical meanings of sensory
descriptors differ between languages.
Further research with a more broad variety of languages is needed to investigate the differences in sensory metaphors. In
cognitive linguistics, sensory metaphors (such as KNOWLEDGE IS VISION or UNDERSTANDING IS GRASPING) are viewed as
conceptual metaphors which are fundamental to language, thought, and experience (Lakoff and Johnson, 1980, 1999;
Sweetser, 1990). Whether such metaphors are universal or vary between languages and cultures is a question of signicant
importance for current cognitive linguistic theory. Most cognitive linguists assume that conceptual metaphors are explicit
abstract representations of embodied cognitive structures. According to Lakoff and Johnson (1999), conceptual metaphors
(such as POSITIVE IS UP, NEGATIVE IS DOWN) could be established as people implicitly learn associations between physical
experiences and emotional states that typically co-occur. Ko vecses (2005) distinguishes between universal metaphors that
are rooted in the universal bodily experience and conventional metaphors that vary across cultural, social, regional, style,
developmental and other dimensions. Some scholars suggest that conceptual metaphors are not merely representative of
universal body experience, but are tied to specic socio-cultural cognition (Kimmel, 2006; Zlatev, 2006). Metaphors could be
learned frompatterns in language and culture (Ibarretxe-Antuano, 2008); they are not necessarily embodied (Boroditsky,
2000; Gentner et al., 2001). Even if direct bodily experience is necessary on the timescale of biological or cultural/linguistic
evolution, it may not be necessary on the timescale of the conceptual development of an individual (Tomasello, 2003;
Vygotsky, 1986). Further research into the cultural differences in sensory metaphors might help to resolve this theoretical
argument.
As concerns practical implications, it should be noted that sensory adjectives are frequently used in advertising to
describe sensory and symbolic properties of products (Ruiz, 2006). Since most sensory adjectives have additional
metaphorical meanings which can differ between the languages, it is important for advertisers who want to use the same
slogans and product descriptions in countries with different languages, to be aware of the polysemy of sensory adjectives
and cross-cultural differences in their metaphorical meanings.
Acknowledgements
The Netherlands Organization for Scientic Research (N.W.O.) is gratefully acknowledged for funding this research
(MAGW OC Grant 400-03-131). We also thank Paul Hekkert and an anonymous journal reviewer for their comments and
suggestions that helped to improve the manuscript.
References
Asch, Solomon, 1955. On the use of metaphor in the description of persons. In: Werner, Heinz (Ed.), On Expressive Language. Clark University Press,
Worcester, MA, pp. 8694.
Boroditsky, Lera, 2000. Metaphoric structuring: understanding time through spatial metaphors. Cognition 75, 128.
Boroditsky, Lera, 2001. Does language shape thought? Mandarin and English speakers conceptions of time. Cognitive Psychology 43, 122.
Boroditsky, Lera, Schmidt, Lauren, Webb, Phillips, 2003. Sex, syntax, and semantics. In: Gentner, Dedre, Goldin-Meadow, Susan (Eds.), Language in Mind:
Advances in the Study of Language and Cognition. MIT Press, Cambridge, MA, pp. 6180.
Bowerman, Melissa, 1996. The origins of childrens spatial semantic categories: cognitive versus linguistic determinants. In: Gumperz, John, Levinson,
Stephen (Eds.), Rethinking Linguistic Relativity. Cambridge University Press, New York, pp. 145176.
Brislin, Richard, 1980. Translation and content analysis of oral and written material. In: Triandis, Harry, Berry, John (Eds.), Handbook of Cross-Cultural
Psychology, vol. 2. Allyn and Bacon, Boston, MA, pp. 389444.
Buck, Carl Darling, 1949. A Dictionary of Selected Synonyms in the Principal Indo-European Languages. Chicago University Press, Chicago.
Byrne, Richard, Whiten, Andrew (Eds.), 1988. Machiavellian Intelligence: Social Expertise and the Evolution of Intellect in Monkeys, Apes, and Humans.
Oxford University Press, New York.
Caplan, David, 1973. A note on the abstract readings of verbs of perception. Cognition 2, 269277.
Chen, Jenn-Yeu, 2007. Do Chinese and English speakers think about time differently? Failure of replicating Boroditsky (2001). Cognition 104, 427436.
Childers, Terry, Houston, Michael, Heckler, Susan, 1985. Measurement of individual differences in visual versus verbal information processing. Journal of
Consumer Research 12, 125134.
Classen, Constance, 1993. Worlds of Sense. Exploring the Senses in History and Across Cultures. Routledge, London.
Classen, Constance, 2005. McLuhan in the rainforest. In: Howes, David (Ed.), Empire of the Senses. Berg, Oxford, pp. 147163.
Davidoff, Jules, Davies, Ian, Robertson, Debi, 1999. Colour categories of a stone-age tribe. Nature 398, 203204.
Evans, Nicolas, Wilkins, David, 2000. In the minds ear: The semantic extensions of perception verbs in Australian languages. Language 76, 546592.
Feldman Barrett, Lisa, Russell, James, 1998. Independence and bipolarity in the structure of current affect. Journal of Personality and Social Psychology 74,
967984.
Fenko, Anna, Schifferstein, Hendrik, Huang, Tse-Chia, Hekkert, Paul, 2009. What makes products fresh: The smell or the colour? Food Quality and Preference
20, 372379.
Fiore, Ann Marie, Kimle, Patricia, 1997. Understanding Aesthetics for the Merchandising and Design Professional. Fairchild, New York.
Foucault, Michel, 1972. The Archaeology of Knowledge (Sheridan Smith, A.M., Trans.). Pantheon, New York.
A. Fenko et al. / Journal of Pragmatics 42 (2010) 33143327 3325
Gallup Jr., Gordon G., Cameron, Patricia, 1992. Modality specic metaphors: is our mental machinery colored by a visual bias? Metaphor and Symbolic
Activity 7, 9398.
Gelman, Rochel, Gallistel, Charles, 2004. Language and the origin of numerical concepts. Science 306, 441443.
Gentner, Dedre, Bowdle, Brian, Wolff, Phillip, Boronat, Consuelo, 2001. Metaphor is like analogy. In: Gentner, Dedre, Holyoak, Keith James, Kokinov, Boicho
N. (Eds.), The Analogical Mind: Perspectives from Cognitive Science. MIT Press, Cambridge, MA, pp. 199253.
Gentner, Dedre, Imai, Mutsumi, 1997. A cross-linguistic study of early word meaning: universal ontology and linguistic inuence. Cognition 62, 169200.
Gibbs Jr., Raymond W., 1994. The Poetics of Mind: Figurative Thought, Language, and Understanding. Cambridge University Press, Cambridge and NewYork.
Gilbert, Aubrey, Regier, Terry, Kay, Paul, Ivry, Richard, 2006. Whorf hypothesis is supported in the right visual eld but not the left. Proceedings of the
National Academy of Sciences of the United States of America 103, 489494.
Gordon, Peter, 2004. Numerical cognition without words: evidence from Amazonia. Science 306, 496.
Govers, Pascalle, 2004. Product Personality. Unpublished Ph.D. Dissertation. Delft University of Technology, Delft, The Netherlands.
Gumperz, John, Levinson, Stephen (Eds.), 1996. Rethinking Linguistic Relativity. Cambridge University Press, New York.
Gyoba, Jiro, Suzuki, Miho, Kawabata, Yamaguchi, Hiroshi, Komatsu, 2005. Analyses of sensory-relevance of adjective pairs frequently used in the semantic
differential method. Paper presented at the 6th Annual Meeting of the International Multisensory Research Forum, Rovereto, Italy, 58 June 2005.
Heider, Eleanor, 1972. Universals in color naming and memory. Journal of Experimental Psychology 93, 1020.
Hekkert, Paul, Schifferstein, Hendrik, 2008. Introducing product experience. In: Schifferstein, H.N.J., Hekkert, P. (Eds.), Product Experience. Elsevier,
Amsterdam, pp. 18.
Hirschman, Elizabeth, 1984. Experience seeking: a subjectivistic perspective for consumption. Journal of Business Research 12, 115136.
Hofstede, Geert, 2001. Cultures Consequences: Comparing Values, Behaviours, Institutions, and Organizations across Nations, 2nded. Sage, Thousand Oaks,
CA.
Hong, Ying-Yi, Morris, Michael, Chiu, Chi-Yue, Benet-Martnez, Vernica, 2000. Multicultural minds: a dynamic constructivist approach to culture and
cognition. American Psychologist 55, 709720.
Hunt, Earl, Agnoli, Franca, 1991. The Whoran hypothesis: a cognitive psychology perspective. Psychological Review 98, 377389.
Ibarretxe-Antuano, Iraide, 1999. Metaphorical mappings in the sense of smell. In: Gibbs, Raymond, Steen, Gerard (Eds.), Metaphor in Cognitive Linguistics.
John Benjamins, Amsterdam, pp. 2945.
Ibarretxe-Antuano, Iraide, 2008. Vision metaphors for the intellect: are they really cross-linguistic? Atlantis. Journal of the Spanish Association of Anglo-
American Studies 30, 1533.
Iordanskaja, Lidija, 1979. The semantics of three Russian verbs of perception. Linguistics 17, 825842.
January, David, Kako, Edward, 2007. Re-evaluating evidence for linguistic relativity: reply to Boroditsky (2001). Cognition 104, 417426.
Johansson, Antero, Strmnes, Frode, 1995. Cultural differences in occupational accidents. Part I: Theoretical background. Presented at the 44th Nordic
Meeting on Work Environment, Nantali, Finland.
Johansson, Antero, Salminen, Simo. 1996. Different languages: different information processing systems? Part I: Why can we expect differences in
occupational accidents between language groups. Presented at the International Symposiumon Work in the Information Society, Helsinki, Finland, 20
22 May 1996.
Kay, Paul, Kempton, Willett, 1984. What is the SapirWhorf hypothesis? American Anthropologist 86, 6579.
Kimmel, Michael, 2006. Culture regained: situated and compound image schemas. In: Hampe, Beate (Ed.), From Perception to Meaning: Image Schemas in
Cognitive Linguistics. Mouton de Gruyter, Berlin, pp. 285312.
Ko vecses, Zoltn, 2005. Metaphor in culture: Universality and variation. Cambridge University Press, New York.
Kroll, Judith, de Groot, Annette, 1997. Lexical and conceptual memory in the bilingual: mapping form to meaning in two languages. In: de Groot,
Annette, Kroll, Judith (Eds.), Tutorials in Bilingualism: Psycholinguistic Perspectives. Erlbaum, Mahwah, NJ, pp. 169199.
Kurath, Hans, 1921. The Semantic Sources of the Words for the Emotions in Sanskrit, Greek, Latin and the Germanic Languages. George Banta Publishing,
Menasha, WI.
LaFromboise, Teresa, Coleman, Hardin, Gerton, Jennifer, 1993. Psychological impact of biculturalism: evidence and theory. Psychological Bulletin 114, 395
412.
Lakoff, George, Johnson, Mark, 1980. Metaphors We Live By. University of Chicago Press, Chicago.
Lakoff, George, Johnson, Mark, 1999. Philosophy in the Flesh. Basic Books, New York.
Lehrer, Adrienne, 1978. Structures of the lexicon and transfer of meaning. Lingua 45, 95123.
Levinson, Stephen, 1996. Language and space. Annual Review of Anthropology 25, 353382.
Levinson, Stephen, 2003. Space in Language and Cognition: Explorations in Cognitive Diversity. Cambridge University Press, New York.
Levinson, Stephen, Jaisson, Pierre (Eds.), 2006. Evolution and Culture: A Fyssen Foundation Symposium. The MIT Press, Cambridge, MA.
Levinson, Stephen, Meira, Srgio, 2003. Natural concepts in the spatial topological domainadpositional meanings in cross-linguistic perspective: an
exercise in semantic typology. Language 79, 485516.
Levinson, Stephen, Wilkins, David (Eds.), 2006. Grammars of Space. Cambridge University Press, New York.
Li, Peggy, Gleitman, Lila, 2002. Turning the tables: language and spatial reasoning. Cognition 83, 265294.
Lucy, John, 1992. Language Diversity and Thought: A Reformulation of the Linguistic Relativity Hypothesis. Cambridge University Press, New York.
Lucy, John, Gaskins, Suzanne, 2001. Grammatical categories and the development of classication preferences: a comparative approach. In: Bowerman,
Melissa, Levinson, Stephen (Eds.), Language Acquisition and Conceptual Development. Cambridge University Press, New York, pp. 257283.
Majid, Asifa, Bowerman, Melissa, Kita, Sotaro, Haun, Daniel, Levinson, Stephen, 2004. Can language restructure cognition? The case for space. Trends in
Cognitive Sciences 8, 108114.
Malt, Barbara, Sloman, Steven, Gennari, Silvia, Wong, Yuan, 1999. Knowing versus naming: similarity and the linguistic categorization of artefacts. Journal of
Memory and Language 40, 230262.
McGuire, William, 1976. Some internal psychological factors inuencing consumer choice. Journal of Consumer Research 2, 302319.
McLuhan, Marshall, 1961. Inside the ve sense sensorium. The Canadian Architect 6, 4954.
Mervis, Carolyn, Rosch, Eleanor, 1981. Categorization of natural objects. Annual Review of Psychology 32, 89115.
Miller, George, Johnson-Laird, Philip, 1976. Language and Perception. Cambridge University Press, New York.
Nez, Rafael, Sweetser, Eve, 2006. Looking ahead to the past: convergent evidence fromAymara language and gesture in the crosslinguistic comparison of
spatial construals of time. Cognitive Science 30, 401450.
Osgood, Charles, Suci, George, Tannenbaum, Percy, 1957. The Measurement of Meaning. University of Illinois Press, Urbana, IL.
Pan, Yigang, Schmitt, Bernd, 1996. Language and brand attitudes: impact of script and sound matching in Chinese and English. Journal of Consumer
Psychology 5, 263277.
Pearson, Pamela, 1970. Relationships between global and specic measures of novelty seeking. Journal of Consulting and Clinical Psychology 34, 199204.
Pica, Pierre, Lemer, Cathy, Izard, Veronique, Dehaene, Stanislas, 2004. Exact and approximate arithmetic in an Amazonian indigene group. Science 306, 499
503.
Rein, Martin, Scho n, Donald, 1977. Problemsetting in policy research. In: Weiss, Carol (Ed.), Using Social Research in Public Policy Making. Lexington Books,
Lexington, MA, pp. 235251.
Robertson, Debi, Davies, Ian, Davidoff, Jules, 2000. Color categories are not universal: replications and new evidence from a stone-age culture. Journal of
Experimental Psychology: General 129, 369398.
Rosch, Eleanor, 1978. Principles of categorization. In: Rosch, Eleanor, Lloyd, Barbara (Eds.), Cognition and Categorization. Lawrence Erlbaum Associates,
Hillsdale, NJ, pp. 2748.
A. Fenko et al. / Journal of Pragmatics 42 (2010) 33143327 3326
Ruiz, Javier Herrero, 2006. The role of metaphor, metonymy, and conceptual blending in understanding advertisements: the case of drug-prevention ads.
Revista Alicantina de Estudios Ingleses 19, 169190.
Salminen, Simo, Hiltunen, Erkki, 1995. Cultural differences in occupational accidents. Part II: A case study of Finnish- and Swedish-speaking workers in
Finland. Presented at the 44th Nordic Meeting on Work Environment, Naantali, Finland.
Schifferstein, Hendrik, 2006. The relative importance of sensory modalities in product usage: a study of self-reports. Acta Psychologica 121, 4164.
Schifferstein, Hendrik, Otten, Jacco, Thoolen, Fien, Hekkert, Paul, 2010. An experimental approach to assess sensory dominance in a product development
context. Journal of Design Research 8 (2), 119144.
Schmitt, Bernd, Zhang, Shi, 1998. Language structure and categorization: a study of classiers in consumer cognition, judgment and choice. Journal of
Consumer Research 25, 108122.
Schmitt, Bernd, Pan, Yigang, Tavassoli, Nader, 1994. Language and consumer memory: the impact of linguistic differences between Chinese and English.
Journal of Consumer Research 21, 419431.
Seeger, Anthony, 1975. The meaning of body ornaments: a Suya example. Ethnology 14, 211224.
Simpson, John, Weiner, Edmund (Eds.), 1989. The Oxford English Dictionary. second ed. Oxford University Press, New York.
Stadtlander, Lee, Murdoch, Latona, 2000. Frequency of occurrence and rankings for touch-related adjectives. Behavior Research Methods, Instruments, and
Computers 32, 579587.
Stein, Barry, Meredith, Alex, 1993. The Merging of the Senses. The MIT Press, Cambridge, MA.
Strmnes, Frode, 1974. No universality of cognitive structures? Two experiments with almost-perfect one-trial learning of translatable operators in a Ural-
Altaic and IndoEuropean language. Scandinavian Journal of Psychology 15, 300309.
Strmnes, Frode, 1973. A semiotic theory of imagery processes with experiments on an Indo-European and a Ural-Altaic language: do speakers of different
languages experience different cognitive worlds? Scandinavian Journal of Psychology 14, 291304.
Suzuki, Miho, Gyoba, Jiro, 2001. Contrastive analysis of sensory-relevance of factors affecting aesthetic impressions. Tohoku Psychologica Folia 60, 8192.
Suzuki, Miho, Gyoba, Jiro, Sakuta, Yuiko, 2005. Multichannel NIRS analysis of brain activity during semantic differential rating of drawing stimuli containing
different affective polarities. Neuroscience Letters 375, 5358.
Sweetser, Eve, 1990. From Etymology to Pragmatics: Metaphorical and Cultural Aspects of Semantic Structure. Cambridge University Press, New York.
Tomasello, Michael, 2003. Constructing a Language: A Usage-based Theory of Acquisition. Harvard University Press, Cambridge, MA.
Ullman, Stephen, 1957. The principles of semantics, 2nd ed. Blackwell, Oxford, UK.
Van Essen, David, Felleman, Daniel, DeYoe, Edgar, Olavarria, Jaime, Knierim, James, 1990. Modular and hierarchical organization of extrastriate visual cortex
in the Macaque monkey. Cold Spring Harbor Symposia on Quantitative Biology 55, 679696.
Viberg, Ake, 1984. The verbs of perception: a typological study. In: Butterworth, Brian, Comrie, Bernard, Dahl, O?sten (Eds.), Explanations for Language
Universals. Mouton, Berlin, pp. 123162.
Vygotsky, Lev Semenovich, 1986. Thought and Language. MIT Press, Cambridge, MA.
Whorf, Benjamin Lee, 1956. In: Carroll, John B. (Ed.), Language, Thought and Reality: Selected Writings of Benjamin Lee Whorf. MIT Press, Cambridge, MA.
Williams, Joseph, 1976. Synaesthetic adjectives: a possible law of semantic change. Language 52, 461478.
Zhang, Shi, Schmitt, Bernd, 1998. Language-dependent classication: the role of classiers in cognition, memory and ad evaluation. Journal of Experimental
Psychology: Applied 4, 375385.
Zlatev, Jordan, 2006. Whats in an image schema? Bodily mimesis and the grounding of language. In: Hampe, Beate (Ed.), From Perception to Meaning:
Image Schemas in Cognitive Linguistics. Mouton de Gruyter, Berlin, pp. 313342.
Anna Fenko is a PhDcandidate at the Department of Industrial Design of Delft University of Technology. She has an MSc degree in psychology fromMoscowState
University (Russia). She worked in the areas of consumer psychology and marketing. In 20032004 she was a Fulbright visiting scholar at Jones School of
Management at Rice University, Houston, USA. She presented her work at various international conferences and published papers in Applied Ergonomics, Food
Quality and Preference and Materials and Design.
Jacco J. Otten has an MSc degree in Industrial Design Engineering fromDelft University of Technology. He has worked as a designer and researcher on various user
interaction topics. While user interaction is often mainly vision-based, he had the opportunity to go beyond this single dimension by performing multisensory
design and research. Currently he works as lead user interaction designer at Mavim, an innovative Dutch software development company.
Dr. Hendrik N.J. Schifferstein is Associate Professor at the Department of Industrial Design of Delft University of Technology. After having worked in the food
realm for several years, he now studies the multisensory experiences evoked by consumer durables. Among others, he published in Perception and Psychophysics,
Acta Psychologica, Marketing Letters, and Journal of Experimental Psychology: Human Perception and Performance. He is co-editor of the books Food, People and Society
(2001; Springer-Verlag) and Product Experience (2008; Elsevier).
A. Fenko et al. / Journal of Pragmatics 42 (2010) 33143327 3327