It sounds like a bright color : the effects of package color saturation and brand sound on perceived product healthiness

(1)

1

It sounds like a bright color

The effects of package color saturation and brand sound on

perceived product healthiness

Study: MSc Business Administration – Marketing Track

Institution: Amsterdam Business School / University of Amsterdam Course: Master Thesis

Thesis Supervisor: Antoon Meulemans Student: Jaap de Wit

Student number: 10894578 Date of submission: 26-01-2017 Word count: 11.136

(2)

(3)

3

Statement of originality

This document is written by J.C. De Wit who declares to take full responsibility

of the contents of this document.

I declare that the text and the work presented in this document is original and

that no sources other than those mentioned in the text and its references have

been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision

of completion of the work, not for the contents.

(4)

(5)

5

2.8 Health interest ... 21 2.9 Conceptual model ... 21 3 Method ... 22 3.1 Brand sound ... 22 3.2 Color saturation ... 23 3.3 Product category ... 24 3.4 Sample ... 25 3.5 Measures ... 25 4 Analyses ... 27 4.1 Recoding variables ... 27 4.2 Reliability analysis ... 27 4.3 Correlation matrix ... 28 4.4 Hypothesis testing ... 30 5 Results ... 37 6 General discussion ... 38

7 Limitations and further research ... 40

(6)

(7)

7

Abstract

Consumers are demanding healthier food options and brands that offer health claims are flourishing. In response, food manufacturers are increasingly introducing new ‘healthy’ brands or reposition established brands to gain a healthier image. To do this, managers can use sensory stimuli to establish a healthy image in the mind of the consumer. The effects of certain colours, sounds and shapes on consumer perceptions are called color-, sound-, and shape symbolism. These symbolism effects are particularly important because these sensory triggers are usually the first cues for consumers on which they base their product judgments. One aspect of sensory marketing that has gained a lot of attention over the last years, is the effect of cross-modal correspondences. These crossmodal correspondences can be defined as a sensory feature, or attribute, that matches another sensory attribute. For instance, there exists a crossmodal correspondence between larger objects and lower-pitched sounds, and smaller objects and higher-pitched sound. Consumers link these different sensory attributes naturally to each other.

In this study, we found an effect between brand sound (heavy vs. light), package color saturation (high vs. low), product type (utilitarian vs. hedonic), general health interest (high vs. low) on perceived product healthiness, taste expectation, purchase intention and product evaluation. The results showed that there exists a crossmodal correspondence between brand sound and package color saturation. More specifically, product evaluation and purchase intention are higher if brand sound and package color are congruent. The interaction between Brand Sound and Package Color shows that healthy sounding products in combination with a healthy appearance (low saturated colors) are better evaluated, are expected to taste better and respondents are more likely to buy them. However, such an effect only occurs when people are relying relatively more on heuristics. This means that consumers that have a low general health interest would prefer a product called Ramune (healthy sound) in a low saturated package (healthy appearance) and a product called Asashi (unhealthy sound) in a high saturated package (unhealthy appearance) over all other incongruent combinations. In contrast, people with a relatively high general health interest would prefer a product with a low saturated package irrespective the extent to which a brand name sounds healthy. These findings are important for product managers who want to use sensory triggers to evoke certain associations in the mind of the consumer.

(8)

(9)

9

1. Introduction

1.1 Background

While there seem to be more people suffering from obesity throughout the world, healthy food is gaining popularity. In the last decade, there seems to be a growing trend of health-conscious consumption. Consumers are demanding healthier food options and brands that offer health claims are flourishing. In response, food manufacturers are increasingly introducing new ‘healthy’ brands or reposition established brands to gain a healthier image.

As consumers buy products and brands that align with their physical, ethical and moral viewpoints, the market for healthy products and brands is estimated to reach $1 trillion by 2017. While the health fads and trends have come and gone from time to time, the category appears to have serious stamina these days. The consumer mindset about healthy food has shifted and people are ready to pay more for products that claim to boost health and weight loss (Gagliardi, 2015).

With a fast growing demand for healthy products, companies need to understand how product features and attributes can facilitate a healthy perception in the mind of the consumer. Perceptions are shaped by experiences and the latter are, in turn, shaped (mostly) by our senses. Most of our everyday experiences are multisensory. ‘A consumer’s brand and product experiences are no exceptions, as many sensory marketers are increasingly coming to realize’ (Spence, 2012).

There are five senses which can affect consumer perceptions: haptics, taste, vision, audition and smell. Sensory triggers aimed at one are more of these senses may result in consumers’ self-generation of (desirable) brand attributes, rather than those verbally provided by the advertiser (Krishna, 2011). This means that consumers give their own interpretation to everything they encounter. For example, previous studies have found effects of certain colours, sounds and shapes on consumer perceptions. It is found, for instance, that low saturated colors, light sounding brands and angular packages are perceived to represent healthier products than high saturated colors, heavy sounding brands and round packages (Becker, van Rompay, Schifferstein, & Galetzka, 2011) (Fenko, Lotterman, & Galetzka, 2016).

The effects of certain colours, sounds and shapes on consumer perceptions are called color-, sound-, and shape symbolism. These effects are particularly important because these sensory triggers are usually the first cues for consumers on which they base their product judgments. This is because haptics and taste often involve actual consumption which is normally after the purchase of a

(10)

10 particular product. Color-, sound-, and shape symbolism can also reinforce each other to boost certain perceptions in the mind of the consumer. This is called ‘crossmodal correspondence’.

Crossmodal correspondences gain increasingly attention from researchers. ‘One aspect of multisensory perception that has started to gain increasing importance over the last couple of years relates to the topic of crossmodal correspondences’ (Spence, 2012). ‘Generally-speaking, crossmodal correspondences can be defined as a tendency for a feature, or attribute, in one sensory modality to be matched (or associated) with a sensory feature, or attribute in another sensory modality’ (Spence, 2012). For instance, there exists a crossmodal correspondence between larger objects and lower-pitched sounds, and smaller objects and higher-lower-pitched sounds (Walker & Smith, 1985) (Parise & Spence, 2009). Crossmodal correspondences make sense to consumers and are therefore easy to interpret and process.

As health consciousness increases over the years and manufacturers continue to introduce new healthy brands, it would be interesting to know how crossmodal correspondences can facilitate a healthy perception in the mind of the consumer.

1.2 Research question and sub questions

It seems that shape-, sound-, and color symbolism are important determinants of a (new) brand’s success. In the light of the growing popularity of healthy brands it is interesting to know how name, shape and color can enhance and leverage the perceived healthiness in the mind of the consumer. It is already known that angular shapes are perceived as healthier in contrast to circular shapes. Moreover, congruence effects between brand name and shape also have a positive impact on a (healthy) product’s evaluation. However, the effects of color symbolism remains unclear as contradictive evidence has been found. Furthermore, the relationship between sound and color symbolism has, to my knowledge, not been studied before. Therefore, I propose the following research question:

Does a crossmodal correspondence between sound symbolism and package color exist and how does it affect the perceived healthiness of a particular product?

In order to answer this research question, the following sub questions are proposed: - How does this field relate to branding?

- What are crossmodal correspondences?

(11)

11 - What is color symbolism and how does it affect perceived product healthiness?

- What is sound symbolism and how does it affect perceived product healthiness? - What is the combined effect of color and sound symbolism on perceived product healthiness?

1.3 Scientific relevance

The field of sensory marketing is not new, however, there is still a lot of research to be done. ‘While a lot of work has been done on sensory marketing in the last two decades as we have seen, there is still need for additional research on many aspects of sensory marketing’ (Krishna, 2011). In her review of sensory marketing, Krishna underlines that there is room for further research on sensory congruence. Similarly, we have seen already that contradictive evidence has been found on the effect of color saturation on consumer perceptions. This leaves a scientific gap in the literature, namely: does color saturation affect consumer perceptions and under what conditions? With this study we try to answer this question and add other variables to find interaction effects that have not been studied or found before. By combining color saturation and brand sound in one study, we aim to shed light on the possible congruence effect between those to variables. This is in line with the need for further research described by Krishna (2011).

(12)

(13)

13

2. Theoretical framework

This study investigates the relationship between sensory aspects of product features on the perceived healthiness of particular brands. Consumers’ senses are often triggered by primary brand elements as these are the first things a consumer is exposed to when they are shopping. These primary brand elements are important in shaping consumers’ perceptions about particular products or brands. Thus, in order to leverage the brand equity of brands that want to be perceived as more healthy, the primary brand elements should elicit the right responses in order to built the desired brand equity.

Brands are important for both marketers and consumers. ’Brands serve several valuable functions. At their most basic level, brands serve as markers for the offerings of a firm. For customers, brands can simplify choice, promise a particular quality level, reduce risk and/or engender trust. Brands are built on the product itself, the accompanying marketing activity, and the use (or non-use) by customers as well as others. Brands thus reflect the complete experience that customers have with products. Brands also play an important role in determining the effectiveness of marketing efforts such as advertising and channel placement. Finally, brands are an asset in the financial sense’ (Keller & Lehmann, 2006).

2.1 The customer-based brand equity model

One author that is an expert in branding and building brand equity is Keller. ’To build brand equity, companies must start with the basics’ (Keller, 2001). Keller developed a comprehensive approach, the customer-based brand equity (CBBE) model, in order to build a strong brand. He thinks of building a brand as a series of steps. The first step is establishing the proper brand identity in customers’ minds with a specific product class or customer need. This is reflected in the extent to which a product or brand is easily recalled by consumers (brand awareness). The second step resolves around creating the appropriate brand meaning by strategically linking a host of tangible and intangible brand associations. Third, eliciting the right brand responses to the brand identity and meaning (responses could be either positive or negative). And the fourth step is forging appropriate brand relationships with customers.

The CBBE model or ’branding ladder’ is shown in figure 1. ’There is an obvious sequence in this branding ladder, that is, meaning cannot be established unless identity has been created; responses cannot occur unless the right meaning has been developed; and a relationship cannot be forged unless the proper responses have been elicited’ (Keller, 2001). This structure depends on all brand related contacts, whether marketer initiated or not (Keller, 2005). This indicates that early branding

(14)

14 decisions considering your brand identity can have a significant impact on future product success. Therefore, crafting the right brand identity from the beginning is very important. Primary brand elements are important drivers of a product’s brand identity and will ultimately drive your brand equity as we will see in the next section.

Figure 1: the customer-based brand equity model consists of 4 sequential steps (Keller K. L., 2012).

2.2 Primary brand elements

According to Keller (2005) there are three main sets of brand equity drivers: (1) the choices for the brand elements or identities, (2) the product and service and all accompanying marketing activities and programs, and (3) other associations marketers indirectly transfer to the brand by linking it to another person, place or thing. While the second set of brand equity drivers, marketing programs, is the most important in shaping consumer perceptions, the first and the third set can be of critical importance. That is, these aspects of branding are often less expensive and can, if managed properly, strengthen your brand immensely. In the light of this research, the focus will be on evaluating and choosing the right brand elements.

‘Brands identify and differentiate a company’s offering to customers and other parties. A brand is more than a name. Other brand elements such as logos and symbols (Nike’s swoosh and McDonalds’ golden arches), packaging (Coke’s contour bottle and Kodak’s yellow and black film box), and slogans (BMW’s ‘’ultimate driving machine’’ and VISA’s ‘’It’s everywhere you want to be’’) play an important role in branding as well’, according to Keller & Lehmann (2006). These are the so called ‘primary brand elements’ that can transfer specific thoughts and feelings and elicit particular responses. Often these responses are subconscious which means that the consumer is not aware of the mental processing that is going on in his or her head. Choosing brand elements that contain the message a

(15)

15 brand wants to tell help to shape the brand in order to elicit the right responses and built brand equity.

Marketers must think about each separate brand element to decide whether it elicits the right responses. ‘’A number of broad criteria are useful for choosing and designing brand elements to build brand equity (Keller, 2005): (1) Memorability, (2) meaningfulness, (3) likeability, (4) transferability (both within and across product categories and across geographical and cultural boundaries and market segments), (5) adaptability and flexibility over time, and (6) legal and competitive protectability and defensibility’’ (Keller & Lehmann, 2006). The first three of these criteria are particularly important in shaping consumers’ perceptions while the last three are more of practical importance.

2.3 Crossmodal correspondences

As we have seen, brand elements must be memorable, meaningful and likeable in order to evoke (positive) perceptions in the consumers’ minds. Brand elements and especially primary brand elements trigger consumers’ senses but primary brand elements can also strengthen one another. If this is the case, the different brand elements are congruent and show a crossmodal correspondence. This means that consumers perceive the states of different brand elements as belonging naturally to each other. ‘Generally-speaking, crossmodal correspondences can be defined as a tendency for a feature, or attribute, in one sensory modality to be matched (or associated) with a sensory feature, or attribute in another sensory modality’ (Spence, 2012). Spence (2012) also describes that people are more positive about products that exist of congruent brand elements (vs. incongruent brand elements). He refers to this as the ‘need for congruence’.

Exposure to one of the matching stimuli will evoke in the other matching but unrelated stimulus. ‘Generally, three different mechanisms have been postulated to account for crossmodal correspondences. First of all, they may have an innate, biological origin. Humans may directly perceive the deeper, synesthetic quality of stimuli. Second, correspondences may develop through perceptual learning. For example, people may have learned that the smell of caramel usually coincides with its brown colour. Third, some correspondences may have a semantic origin: an odour may activate a certain association, and this may activate the image of a certain colour. Even when people experience difficulty in identifying the smell of a banana, this odour may elicit the association with fruit or, even more general, with food’ (Schifferstein & Tanudjaja, 2004).

Numerous crossmodal correspondences have been described in literature. It has been shown for instance that if scent and music are congruent, consumer approach behaviors are improved (Mattila & Wirtz, 2001). Similarly, if Christmas songs are played throughout a retail store during the holiday

(16)

16 season, sales for Christmas related products significantly increases as well as store evaluations (Spangenberg, Crowley, & Henderson, 2005). Previous results indicate that the field of sensory marketing is fairly broad. However, in a branding context, most of the sensory congruence effects resolve around shapes, sounds and colors. For marketers these sensory aspects are also the most easy to influence. In the following sections we will discuss shape, color and sound symbolism respectively.

2.4 Shape symbolism

A variety of stimuli are important when consumers are exposed to a (new) product. Attitudes towards a particular product are formed within a split second. This is because consumers immediately give meaning to the products they see. Product shape is, for instance, one of the first stimuli a person is encountered with. Different product shapes can elicit different emotions and perceptions regarding the actual product or brand. It is shown, for instance, that angular product shapes are perceived to reflect intense taste sensations, thereby supporting the idea of cross-modal correspondence (Becker, van Rompay, Schifferstein, & Galetzka, 2011). This is referred to as: shape symbolism.

Shape symbolism refers to the similar crossmodal mapping that exists between abstract shapes and

other sensory attributes (e.g., as between sharp pointy shapes and bitterness or carbonation in foods and beverages)’ (Spence, 2012). Moreover, angular products (as opposed to circular product types) are perceived as more healthy. ‘Package designs can look like human bodies that are associated with a healthy or unhealthy person. Designs that are taller than they are wide are more associated with a healthy body and therefore can influence perceived product healthfulness’ (Fenko, Lotterman, & Galetzka, 2016).

Next to perceived product healthfulness, package shape also shows a correspondence with taste. For instance, people tend to match carbonated water with different product shapes than still water. This means in practice that, more angular shapes have a better match with carbonated water versus more rounded shapes for still water (Spence & Gallace, Tasting shapes and words, 2010). Product shapes and names can also be combined in order to convey information about the taste (sweetness and sourness) of a product (Velasco, 2013). Velasco (2013) showed that ‘’sweet’’ tastes are better expressed by means of rounded shapes, typefaces, and names, and low pitched-sounds, whereas ‘’sour tastes’’ are better conveyed by means of angular shapes, typefaces, and names, and high-pitched sounds. These findings underline the results of Fenko and colleagues (2016) who showed that angular shapes are perceived to contain healthier products. This is because these products tend to be less sweet and therefore are less associated with round shapes.

(17)

17 This study investigates the cross modal relationship between color and sound symbolism on perceived product healthfulness. Therefore, this study will keep product shape as a constant as it is proven by various studies that products with angular shapes are indeed perceived as more healthy.

2.5 Color symbolism

Another stimulus on which consumers can draw inferences regarding particular product attributes is color. For instance, in one study it was predicted that packaging color would have the same effect on product evaluations regarding healthiness and taste impressions as particular shapes have. However, no interaction between particular colors and consumer perceptions was found in that particular study (Becker, van Rompay, Schifferstein, & Galetzka, 2011).

That does not mean that color is not important in marketing. Color can indeed play an important role in shaping consumer perceptions. ‘For savvy marketers, color is more than a graphic artist’s concern. It’s an integral aspect of a cohesive marketing plan. Color is a valuable messaging tool that cements a brand in consumers’ visual memories and can even elicit emotions or physical reactions. Color speaks volumes about who your company is and what you’re marketing. At the very least, color can make a brand more cohesive and a product more pleasing. And at best, color can make your product stand out from the rest’ (Sullivan, 2015).

Hoegg and Alba (2007) found that color manipulations do have a strong effect on product ratings. These results contrast with the results from Becker, et al. (2011) who did not find such an effect. In their study, Hoegg and Alba (2007) studied consumer responses to different packages of orange juice. They manipulated the color intensity of the packaging. The results demonstrated that the experienced ‘sweetness’ of the orange juice was indeed related to the color intensity of the packaging. Meaning that consumers perceive more intense colors as an indication of a sweeter product (in this case orange juice) even if the juice is exactly the same. Thus, according to this study, color can indeed have the ability to influence taste experiences. However, another study revealed quite the opposite results. Becker, et al. (2011) researched the effects of color saturation on perceived product sweetness and overall consumer attitude towards the product. They used yoghurt in their study and they manipulated the color saturation (high vs. low saturation). They proposed that consumers perceive product packages with harsh, saturated, shouting colors as containing more artificial ingredients. Consumers could believe that saturated colors indicate a lot of added sugar extract which lead them to experience a sweeter (and unhealthier) product and therefore affecting the taste experience. However, the main effects of packaging color on the taste experience did not reach significance in that particular study. This means that consumers did not taste any (significant) difference in taste between products in packages with highly saturated and low saturated colors. As

(18)

18 for product attitude, packaging color (saturation) did not influence the participants’ product attitudes (Becker, et al. 2011).

The effects of color symbolism in current literature, thus, remains contradictory. These studies, however, let the participants actually taste the products. This extra sensory aspect (taste) is very likely to result in different results than when respondents are only expected to look at a fictional product. That is, consumers often rely on heuristics prior to purchase. Therefore, this study will only show different packages to respondents and exclude the actual tasting. It is expected that excluding actual consumption will strengthen the effect. That is, more intense package colors will lead people to perceive the product as sweeter, more artificial and unhealthier than products with less intense package colors. The following is hypothesized:

H1: More saturated package colors will lead to stronger and more intense perceived (artificial) taste sensations and lower perceived healthiness.

2.6 Sound symbolism

Sound as a sense can have a significant effect on consumers in a variety of ways. This is often referred to as sound symbolism. ‘When we hear the sound of a word, we attach meaning to it, even perceiving physical features for the source of the sound, be it animate (human, dog, cat) or inanimate (box, robot, car, ice cream). For instance, a high frequency bark/yelp is associated with a small dog whereas a deep, low frequency growl with a large, ferocious looking one’ (Krishna, 2011). Sound symbolism is the name given to the association that people experience between specific sounds (including speech sounds) and particular stimulus attributes (e.g., as when they associate words containing the ‘i’ sound with smallness). It is shown, for instance, that the sounds of the consumption of food (e.g., chips and crackers) is affecting product evaluations (e.g., quality and freshness), which is a typical example of sound symbolism (Zampini & Spence, 2005).

We have already discussed the possible effects of product shape and color on consumer perceptions. Next to package shape and package color, the brand name is one of the first stimuli a consumer faces when he or she encounters a product. The sound of a brand name has the power to influence the attitude of a consumer. For example, Yorkston and Menon (2004) researched the effect of sound symbolism on the perceived creaminess of ice cream. They found for instance that ice cream called ‘Frosh’ is perceived to be creamier than ‘Frish’ (Yorkston & Menon, 2004). Fenko, Lotterman & Galetza (2016) also focussed on the reaction of people when being exposed to an unknown brand. Sound symbolism has been researched for many years. According to Fenko, Lotterman & Galetza (2016), Köhler investigated already in the year 1929 cross-modal correspondences between

(19)

19 meaningless words and particular shapes. He found out that people link more rounded shapes with ‘heavy’ sounding words like ‘maluma’ and more edgy shapes with words that sound more ‘light’ like ‘takate’. In the light of these findings, Fenko, et al. (2016) linked an unknown brand name (soft vs. sharp) to a particular package shape (round vs. angular) and found different congruence effects between brand name and package design. For instance, products with sharp sounding brand names and angular package designs are rated more positively than when there is no congruence between those two variables according to Fenko, et al. (2016). The need for congruence is also underlined by Spence (2012), who argues that marketers can enhance their consumers’ product experiences by ensuring that the brand name, the shape of the packaging and product-related sensory expectations like taste, are all in line within the mind of the consumer.

Based on the research of Köhler on crossmodal correspondences which has been mentioned by Fenko, et al. (2016), it is expected that people link soft sounded brands to unhealthy products. It is expected that the mechanism for sound symbolism works exactly the same as for shape symbolism. The following hypothesis is proposed:

H2: The ‘heavier’ the sound of a brand name, the lower the perceived healthiness of the product.

2.7 Product category

When making purchase decisions, consumers are always evaluating products on hedonic or more functional traits. In the light of growing health interest, this is particularly interesting. An increasing number of research indicates a health paradox or ‘unhealthy = tasty intuition’ (Raghunathan, Naylor, & Hoyer, 2006). This ‘unhealthy = tasty intuition’ of Raghunathan and collegues resolves around the fact that when information pertaining to the assessment of the healthiness of food items is provided, the less healthy the item is portrayed to be, (1) the better is its inferred taste, (2) the more it is enjoyed during actual consumption, and (3) the greater is the preference for it in choice tasks when a hedonic goal is more (versus less) salient (Raghunathan, et al. 2006).

However, it has been found that the ‘unhealthy = taste intuition’ is not universal (Werle, Trendel, & Ardito, 2013). Werle and colleagues argued that differences between food culture between France and the United States would lead to different results for the perceived tastiness and healthfulness of food. In fact, the authors found that the opposite intuition exists in France. ‘’Unhealthy food is spontaneously associated with bad taste, while healthy food is linked to tastiness’’ (Werle, et al. 2013). According to these results, it seems that product category can have different effects on perceived tastefulness of food. Therefore, it might also affect purchase intention and product evaluation. However, on average, we expect that products representing a more hedonic goal will be perceived as less healthy than products representing a more utilitarian goal.

(20)

20

H3: Products representing a more hedonic goal (versus less) will be perceived as less healthy.

2.6 Congruence effect between color, sound and product category

This study focuses on crossmodal correspondences and particularly the correspondence between package color and the sound of a brand. Both color and sound, in isolation, can affect taste sensations and influence product expectations. However, it remains unclear whether there exists a congruence effect between the color of a package and the sound of a brand. The need for congruence is underlined by Spence (2012). He argues that all sensory stimuli a consumer encounters should reinforce each other. If all the stimuli are in line with each other, the product will be rated more positively, than when certain stimuli are not balanced.

For example, the congruent combination of product shape and brand name (round ‘’ramune’’ cookie) was preferred to incongruent combinations. Furthermore, the congruent combinations (‘’Ramune’’ butter cookie and ‘’Asashi’’ muesli cookie) are expected to taste better and are more likely to be purchased compared to incongruent combinations as been described by Fenko, et al. (2016).

It is expected that color saturation can affect perceived product healthfulness. Lower saturation is thought to be linked to natural flavours as opposed to highly saturated colors (artificial flavours). Building on the study of Fenko, et al. (2016) it could be expected that more saturated colors are congruent with the ‘unhealthy’ combination of round shaped products which sound ‘heavy’, ‘dark’ and ‘fat’. In contrast, colors with low saturated are expected to be congruent with angular shaped products which sound ‘slim’, ‘light’ and ‘soft’.

These congruent combinations of product category, color and sound should lead to higher product evaluations. The healthy congruent variant should, of course, lead to a product that is perceived as more healthful than an incongruent combination. Therefore, the following hypothesis is proposed:

H4a: Congruence between color and sound will lead to (a) better perceived healthfulness and (b) better perceived taste, (c) better product evaluation and, (d) greater purchase intention than incongruent combinations.

H4b: Congruence between sound and product category will lead to (a) better perceived healthfulness and (b) better perceived taste, (c) better product evaluation and, (d) greater purchase intention than incongruent combinations.

H4c: Congruence between product category and color saturation will lead to (a) better perceived healthfulness and (b) better perceived taste, (c) better product evaluation and, (d) greater purchase intention than incongruent combinations.

(21)

21

H4d: The combined congruence effect of color saturation, brand sound and product category will be greater than the effects described in H4a, b and c.

2.8 Health interest

The previously mentioned stimuli are all peripheral cues. These peripheral cues, however, are only dominant if a consumer is not processing information in the central route, thus, when a person is low in his need for cognition (Cacioppo, Petty, Kao, & Rodriquez, 1983). ‘One of the major sources of variance in attitude research is that attributable to individual differences among subjects. In particular, it is reasoned that there are stable individual differences in intrinsic motivation to engage in effortful cognitive endeavours generally, just as there are stable individual differences in intrinsic motivation to engage in effortful physical endeavours’ (Cacioppo, et al. 1983). This means that people with no interest in healthy food will be influenced more by peripheral cues as color, sound and shape. In contrast, such stimuli would have less impact on people with a great interest in a healthy lifestyle.

H5: General Health Interest will influence the effects described in H1 till H4d.

2.9 Conceptual model

(22)

22

3 Method

This quantitative research was carried out using a survey design. The quantitative origin is due to the deductive approach of this research. Hypotheses were tested using a survey which makes findings generalisable to the whole population. The conceptual model consists out of three variables which can vary in two ways and one moderator. This resulted in a 2x2x2 research design or 8 different conditions (see table 1). Respondents were exposed to only one condition (between subjects design) in order to make the questionnaire (see table 2) as short as possible. This was expected to increase the overall response rate and reduce any bias due to a ‘learning effect’ which would be the case in a within subjects design. The following conditions were proposed:

Table 1

Research conditions in a 2x2x2 research design

Color saturation

Brand

sound

Hedonic product

High saturation/ heavy brand sound Low saturation/ heavy brand sound High saturation/ light brand sound Low saturation/ light brand sound

Utilitarian product

High saturation/ heavy brand sound Low saturation/ heavy brand sound High saturation/ light brand sound Low saturation/ light brand sound

The conditions used in this research consist of variables that were pre-tested in other studies. These pre-tests are described below. These pre-tests were conducted in a proper manner and are not expected to bias the results in this study.

3.1 Brand sound

’A pre-study was conducted to select brand names that are associated with health. Participants (N=125; 80 females; aged from 18 to 65; mean age 37) evaluated 12 unfamiliar names of food products on 15 bi-polar 5-point scales (including healthy/unhealthy, warm/cold, dark/light, good/bad, light/heavy, slim/fat, etc.). The product names unfamiliar in Dutch language were selected from Japan, Indonesia, Marocco, Norway, Sweden, Italy and the United States via online search. The results showed that the words ‘Asashi’ and Tahini’ were perceived as significantly softer, lighter, brighter, slimmer and healthier than other names, while ‘Ramune’, ‘Kex’ and ‘Manner’ were perceived as harder, heavier, darker and fatter. ‘Asashi’ was selected as the name with healthy associations because some participants were familiar with the name ‘Tahini’. ‘Ramune’ was selected as the name with unhealthy associations, because it scored low on healthiness and high on heavy and fat. ‘Ramune’ and ‘Asashi’ did not significantly differ on attractiveness and good/bad scales’ (Fenko, Lotterman, & Galetzka, 2016).

(23)

23

3.2 Color saturation

The second pre-test was conducted to find an optimal perceived difference in color saturation between low saturated and a high saturated variant. ’20 participants (8 male, 12 female; mean age 32,9 years) rated two saturation variants of two lemon-greenish color pairs (color pair 1: hue 62, saturation 100, brightness 95 versus hue 62, saturation 50, brightness 95; color pair 2: hue 68, saturation 100, brightness 95 versus hue 68, saturation 50, brightness 95), (Becker, et al. 2011). ’Analysis of the means showed that within the first color pair, the saturation variants differentiated most clearly on the potency construct (color 1: Mlow saturation = 3.07, SD = .78 versus Mhighsaturation =

5.06, SD = .57; color 2: Mlowsaturation = 3.42, SD = .80 versus Mhighsaturation = 5.21, SD = .58). An analysis

of variance confirmed that this difference is significant (F(1,18) = 85.18, p < .01), indicating that the 50% saturated color is rated as less potent than the 100% saturated color’ (Becker, et al. 2011).

(24)

24

3.3 Product category

This study makes a distinction between utilitarian and hedonic product categories. We have chosen chocolate bars as representing a hedonic goal and a muesli bar for representing a more utilitarian goal. We have chosen these two products because we expect that chocolate bars commonly known as an unhealthy snack serving hedonic goals. In contrast, we have chosen muesli bars to represent healthy products. In figure 3-6 the different conditions are visualized. Respondents were shown one of the visualized fictional products below and were either told it was a chocolate bar (hedonic product category) or a muesli bar (utilitarian product category).

Figure 3: Heavy Brand Sound/High Color Saturation Figure 4: Light Brand Sound/High Color Saturation

(25)

25

3.4 Sample

The population of this research is basically every shopper in the Netherlands. Therefore, a non-probability sampling technique has been used as data cannot be collected from the complete population and no sampling frame is available. A convenience sample was used in order to reach as much respondents as possible. Respondents were exposed to only one condition (between subjects design). In total 186 respondents filled in the survey from which 12 didn’t complete all questions, thus, only 174 responses were used to analyse the data.

3.5 Measures

The questionnaire (see table 2) asked respondents to evaluate a fictional product on (1) perceived healthiness, (2) overall evaluation, (3) taste expectation, and (4) purchase intention. Evaluation is done on a 5-point Likert scale (1 = fully disagree, 5 = fully agree). The products (or conditions) differ in product category (hedonic or utilitarian), brand name (sound), and package color (saturation). In the end of the questionnaire, respondents will answer eight general questions about their health interest in order to measure the effect of the moderator. General health interest will be measured by a validated Taste and Health attitude scale (Roininen, Lahteenmaki, & Tuorila, 1999). Other scales are validated by Fenko, et al (2016).

(26)

26

Table 2:

Items and reliability of the scales (Fenko & et, What's in a name? The effects of sound symbolism and package shape on consumer responses to food products, 2016)

Scale Items Cronbach’s

Perceived product healthfulness

1. I expect this product to be healthy

2. I would consider this product as good for me 3. This product sounds healthy

4. The product looks healthy

5. The product looks low on calories

6. I have an impression that this product is healthy 7. This product looks healthier than similar products

.913

Product evaluation 1. This product looks good

2. This product will be of good quality 3. This product looks attractive 4. I would notice this product 5. The name sounds attractive 6. The product has a nice color

.833

Taste expectation 1. The texture of this product will be soft 2. This product will have a nice smell 3. This product will have a nice texture 4. I will enjoy this product

5. This product will be crunchy

6. I think this product will be very tasty

.819

Purchase intention 1. I would like to try this product

2. I would seriously consider buying this product 3. I would buy this product

.891

General health interest 1. The healthiness of food has little impact on m food choices (R)

2. I am very particular about the healthiness of food I eat 3. I eat what I like and I do not worry much about the

healthiness of food (R)

4. It is important for me that my diet is low in fat 5. I always follow a healthy and balanced diet

6. It is important for me that my daily diet contains a lot of vitamins and minerals

7. The healthiness of snacks makes no difference to me (R) 8. I do not avoid foods, even if they may raise my

cholesterol (R)

(27)

27

4 Analyses

This research is carried out by using an experimental 2x2x2 survey design which resulted in eight different conditions. Respondents were exposed to only one condition (between subjects design). In total 186 respondents filled in the survey from which 12 didn’t complete all questions, thus, only 174 responses were used to analyse the data.

4.1 Recoding variables

The survey included the Taste and Health attitude scale from Roininen et, al. (1999). This scale consists of eight questions from which four are counter indicative (see table 2). The scale measures participants’ General Health Interest (GHI) on a 5-point Likert scale. Whereas question 2, 4, 5 and 6 are about the extent to which participants are healthy, question 1, 3, 7 and 8 are about the extent to which participants are unhealthy. The ‘unhealthy’ questions are re-coded in order for these to measure the extent to which a participant is considered to have a ‘healthy lifestyle’. Concrete this means that scales were re-coded in the following way: (1=5), (2=4), (3=3), (4=2), and (5=1).

4.2 Reliability analysis

In order to check whether the used measures are really consistent, reliability checks were run for Perceived Product Healthiness (PPH), Product Evaluation (PE), Taste Expectation (TE), Purchase Intention (PI), and General Health Interest (GHI). The results are shown in table 3. Almost all measures score above .7 which indicate a high internal consistency for that particular measure. Only taste expectation (TE) is slightly below this threshold (.655 < .7). However, if question 1 from this measure is deleted, the Cronbach’s Alpha increases to .763 and we can further increase the Cronbach’s Alpha when we delete question 5 as well to .819 (see table 2). The deleted items both were about product texture. It was decided to not to include these questions in any further analysis, since respondents were asked to answer these questions about two very distinct products with different textures (muesli bar and chocolate bar).

Variable Cronbach’s Alpha Items deleted New Cronbach’s Alpha

Perceived Product Healthiness (PPH) .913 0 -

Product Evaluation (PE) .833 0 -

Taste Expectation (TE) .655 2 .819

Purchase Intention (PI) .891 0 -

General Health Interest (GHI) .769 0 -

(28)

28

4.3 Correlation matrix

In order to test the hypotheses, new variables have been created by calculating the mean and standard deviation of all items describing a variable. With these means we ran a correlation to see if there are any patterns observable between the different variables (see table 4). From the correlation matrix we can conclude that from the independent variables only Brand Color (.237, P <.01) and Product Type (.163, P <.05) are significant predictors of Perceived Product Healthiness. Product type also has a weak significant correlation with Product evaluation (.150, P <.05). In contrast, the dependent variables are all moderately to strongly correlated. Purchase Intention and Product Evaluation exhibit the strongest correlation (.766, P < .01). Furthermore, Perceived Product Healthiness is most strongly correlated with Purchase Intention (.511, P < .01), followed by Product Evaluation (.388, P < .01) and Taste Expectation (.372, P < P .01).

(29)

29

Variable Mean SD 1. 2. 3. 4. 5. 6. 7. 8. 9.

1. Perceived Product Healthiness (PPH) 2,74 ,94 (,913)

2. Product Evaluation (PE) 2,73 ,85 ,388** (,833)

3. Taste Expectation (TE) 2.98 ,76 ,372** ,523** (,819)

4. Purchase Intention (PI) 2.73 1,07 ,511** ,766** ,513** (,891)

5. General Health Interest (GHI) 3,23 ,68 ,045 -,025 ,076 -0,27 (,769)

6. Gender 1,60 ,49 -,050 -,072 ,012 ,035 ,198** -

7. Brand Sound 1,48 ,50 -,127 ,009 ,022 -,095 ,090 -,099 -

8. Product Color 1,49 ,50 ,237** ,066 ,023 ,090 ,048 -,127 ,011 -

9. Product Type 1,50 ,50 ,163* ,150* ,023 ,032 -,070 -,108 -,057 -,034 -

**. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-taled).

(30)

30

4.4 Hypothesis testing

Different analyses of covariance were conducted in order to see if there are any main effects between the independent variables (Brand Sound, Brand Color and Product Type) and the dependent variables (Perceived Product Healthiness, Product Evaluation, Taste Expectation and Purchase

Intention) with General Health Interest as covariate. The results are shown in table 5. Multiple

ANOVA’s were conducted as well, in order to be able to compare means within each independent variable.

Dependent variable Independent factor F-value Sig.

Perceived Product GHI 1,207 ,274

Healthfulness Brand Sound 3,259 ,073

Product Color 10,153 ,002** Product Type 6,047 ,015* Sound*Color ,136 ,712 Sound*Product Type 8,633 ,004** Color*Product Type ,005 ,945 Sound*Color*Type ,287 ,593

Product Evaluation GHI ,539 ,464

Brand Sound ,133 ,716 Product Color ,880 ,350 Product Type 4,482 ,036* Sound*Color 8,794 ,003** Sound*Product Type ,359 ,550 Color*Product Type 4,975 ,027* Sound*Color*Type ,301 ,584

Taste Expectation GHI ,939 ,334

Brand Sound ,044 ,834 Product Color ,017 ,896 Product Type ,281 ,597 Sound*Color 7,688 ,006** Sound*Product Type 2,256 ,135 Color*Product Type ,643 ,424 Sound*Color*Type ,005 ,944

Purchase Intention GHI ,213 ,645

Brand Sound 1,409 ,237 Product Color 1,271 ,261 Product Type ,226 ,635 Sound*Color 7,326 ,008** Sound*Product Type 1,740 ,189 Color*Product Type 5,035 ,026* Sound*Color*Type ,186 ,667 **p<.001 and *p<.05 Table 5: Results of ANCOVA with GHI as covariate

(31)

31 In order to find the moderating effect of general health interest (GHI) on the dependent variables, we calculated the median of general health interest. The median is 3,3750 and based on this number we split the sample in two groups. The first group represent people low in general health interest (N=82, M=2,65; SD=,42) and the second group is high in general health interest (N=92, M=3,75; SD=,38) and the difference in GHI between the two groups is significant (F=330,574, P<.01). With this new variable we conducted additional ANOVA’s in order to find the moderating effect of General Health Interest.

Perceived Product Healthiness

The results in table 5 show a significant main effect of Product Color (F = 10.153, P < .001). Packages with colors high in saturation (M = 2,51; SD ,86) are perceived as less healthy than packages with colors low in saturation (M = 2,96; SD = ,96). This effect is greater for people with a high general health interest (see figure 7). Product Type (F = 6.047, P < .05) also shows a significant main effect on

Perceived Product Healthiness. The chocolate bar (M = 2,58; SD = ,88) is perceived as less healthy

than the muesli bar (M = 2,88; SD = ,98). In contrast to our predictions (although this main effect is not significant), products called ‘’Ramune’’ which represent a heavy sounding brand (M = 2,85; SD = ,85) are perceived as healthier than products called ‘’Asashi’’ which represent a light sounding brand (M = 2,62; SD = 1,02). This means that, against our predictions, ‘’Ramune’’ is perceived as healthier than ‘’Asashi’’. However, when we control for General Health Interest, a different pattern emerges for people with a high general health interest (versus people with a low general health interest). We found that people with a relatively low General Health Interest perceive products called ‘Ramune’ (M=2,94; SD=,14) as significantly healthier than products called ‘Asashi’ (M=2,46; SD=,16). In contrast, people with a high General Health Interest don’t perceive a difference in healthiness for products called ‘Ramune’ (M=2,76; SD=.14) and ‘Asashi’ (M=2,73; SD=,14). See also figure 8. General

Health Interest (F = 1.207, P > .05) and Brand Sound are non-significant (F = 3.259, P > .05). From the

interaction effects, only Product Type*Brand Sound seem to show a significant effect (F = 8,633, P > .001) on Perceived Product Healthiness.

(32)

32

Figure 7: Package color saturation has a greater effect

on perceived product healthiness for people with a high (vs. low) general health interest.

Figure 8: Brand sound only has a significant effect on

perceived product healthiness for people with a low (vs. high) general health interest.

Product Evaluation

General Health Interest (F = .539, P > .05), Brand Sound (F = .133, P > .05) and Product Color (F = .880,

P > .05) do not show significant main effects on Product Evaluation. Product Type (F = 4,482, P < .05), however, does show a significant main effect on Product Evaluation. Muesli bars (M = 2,86; SD = .80) are evaluated more positively than Chocolate bars (M = 2,60; SD = .80). From the interaction effects, only Brand Sound*Product Color (F = 8.794, P < .001) and Product Color*Product Type (F = 4.975, P < .05) show a significant interaction. Products named ‘’Ramune’’ were better evaluated when the

Package Color was low in saturation (M = 2,95; SD = .12) as opposed to high in color saturation ( M =

2,50; SD = .13). In contrast, products with the name ‘’Asashi’’ were better evaluated when the

Package Color was high in saturation (M = 2,87; SD = .13) as opposed to low in saturation (M = 2,61;

SD = .13).

However, this congruence effect between Brand Sound and Color Saturation only holds for people that have a low General Health Interest (see figure 9). Respondents high in GHI evaluate products called ‘’Ramune’’ (M=3.04; SD=.162) and ‘’Asashi’’ (M=2.77; SD=.15) in packages with low saturated colors better than products called ‘’Ramune’’ (M=2.61; SD=.18) or ‘’Asashi’’ (M=2.30; SD=.18) in packages with high saturated colors (see figure 10). Thus, the congruence effect disappears for people that are on average more concerned with their health due to their high general health interest.

(33)

33

Figure 9: Congruence effect between brand sound

and color saturation for people low in General Health Interest.

Figure 10: The congruence effects of brand sound and

color saturation disappears for people high in General Health Interest.

For the Product Color*Product Type interaction, chocolate bars are significantly better evaluated when they have a low-saturated package (M = 2,78; SD = ,78) as opposed to a high-saturated package (M = 2,41; SD = ,96). This is surprising, as we expected the exact opposite effect since the low saturated package is not corresponding with the ‘unhealthy’ chocolate bar. For muesli bars there is no significant difference in product evaluation for difference in color saturation.

Figure 11: The effect of Product category*Color

saturation on Product evaluation for people low in general health interest.

Figure 12: The effect of Product category*Color

saturation on Product evaluation for people high in general health interest

(34)

34 However, when we control for GHI we find a different effect. Respondents low in GHI evaluate chocolate bars (M=2.65; SD .17) and muesli bars (M=3.06; SD .17) better when they have a package high in color saturation as opposed to chocolate bars (M=2,47; SD .24) and muesli bars (M=2.70; SD=.18) low in color saturation (see figure 11).

In contrast, respondents high in GHI evaluate muesli bars (M=2.75; SD=.18) better than chocolate bars (M=2.09; SD=.19) when they have a package high in color saturation. Respondents high in GHI do not evaluate chocolate bars (M=2,90; SD=.15) and muesli bars (M=2.90; SD=.18) differently when they have a package low in color saturation (see figure 12). Despite these effects, we cannot confirm that there exists a congruence effect or a crossmodal correspondence between Product Color and Product type for Product Evaluation.

Taste Expectation

Within Taste Expectation, only Brand Sound*Product Color shows a significant interaction (F = 7.668), P < .01). There are no main effects on Taste Expectation. Products called ‘’Ramune’’ are expected to taste better when the package color is low in saturation (M = 3,10; SD = ,64) as opposed to high in saturation (M = 2,83; SD = ,61). For products called ‘’Asashi’’ the effect is the other way around. ‘’Asashi’’ is expected to taste better when the package is high in saturation (M = 3,12; SD = ,45) as opposed to low in saturation (M = 2,91; SD = ,58). This means that there exists a congruence effect or a so called crossmodal correspondence between Brand Sound and Product Color. Respondents expect that congruent combinations (healthy sounding brand with low saturated colors or unhealthy sounding brand with high saturated colors) taste better. However, this only holds for people low in General Health interest (see figure 13). Respondents high in GHI expected that products called ‘’Ramune’’ (M=3.14; SD=.52) or ‘’Asashi’’ (M=2,99; SD=.57) both tasted better when the package was low in color saturation as opposed to products called ‘’Ramune’’ (M=2.88; SD=.62) or ‘’Asashi’’ (M=2.93; SD=.54) in packages high in color saturation (see figure 14).

(35)

35

Figure 13: The congruence effect of Product

category*Color saturation on Taste Expectation for people low in general health interest.

Figure 14: The congruence effect between Product

Category and Color Saturation on Taste Expectation disappears for people that have a high general health interest.

Purchase Intention

For Purchase Intention there are no main effects of Brand Sound, Package Color and Product Type. However, there are two significant interaction effects namely: Brand Sound*Package Color

Saturation (F = 7.326, P < .01) and Product Type*Package Color Saturation (F = 5.035, P < .01).

Respondents show a higher Purchase Intention for products called ‘’Ramune’’ when they have a low saturated package (M = 3,14; SD = ,99) as opposed to a high saturated package (M = 2,54; SD = ,95). In contrast, respondents show a higher Purchase Intention for products called ‘’Asashi’’ when they have a high saturated package (M = 2,75; SD = 1,14) as opposed to a low saturated package (M= 2,51; SD = 1,11). However, this only holds for people low in General Health Interest (see figure 15). Respondents high in GHI showed a higher purchase intention for products called ‘’Ramune’’ (M=3.03; SD=1.11) or ‘’Asashi’’ (M=2.92; SD=.98) in packages with low saturated colors as opposed to products called ‘’Ramune’’ (M=2.63; SD=1,11) or ‘’Asashi’’ (M=2.20; SD=.72) in packages with high saturated colors (see figure 16).

(36)

36

Figure 15: The congruence effect of Brand Sound*Color

Saturation on Purchase Intention for people low in general health interest.

Figure 16: The congruence effect between Brand

Sound and Color Saturation on Purchase Intention disappears for people that have a high general health interest.

The Product Type*Package Color Saturation interaction shows that respondents have a significantly higher Purchase Intention for chocolate bars when they have a low-saturated package (M = 2,78; SD = ,78) as opposed to a high-saturated package (M = 2,41; SD = ,96). For muesli bars there is no significant difference in Purchase intention for difference in color saturation. However, this only holds for people that have a high General Health Interest (see figure 17). People low in GHI show a greater purchase intention for muesli bars (M=2.88; SD=1.04) and chocolate bars (M=2.75; SD=1.17) with high saturated package colors as opposed to muesli bars (M=2.70; SD= 1.13) and chocolate bars (M=2,44; SD=1.28) with a low saturated package (see figure 18).

Figure 17: The effect of Product Type*Color Saturation

on Purchase Intention for people low in general health interest.

Figure 18: The effect between Product Type and

Color Saturation on Purchase Intention disappears for people that have a high general health interest.

(37)

37

5 Results

Previous analysis revealed a significant main effect of Package Color Saturation on Perceived Product

Healthiness which is in line with H1. Products that have a package color that is highly saturated (very

bright colors) are perceived to contain unhealthier products than when the same product has a package color with low saturated colors. This finding is in line with previous research findings (Hoegg & Alba, 2007). Hoegg & Alba (2007) demonstrated that the experienced ‘sweetness’ of orange juice was related to the color intensity of the packaging. Meaning that consumers perceive more intense colors as an indication of a sweeter product. Brand Sound, in contrast, does not show such an effect. We proposed that ‘‘heavy sounding’’ brands (Ramune) would be perceived as less healthy than ‘’light sounding’’ brands (Asashi). In our study no such effect occurred. In fact, we found a tendency towards significance which showed that products called ‘’Ramune’’ (which were supposed to sound heavy) are being perceived as healthier than products called ‘’Asashi’’ (which were supposed to sound light). Thus, H2 is not proven. Product Type does show a significant effect on Perceived Product

Healthiness. Chocolate bars which represent more hedonic goals are perceived to be less healthy

than muesli bars. This finding is in line with H3.

Despite the absence of a main effect, Brand Sound does show significant interaction effects with

Package Color Saturation on Product Evaluation, Taste Expectation and Purchase Intention. This is in

line with H4a. We predicted that congruence between Brand Sound and Package Color Saturation would lead to better Product Evaluation, better Taste Expectation and higher Purchase Intention than incongruent combinations. The interaction between Brand Sound and Package Color shows that healthy sounding products (Ramune) in combination with a healthy appearance (low saturated colors) are better evaluated, are expected to taste better and respondents are more likely to buy them. The same holds for unhealthy sounding products (Asashi) which are better evaluated, are perceived to taste better and show a higher purchase intention when the package color is high in saturation. Thus, we can say that there exists a congruence between high saturated colors and ‘unhealthy sounding’ brands (and also between low saturated colors and healthy sounding brands). The need for congruence is underlined in various studies (Becker, van Rompay, Schifferstein, & Galetzka, 2011) (Spence, 2012) (Fenko, Lotterman, & Galetzka, 2016), in which it is argued that marketers can enhance their consumers’ product experiences by ensuring that the brand name, the shape of the packaging and product-related sensory expectations like taste, are all in line within the mind of the consumer.

(38)

38 Another interesting interaction effect can be found between Product Category and Package Color

Saturation which have a significant interaction effect on Product Evaluation and Purchase Intention.

We predicted that Congruence between Product Category and Package Color Saturation will lead to (a) better perceived taste, (b) higher purchase intention, and (c) better product evaluation than incongruent combinations. Congruent combinations are (1) chocolate bars with a high saturated package color and (2) muesli bars with a low saturated package color. However, we found that chocolate bars are significantly better evaluated when they have a low-saturated package as opposed to a high-saturated package. For muesli bars there is no significant difference in product evaluation for difference in color saturation. The same holds for Purchase Intention. No effect was found for perceived taste, thus, H4c is partly supported. Hypothesis H4b and H4d have not been proven by the results of this study. We couldn’t find an effect

As we have seen, the described effects differ for people that have a low general health interest (versus people with a high general health interest). This supports H5. The congruence effects described in H4a and H4c only hold for people with a low general health interest. This means that people low in general health interest rely more on heuristics and peripheral cues than people high in general health interest.

6 General discussion

Taken together, the findings of this study show an important contribution to the literature related to sensory marketing and cross-modal correspondences. We sought to shed light on the effects of different sensory stimuli on consumer perceptions and their buying behavior. In particular, we aimed at a very specific problem which is best described as a cross-modal correspondence between package color and brand sound and the effect it has on consumers when considering a more hedonic or utilitarian food option. The results of this study demonstrated that such a cross-modal correspondence exists and that it affects the mind of the consumer. More specifically, product evaluation and purchase intention are higher if brand sound and package color are congruent. However, such an effect only occurs when people are relying relatively more on heuristics. This means that consumers that have a low general health interest would prefer a product called Ramune (healthy sound) in a low saturated package (healthy appearance) and a product called Asashi (unhealthy sound) in a high saturated package (unhealthy appearance) over all other incongruent combinations. In contrast, people with a relatively high general health interest would prefer a product with a low saturated package irrespective the extent to which a brand name sounds healthy. Another cross-modal correspondence can be found between product category (hedonic vs. utilitarian) and package color (high vs. low saturation) which affects taste expectation. This effect

(39)

39 also holds only for people with a low general health interest. These results imply that cross-modal correspondences are only affecting those consumers which rely relatively more on external cues to make inferences about a particular product, since these product related cues help consumers to evaluate trade-offs without intensive cognitive processing (Gottfried & Dolan, 2003).

The observation that cross-modal correspondences are only affecting those consumers which rely relatively more on external cues to make inferences about products does not mean that sensory stimuli in isolation are not important. Although not the main focus of this study, you can see that package color saturation and brand sound show quite the opposite effect for people with a low (vs. high) general health interest (see figure 7 and 8). It seems for instance that the sound of a brand rather affects people with a low (vs. high) general health interest and that color saturation has a greater effect for people with a high (vs. low) general health interest. This is an interesting finding, because it indicates that people who rely relatively more on heuristics are not always those who are less informed/interested and therefore try to avoid cognitive effortful processing. If that was the case, than the effects of brand sound and package color would show the same pattern for people with a low general health interest (and for people with a high general health interest).

Practical implications

This study helps mangers to understand the consequences of seemingly unimportant aspects of marketing. We have shown that it is important to not only focus on the brand name and package color of a particular product but also on the sound of the brand name and the saturation of colors. And maybe even more important, the interaction between brand sound and color saturation. These aspects lie at the very core of (new) product development and should not be underestimated. Therefore, when developing a new product, managers should emphasize congruent sounds and color saturation for a particular product category (either hedonic or utilitarian). This will enhance the ability for consumers to process information more fluently which makes them more open to actually purchasing the product. Moreover, for marketing managers it is easier and less expensive to advertise a product which already holds the desired associations in its primary brand elements. Next to the positive effects of congruency effects or cross-modal correspondences, it is also important to underline the limitations of using these symbolism effects. Some consumers are less triggered by these heuristics and potential ‘problem brands’ may not be saved by simply reposition or relaunch a product. In our study, for instance, people with a high general health interest are not influenced by cross-modal correspondences. This means that if your customer base mainly consists

It sounds like a bright color : the effects of package color saturation and brand sound on perceived product healthiness