Crafting to perfection:
The effect of Perceptual Fluency in craft beer labels
on Taste Expectations and Purchase Intention
2
Crafting to perfection:
The effect of Perceptual Fluency in craft beer labels on Taste Expectations and
Purchase Intention
18th June 2018
Master Thesis Marketing University of Groningen
Faculty of Economics and Business
MSc Marketing Management Author Mitchell Oost S3260968 m.l.oost@student.rug.nl Meidoornpad 44 9713NP Groningen Supervisors
First supervisor: Dr. Y. Joye
Second supervisor: Dr. M. Moeini-Jazani
3 ABSTRACT
Product design helps in attracting consumers’ attention and adds value to a product, therefore the design of a product can be an important element of its success on the market. Consumers use the design and attractiveness of a product in their decision making process. An element that is used to evaluate or process a product’s visual appearance is perceptual fluency. Perceptual fluency is involved with identifying the physical identity of a stimulus through low-level processing. This research aims to find out more about the relationship between perceptual fluency in craft beer labels, deriving from visual complexity and colour contrast, and consumer behaviour. More specifically, how perceptual fluency in craft beer labels is related to consumer taste expectations and purchase intention. Earlier research showed that objects are more positively assessed when they are processed with higher levels of fluency rather than lower levels of fluency. During an experiment (N = 130), perceptual fluency was manipulated by two element, visual complexity and colour contrast. The findings of this research suggest that perceptual fluency, facilitated by perceived colour contrast, is positively related to the strength of taste expectations, attractiveness and purchase intention.
Furthermore, consumers seem to prefer a moderate level of complexity in the design of craft beer labels which suggests that there might be an ideal point in terms of complexity and liking. Overall, this research provides more insights on how perceptual fluency is related to consumer behaviour and taste expectations. In addition, it gives marketers, managers and brewers more insights on how they can use the visual design of their product packaging to influence consumers’ taste expectations and consumer purchase behaviour.
4 PREFACE
This thesis is my last hurdle in order to complete the Master Marketing programme at the University of Groningen. Together with the Pre-Master, it has been an incredibly valuable and enlightening experience. Two years of hard work definitely paid off and now the time has come to apply the acquired knowledge in business.
I have always been fascinated by the design of products and product packaging. I am also a craft beer enthusiast. Therefore, I am glad that I had the opportunity to combine both subjects to one thesis topic. The process of writing this thesis gave me the opportunity to learn much more about how visual design influences consumer behaviour and product evaluations, and about the interesting concept of perceptual fluency.
I would like to thank my supervisor Dr. Yannick Joye for his constructive feedback, guidance, motivation and for giving direction to my research. His help was most valuable and he has been a great supervisor during the entire thesis writing process. Furthermore, I would like to thank all the members of my thesis group for the moments that we shared our experiences and for keeping each other motivated along the road. I also want to thank Justin for his help with designing the craft beer labels that were used in the experiment. Lastly, I would like to thank my parents and Ilse for their support during the journey.
Mitchell Oost
5 LIST OF FIGURES AND TABLES
1. Figures
Figure 1: Conceptual model 19
Figure 2: Condition mean 26
Figure 3: Graph interaction H4c 33
Figure 4: Graph interaction H4d 34
2. Tables
Table 1: Research design 21
Table 2: Results factor analysis. 25
Table 3: Interaction H4b 31
Table 4: Interaction H4c 32
6 TABLE OF CONTENTS
1. INTRODUCTION ... 8
2. LITERATURE REVIEW ... 11
2.1 Packaging and product label design ... 11
2.2 Perceptual fluency ... 12
2.2.1 Simplicity/complexity ... 13
2.2.2 Colour contrast ... 14
2.3 Purchase intention ... 15
2.4 Perceived taste expectations ... 15
2.5 Attractiveness ... 17
2.6 Centrality of Visual Product Aesthetics ... 18
2.7 Conceptual model ... 20
2.7.1 Overview of all the hypothesis ... 20
3. METHODOLOGY ... 21
3.1 Data collection ... 21
3.2 Research design and data measurements ... 21
3.3 Data analysis... 24 4. RESULTS ... 25 4.1 Descriptive statistics ... 25 4.2 Factor analysis ... 25 4.3 Manipulation checks ... 26 4.4 Hypothesis testing ... 27
4.4.1 Direct effects on purchase intention ... 27
4.4.2 Direct effects on taste intensity ... 28
4.4.3 Effects involving attractiveness ... 29
4.4.4 Effects involving Centrality of Visual Product Aesthetics ... 31
4.5 Overview of the findings ... 35
5. DISCUSSION... 36
5.1 Conclusions ... 36
5.2 Managerial implications ... 38
5.3 Scientific implications ... 39
5.4 Limitations... 39
5.5 Suggestions for future research ... 40
7
APPENDICES ... 47
Appendix I: Beer label conditions in the experiment ... 47
Appendix II: Survey flow ... 51
Appendix III: Scales ... 62
8 1. INTRODUCTION
Craft beer is booming! After being dominated by lager for the past 70 years, craft beer is gaining ground in the beer market. And so are the number of breweries, especially in Western countries. In the United Kingdom alone, the number of breweries rose with 64 percent in the past five years. Adding up to a total of 2,000 breweries (Monaghan, 2017). According to the Cambridge Dictionary, craft beer (or craft ale) can be defined as ‘beer made using traditional methods in small independent breweries.’ In most Western countries, consuming and brewing craft beer is on the rise. In the Netherlands, four out of five consumers drink craft beer on a regular basis. This is 40 percent more compared to three years ago (Ruigrok NetPanel, 2017). Furthermore, in the past ten years the number of breweries in the Netherlands has quadrupled (CBS, 2018).
9 Earlier research showed that package design plays, especially when the decision is made with low cognitive load, an important role in consumer decision making (Milosavlevic, 2012). There are several studies that focus on the effect of beer packaging on preferences and taste perception. However, the concept of fluency has not been taken into account yet. Fluency refers to ‘the ease of which information is being processed’ (Reber, Schwarz & Wienkelman, 2004, p. 368). People seem to hold positive attitudes, reactions and behaviour towards stimuli that take little effort to process, and are thus perceived as fluent (Reber et al., 2004). Since most everyday purchase decisions are made on the spot, consumers might prefer products that are easy to process. Therefore, perceived fluency of the beer label might play a role in
persuading the consumer to purchase a certain beer. Furthermore, taste is an important indicator of whether consumers want to purchase a craft beer (Ruigrok NetPanel, 2017). Earlier research showed that fluency, in terms of easy versus difficult to read or pronounce brand names, influences a consumer’s expected taste experiences (Mantonakis et al., 2013). It is interesting to know whether this also counts for other design elements that facilitate
perceptual fluency. Furthermore, could it be that fluency in labels plays a role in determining a consumer’s taste expectations when it comes to craft beer? Especially since taste is such an important factor for consumers when it comes to purchasing craft beer.
Although there is a lot of research available in which perceptual fluency shows a positive effect on consumer responses, there are also studies showing that disfluency in visual stimuli is occasionally more preferred (Berlyne, 1971). However, there seems to be little
understanding on when, where and for which stimuli and products this is the case. More research on how perceptual fluency is used by consumers to evaluate products might provide more insights on the effects of perceptual fluency. Since consumers do tend to use the
physical appearance of products to evaluate it in various ways. Marketing can benefit from more knowledge on packaging design, the role of perceptual fluency and the effects it has on consumer evaluations and consumer behaviour (Clement, 2007). Especially in the highly competitive yet fast growing market for craft beer. Therefore, this research aims to answer the following research question:
10 For this research, two dimensions and visual design elements that facilitate perceptual fluency were chosen to study its effects on taste perception and purchase intention when it comes to craft beer. The first one being visual complexity (and the opposite end of the continuity: simplicity). The second one being colour contrast. Both dimensions facilitate visual processing (Reber et al., 2004).
This paper is structured as follows; after the introduction, an overview of relevant literature is presented. This review discusses earlier research about product packaging, perceptual fluency, the effects of attractive product design and taste expectations. It provides an overview of all the constructs, their expected relations with each other and the formulated hypotheses. The conceptual model can be found at the end of this section. After the literature review comes the methodology section. The data collection method and the experimental design are explained in this section. This section also discusses which tests were used to analyse the collected data. After the methodology section comes the results chapter. This chapter describes and analyses the relevant results. Thereafter comes the discussion chapter in which the results are reflected based on the literature that was discussed earlier. Furthermore, limitations of this research and suggestions for future research are also addressed in this section.
11 2. LITERATURE REVIEW
2.1 Packaging and product label design
The physical design of a product is an important element of its success on the market. Good product design helps in attracting the consumer’s attention and adds value to the product (Bloch, 1995). Especially when it comes to fast moving consumer goods (FMCG), product packaging can play a crucial role (Wansink & Huffman, 2001). A study by Cooper and Kleinschmidt (1987) showed that product design was the most, and thus more important than price, determinant of product success regarding sales. People tend to choose with their eyes. Nine out of ten consumers base their in-store purchase decision after viewing the front of the package, without even touching the actual product (Urbany, Dickson, & Kalapurakal, 1996). One could agree that price is likely to play an important role in the decision making process, however according to Vanhuele and Drèze (2002), consumers do not put much effort in cognitive processing like analysing prices while in-store. Furthermore, when consumers have to choose between two products that have the same product characteristics and are equal in price, they will purchase the product they regard as being more attractive (Bloch, 1995). In addition, the packaging of products is an important tool which enables firms to position themselves in the market, communicate their image to the consumer and differentiate themselves from the competition (Underwood, Klein, & Burke, 2001). In the eyes of the consumer, attractive packaging design also generates positive product and brand evaluations. Furthermore, visual packaging influences the perceived quality of the product and has a positive influence on brand preference (Wang, 2013).
12 2.2 Perceptual fluency
Perceptual fluency is an element of the concept of processing fluency or simply fluency, which is about the processing of any stimulus. This refers to the processing of internal mental events that are not assignable to the content of the stimulus, like the rate, amount of effort and precision of how the stimulus is processed. Thus, processing fluency or fluency refers to ‘the ease of which information is being processed’ (Reber et al., 2004, p. 368). This phenomenon is also known as hedonic fluency. Because the ease of processing is experienced as a moderate positive emotion which can be used as information to evaluate an object or stimulus (Fennis & Stroebe, 2016; Haugtvedt, Herr & Kardes, 2008). According to Winkielman et al. (2003), Objects and stimuli are more positively assessed when they are processed with higher levels of fluency rather than lower levels of fluency. Generally, higher levels of fluency are more symptomatic for positive states of the cognitive system or the environment, while lower levels of fluency are symptomatic for negative states in those areas (Winkielman et al., 2003). Furthermore, higher rates of perceptual fluency stimulate people to elaborate vivid information and to process visual information (Jacoby et al., 1988). Due to an increase in attention, perceptual fluency may have a positive effect on cognitive effort. McGill and Anand (1989) found out that people appoint cognitive effort to process visual information when the object attracts attention, which is enhanced by perceptual fluency (represented by clarity and colour).
Fluency can take place at two different levels, perceptual fluency and conceptual fluency. Perceptual fluency ‘reflects the ease of low-level, data-driven operations that deal primarily with "surface" features of the stimulus, or its perceptual form’ (Winkielman et al., 2003, p. 6). Thus, how easy it is for a person to identify and process visual elements of a stimulus.
Conceptual fluency refers to the ‘ease of mental operations concerned with stimulus meaning and its relation to semantic knowledge structures’ (Reber et al., 2004, p. 368). Or in other words; ‘the ease with which associations come to mind’ (Tversky & Kahneman, 1973, p. 211). While perceptual fluency is involved with identifying the physical identity of the stimulus through low-level processing, conceptual fluency is involved with the identification of its meaning. This involves high-level, thus more conscious, processing (Winkielman et al., 2003). For example; one sees a picture of a car. Perceptual fluency identifies that the object in the picture is a car. Conceptual fluency then identifies what the car is, a mean of
13 separate constructs with different predecessors and consequences (Lee, 2002; Tulving & Schacter, 1990).
Since this research is about the effects of certain visual stimuli on purchase intention and taste perception of craft beer, the focus of this research will be on perceptual fluency. Research shows that being exposed to a stimulus or object (e.g. a brand name, logo or advertisement) multiple times can increase fluency. This is because every time a consumer gets exposed, it becomes easier to recognize and process that stimulus or object (Lee & Labroo, 2004). This is also known as the mere exposure effect, which boosts the attitude of an individual towards an object due to mere repeated exposure (Zajonc, 1968). This also holds for exposures that are incidental or when consumers are not consciously paying attention towards the exposure. Thus, fluency is not always a conscious experience (Fennis & Stroebe, 2016; Winkielman et al., 2003). Furthermore, Reber et al. (2004) suggest that the more an object is perceived as being fluent, the greater the positive aesthetic response of the perceiver towards that object will be. In their research, Lee and Labroo (2004) already found out that an increase in fluency positively influences consumers’ choices. Furthermore, research by Janiszweki and Meyvis (2001), Reber, Winkielman and Schwarz (1998) and Winkielman and Ciaciopo (2001), concluded that fluency has a positive effect on the attitudes that consumers have towards brands or products which results in increased liking and preference for those brands and products.
Elements that facilitate in fluent visual processing are; simplicity/complexity, contrast/clarity, symmetry, prototypically, semantic priming, duration of exposure and previous experiences with the stimulus (Reber et al. 2004; Batra, Seifert & Brei, 2015). They influence perceptual fluency since these elements affect the accuracy and speed of the processes that are involved in identifying the physical form of the stimulus (Schwarz, 2004). Elements that assist in fluent processing result in enhanced liking, even when the perceiver has only been exposed to the element one single time (Schwarz, 2004; Reber et al., 1998; Winkielman et al., 2003).
2.2.1 Simplicity/complexity
14 presented in a stimulus (Reber et al. 2004; Batra et al., 2015). Complexity is also an identified element of universal design factors. It captures the richness of a certain design and its ability to grasp the meaning of something (Orth & Malkewitz, 2008). The number of different shapes in a stimulus can indicate the level of simplicity or ‘figural goodness’ of a stimulus (Garner, 1974; Reber et al., 2004). More different shapes mean that the viewer needs to take more time and has to put in more cognitive effort to extract the information in order to perceive it. Thus making the stimulus more complex, because of conflicting elements, and resulting in a
decrease of fluency (Schoormans & Robben, 1997). This is also in line with the preference for symmetry that people have because stimuli with identical shapes contain less information, than stimuli that are non-symmetric, and are thus easier to process (Garner, 1974; Reber et al., 2004). Although earlier on we have discussed evidence that a higher level of visual simplicity enhances fluency and thus the liking of that stimulus, there are also researchers who show that this is not always the case. Landwehr, Labroo and Herrman (2011) showed that car types that have a complex, yet also prototypical, designs are more fluently processed and evoke more positive reactions towards the design among consumers. Car types with a complex and prototypical design were also more successful in terms of car sales. Cox and Cox (2002) showed that simple designs were being processed more fluently and were more preferred over complex designs, but that this effect reverses due to mere-exposure. Furthermore, research by Joye, Steg and Ünal (2016) confirms that repetition of visual stimuli of objects that are complex can enhance perceptual fluency. Although there is also evidence showing that a moderate level of complexity is preferred (Palmer, Schlos & Sammartino, 2013). This is coherent with the inverted U-shape relationship between liking and simplicity/complexity in Berlyne’s (1971) model of arousal (Palmer et al., 2013; Batra et al., 2015). Thus, an ideal point in terms of complexity in visual design possibly exists depending on the circumstances. Overall, one can conclude that research results vary in the way that simplicity and complexity influences perceptual fluency within different settings and for different product categories. More research about the role of simplicity and complexity in visual design could provide additional insights on their role in perceptual fluency.
2.2.2 Colour contrast
15 an indicator of perceptual fluency. A higher contrast level results in a higher degree of
perceptual fluency and increases liking for the stimulus (Reber et al, 1998; Rebert et al. 2004). Studies conducted by Reber and Schwarz (1999) and Unkelbach (2007) showed that
respondents perceived statements that were presented in high-contrast colours as being more true than statements that were presented in low-contrast colours. Furthermore, respondents were able to judge statements that were presented in high-contrast colours at a quicker rate, which is an indication of improved fluency compared to the condition with low-contrast colours.
2.3 Purchase intention
According to Fishbein and Ajzen (1975), a consumer’s intentions to purchase something are influenced by external components, attitudes of the consumer and the consumer’s evaluations. This relates back to perceptual fluency since easy processing of visual (external) elements results in positive attitudes and evaluations towards the stimulus (Reber et al., 1998). A study by Storme et al. (2015) showed that fluency in advertisements has a positive effect on
purchase intention. Thus, a higher level of fluency results in an increase of purchase intention. Also because of the positive effects of high-level fluency in the multiple studies that have been discussed, the expectation is that this will also be the case for the label design of craft beer. Especially since purchase decisions are often made in-store under the conditions of fast everyday decision making with little cognitive load (Vanhuele & Drèze 2002; Milosavlevic et al., 2012). Thus it is likely that consumers tend to choose a product with a higher level of perceptual fluency. Therefore, the following is hypothesised:
H1a. Higher colour contrast in craft beer labels is positively related to purchase intention.
H1b. Complexity in craft beer labels is negatively related to purchase intention.
2.4 Perceived taste expectations
16 experience and taste perception, but that studies on specific impact of particular colours view mixed results. Although there is also strong evidence that learned colour-taste associations exist (Pangborn, Berg & Hansen, 1963; Spence et al., 2015). Research on the effect of beer labels on taste perception and purchase intention has been conducted before. Research by Allison and Uhl (1964) showed that beer labels, and the associations consumers have with those labels, influence consumer taste evaluations of lager beers. The perceived taste
differences of consumers between different lager brands were a result of the marketing efforts of those brands, rather than through actual taste differences. Barnett and Spence (2016) found out that, when emphasizing citruses and fruitiness on the label, green labels received higher ratings on taste, perceived quality, purchase intention and the presence of citrus/fruity notes compared to a brown label or no label at all. It appears that brighter colours (e.g. green, yellow and red) of packaging design evoke associations with taste perceptions like sweetness and sourness. While darker colours like black and grey evoke taste perceptions like bitterness (Spence et al., 2015). What seems to be empirically evident is that colour influences
consumers’ perception on product freshness and taste intensity (Delwiche, 2004). Taste intensity refers to the intensity or strength to which taste stimuli, both separately and/or combined (e.g. sweetness and bitterness), are perceived by an individual (Marks & Bartoshuk, 1979). Furthermore, a study on the effects of colours on emotions found that colour brightness and saturation results in objects being perceived as more pleasant and evoke more emotions (Valdez & Mehrabian, 1994). Hence, we assume that higher colour contrasts can result in stimuli being experienced as more intense. This may result in higher perceived taste sensations as well. Therefore the following is hypothesised:
H2a. Higher colour contrast in craft beer labels is positively related to perceived taste intensity.
17 that there is evidence that, in some cases, consumers prefer a moderate level of complexity (Landwehr et al., 2011; Palmer et al., 2013). Since consuming craft beer is more associated with pleasure rather than that it is very useful, it seems obvious that craft beer is more a hedonic kind of product. Thus making it more likely that complexity, to a certain amount, in label design could result in higher perceived taste evaluations. This does not immediately have to mean that complexity, and thus disfluency, is better or more preferable when it comes to taste. The essence is that it could lead to expected taste being more intensely perceived (i.e. stronger doesn’t always has to imply better). Whether consumers prefer strong taste depends on individual preferences. Therefore, the following is hypothesised:
H2b. Complexity in craft beer labels is positively related to perceived taste intensity.
2.5 Attractiveness
The importance of packaging design has already been emphasised earlier on. One reason why packaging design is important was that the physical appearance of products is used by
consumers to evaluate brands and products and that it is an important influencer of consumer behaviour (Bloch, 1995; Wang, 2013). Therefore, it is important for the product to appear attractive in the eyes of the consumer. Marketers and firms can control the physical
appearance of their products and thus also the attractiveness of the products perceived by the consumer. Making the visual design of a product an effective tool to influence consumer behaviour. Although there are multiple ways to influence aesthetic responses towards visual design, the emphasis of this research lays on the effects of fluency. Evidence exists that fluency enhances positive aesthetic responses, affective judgements like attractiveness, pleasure, positive consumer attitudes and increases consumer liking (Janiszweki & Meyvis, 2001; Reber et al., 1998; Winkielman & Ciaciopo, 2001; Reber et al., 2004). Furthermore, an attractive physical appearance influences consumers when they are making purchase
decisions. (Cooper & Kleinschmidt, 1987). Not only do consumers use the physical appearance of the product to simply evaluate the quality of products, they also use it to
18 H3a. Higher colour contrast in craft beer labels is positively related to attractiveness.
H3b. Complexity in craft beer labels is negatively related to attractiveness
H3c. A beer label’s attractiveness is positively related to purchase intention.
H3d. A beer label’s attractiveness is positively related to perceived taste intensity.
H3e. Attractiveness mediates the effect of colour contrast and complexity in beer labels on purchase intention.
H3f. Attractiveness mediates the effect of colour contrast and complexity in beer labels on perceived taste.
2.6 Centrality of Visual Product Aesthetics
The previous sections described that fluency has a positive effect on aesthetic responses from consumers. Furthermore, it provided evidence that product design matters for consumers when they are in the process of making purchase decisions (Bloch, 1995). However, it is certainly not assumable that product design matters just as much for every consumer. Differences in design orientation exist between markets and consumer segments, as well as between individual consumers. Some consumers attach more value to design and are more sensitive for design than others (Bloch, Brunel & Arnold, 2003). This can be explained with the Centrality of Visual Product Aesthetics (CVPA). This refers to ‘the overall level of significance that visual aesthetics hold for a consumer in his relationships with products’ (Bloch et al., 2003, p. 551). CVPA consists of four related dimensions. The first is the value a consumer assigns to product experiences, the second being the ability to evaluate and
recognize product appearances, the third is the response to aspects of product design while the fourth and last refers to the impact of visual design on product satisfaction and preferences (Bearden, Netemeyer & Haws, 2011). Consumers that score high on CVPA are more
19 fluency on attractiveness is enhanced for consumers that have a high level of sensitivity for visual aesthetics. Therefore, the following is hypothesized:
H4a. The effect of colour contrast and complexity in craft beer labels on attractiveness is greater for people with a higher CPVA.
In their study, Bloch et al. (2003) recognized that consumers with high levels of CVPA found product design more important when making purchase decisions than consumers with low CVPA. Thus, CVPA plays a role in consumer decision making when it comes to buying products. Because consumers with high levels of CPVA find product design more important, it is expected that they care more about the level of perceptual fluency of a product or label when making a purchase decision. Therefore, the following is hypothesized:
H4b. The effect of colour contrast and complexity in craft beer labels on purchase intention is greater for people with a higher CVPA.
DeBono et al. (2003) conducted a research where they asked participants to evaluate the taste of chocolate, coffee and bottled water. They found out that people who find the visual
appearance of products very important, give attractive products a higher taste rating than non-attractive products. As Bloch (1995) and Wang (2013) already noted, people evaluate
products based on their physical appearance. Therefore, there is reason to believe that people who are more sensitive for visual aesthetics give higher taste evaluations and have higher purchase intentions for attractive craft beer labels. This leads to the following hypotheses:
H4c. The effect of colour contrast and complexity on perceived taste intensity is greater for people with a higher CVPA.
20 2.7 Conceptual model
Figure 1: Conceptual model. 2.7.1 Overview of all the hypothesis
H1a. Higher colour contrast in craft beer labels is positively related to purchase intention. H1b. Complexity in craft beer labels is negatively related to purchase intention.
H2a. Higher colour contrast in craft beer labels is positively related to perceived taste intensity.
H2b. Complexity in craft beer labels is positively related to perceived taste intensity. H3a. Higher colour contrast in craft beer labels is positively related to attractiveness. H3b. Complexity in craft beer labels is negatively related to attractiveness
H3c. A beer label’s attractiveness is positively related to purchase intention.
H3d. A beer label’s attractiveness has is positively related to perceived taste intensity. H3e. Attractiveness mediates the effect of colour contrast and complexity in beer labels on purchase intention.
H3f. Attractiveness mediates the effect of colour contrast and complexity in beer labels on perceived taste.
H4a. The effect of colour contrast and complexity in craft beer labels on attractiveness is greater for people with a higher CVPA.
H4b. The effect of colour contrast and complexity in craft beer labels on purchase intention is greater for people with a higher CVPA.
H4c. The effect of colour contrast and complexity on perceived taste intensity is greater for people with a higher CVPA.
21 3. METHODOLOGY
3.1 Data collection
A survey was created in order to collect data for this research. The questionnaire was designed using Qualtrics. This is an online research platform that is freely available for students and employees of the University of Groningen, which makes it a cost-efficient method to collect data. With Qualtrics it is possible to easily distribute the questionnaire through online communication channels. Allowing researchers to reach a large group of respondents. Another advantage with online questionnaires is that respondents can fill out the questionnaire at their own convenience on their computer or smartphone. For this research, the survey was distributed via e-mail and social media channels like Facebook, LinkedIn and WhatsApp. The target population consists of all adults living in the Netherlands who are eighteen years or older. The reason why this research is only interested in adult citizens is because people in the Netherlands are not allowed to drink alcohol, and thus craft beer, until they reach the age of eighteen. The sampling method that was used for collecting the data was convenience sampling. This is a time and cost-efficient method of collecting data. A
disadvantage of using this sampling technique is that some groups in the population might be underrepresented. The data was collected in May 2018.
3.2 Research design and data measurements
Participants were randomly exposed to one of the four craft beer label conditions, where perceptual fluency in the beer label was manipulated using two elements, complexity and colour contrast. Each element contained two levels: high vs. low. Complexity is manipulated by the amount of details and figures (i.e. hop flowers) that are present in the label. Colour contrast is simply manipulated by changing the contrast level in the label, as can be seen in Table 1 and Appendix I. The labels were designed with the help of a professional designer and the name of the craft beer is fictional. All the other design elements of the label are the same. The manipulation was checked and measured using a 7-point Likert scale by asking the participant about the by them perceived level of complexity: “I find the design of the beer
label...” 1 = very simple to 7 = very complex. This was also the case for perceived colour
between-22 subjects design, as can be seen in Table 1. Participants were only exposed to their assigned conditions when they were answering the scales regarding taste intensity, purchase intention and attractiveness. Simple Complex H igh c ol ou r c on tr as t L ow c ol ou r c on tr as t
23 After a short introduction, participants were asked questions about their consumption and preference for craft beer, as can be seen in Appendix II. Then Centrality of Visual Product Aesthetics was measured first. Also known as ‘the overall level of significance that visual aesthetics hold for a consumer in his relationships with products.’ At this point, participants were not exposed to their assigned beer label yet. This was done to avoid that their
sensitiveness for aesthetics maybe became triggered by being exposed to visual stimuli. CVPA was measured using a scale that measures three dimensions and consists of eleven items. Consumers that score high on CVPA are more concerned with the visual aesthetics of a product. The scale was developed (N = 318) by Bloch et al. (2003) and was measured using a 7-point Likert scale rating from 1 = strongly disagree to 7 = strongly agree.
The attractiveness construct was measured using a scale from Argo and Main (2008). The scale to measure attractiveness consists of three items which were measured using a 7-point Likert scale ranging from 1 = strongly disagree to 7 = strongly agree. Participants had to rate the following statements: “1) This label seems attractive to me. 2) This label is good-looking.
3) This label seems appealing to me.” Then, their purchase intention for their assigned label
condition was measured using a 7-point Likert scale item from Kozup, Creyer and Burton (2003): “How likely would you be to purchase the product, given the information shown?” Purchase intention and perceived taste intensity were the dependent variables on which the influence of perceptual fluency was measured. Perceived taste intensity was measured using a scale from Becker et al. (2011) and Marks and Bartoshuk (1979). Participants had to rate three items: mild (reversed), sharp and bitter. It’s possible to combine these items as well as using them separately. The items are measured using a 7-point Likert scale ranging from 1 = not at all to 7 = very. Participants were also asked to rate perceived sweetness and sourness. An overview of the entire survey that was used to collect the data for this research can be found in Appendix II.
24 3.3 Data analysis
Data was analysed using the SPSS 24 software package. After collecting the data using Qualtrics, the data was exported into SPSS for data cleansing and further analysis. The validity and reliability of the scales were assessed with the help of factor analysis and a Cronbach’s Alpha test. This was done in order to test whether the items that measured the different constructs truly represented the underlying constructs.
The effects of colour contrast and complexity, the independent variables, on the dependent variables were tested in two different ways. First, a two-way ANOVA was conducted to test whether differences existed between the simple vs. complex and low-contrast vs. high-contrast conditions. In cases were no significant results were found, the relationship between the independent and dependent variables was assessed using a Pearson correlation. With perceived complexity and perceived colour contrast, both continuous, as the independent variables. Since how participants perceived the manipulations was also measured. This way, it was still possible to make implications about the relationships between variables even when participants did not find the conditions significantly different from each other.
The mediation effect of attractiveness was tested by using the Preacher and Hayes' bootstrap method, employing the SPSS macro PROCESS (Model 4) developed by Hayes (2016). The moderating effect of Centrality of Visual Product Aesthetics was tested using linear
25 4. RESULTS
4.1 Descriptive statistics
A total of 131 participants finished the experiment. However, one participant was not an adult yet and was therefore deleted from the dataset. Of all the participants, 79 were male and 51 were female. The age of all the participants ranged from 18 till 69 with an average age of 28.5 years, SD = 9.21. Over 91 percent of the participants had consumed craft beer at least once in their life while a total of 78.5 percent enjoyed drinking craft beer. 68.5 percent of the
participants drink craft beer at least once per month. Most participants, 73.9 percent, had either an HBO or University Bachelor degree, or a University Master’s degree.
4.2 Factor analysis
To test whether the hypothesised constructs existed in the dataset and the items in the experiment represented the construct in the model, a factor analysis was conducted. This allowed the data to be properly inspected and prepared for the final analysis. If the items represent the underlying construct, they can be combined into one variable. This results in more parsimony and less multicollinearity for the final analysis. With the use of the Kaiser-Meyer-Olkin (KMO) test and the Bartlett’s test of sphericity, one can assess whether a factor analysis is appropriate. For the KMO test, a score higher than 0.5 is needed to proceed with the factor analysis. Furthermore, the Bartlett’s test must be significant (p < .05) and the
26 Table 2: Factor analysis. The above table illustrates the results from the factor analysis and the Cronbach Alpha scores from the scales that measure the constructs.
Table 2 shows the results of the factor analysis and the Cronbach Alpha scores of all the constructs. Conducting a factor analysis was appropriate in all cases. All the different items loaded properly onto their designated construct. Cronbach Alpha scores for testing reliability were sufficient for every construct except for perceived taste intensity (0.57). Although the factor analysis provides evidence that the construct was indeed measured in the dataset.
4.3 Manipulation checks
The manipulations were checked after the factor analysis. There were four conditions during the experiment: simple – high contrast, complex – high contrast, simple – low contrast and complex – low contrast. Participants were randomly exposed to one condition. Both perceived level of complexity and contrast were also measured in the experiment.
The overall means of the conditions (i.e. simple vs. complex and low vs. high colour contrast) were calculated, so that the effects of the different conditions could be tested. As can be seen in Figure 2, the simple condition was perceived as less complex than the complex condition. To test whether this difference was significant, an independent samples t-test was conducted. Levene’s test for equality of variances indicated that the variance between the two conditions was unequal, F = 4.22, p = .04. Therefore a t-test that does not presuppose equality of
variances was used. This test showed that there was a marginally significant difference
between the two experimental conditions t(123) = -1.74, p = .08. Thus, participants’ perceived the complex condition, M = 3.15, SD = 1.46, as more complex than the simple condition, M = 2.75, SD = 1.16. Though this difference is not truly significant.
Although the complex label was perceived as being more complex than the simple label, the complex label was not per se perceived as actually complex. Because only 22.7 percent, in both low and high-contrast conditions, of the participants who were exposed to the complex label rated the label as being actually complex (Appendix IV). The majority of the
participants rated the complex label as being simple, or neither simple nor complex.
Construct Source KMO Bartlett's test of Sphericity Communalities Factor loadings CA
27 Figure 2 also shows a difference between the low versus high colour contrast conditions. To test whether this difference was significant, an independent samples t-test was conducted. Levene’s test for equality of variances indicated that the variance between the two groups was equal, F = .12, p = .73. Thus, a t-test that does not presuppose equality of variances was used. This test showed that there was a significant difference between the two experimental
conditions t(128) = 2.24, p = .03. Therefore, participants perceived the low-contrast, M = 3.46, SD = 1.46, to be of less colour contrast than the high-contrast condition, M = 4.03, SD = 1.47.
Figure 2: Means of the different condition. The above figure illustrates the means of the conditions in terms of visual complexity (blue) and colour contrast (yellow). A higher mean means more complexity or more colour contrast.
Furthermore, perceived colour contrast and perceived level of complexity significantly correlate with each other. The more colour contrast a manipulation has, the lower the
perceived level of complexity, r = -.25, p = .00. Hence the manipulation will be perceived as more simple instead of complex when the amount of colour contrast is high.
4.4 Hypothesis testing
4.4.1 Direct effects on purchase intention
To analyse the influence of the simple versus complex and the low versus high-contrast conditions on purchase intention, a two-way ANOVA was conducted. The conditions did not have significant effects on purchase intention. Neither was there an interaction effect between
0 1 2 3 4
Simple Complex Low contrast High contrast
28 the conditions of both manipulations, F(3, 1) = .19, p = .66. Simple versus complex condition,
F(1, 126) = .33, p = .57, low versus high contrast F(1, 126) = .02, p = .90. Thus the
conditions, to which the participants were exposed, did not affect respondents purchase intention.
Hypothesis 1a: Higher colour contrast in craft beer labels is positively related to purchase intention.
To analyse whether perceived colour contrast and purchase intention correlate with each other, a Pearson correlation was used. There was a significant positive correlation between perceived colour contrast and purchase intention, r = .21, p = .02. Meaning that participants were more willing to purchase a craft beer when colour contrast is higher. This is in line with hypothesis 1a.
Hypothesis 1b: Complexity in craft beer labels is negatively related to purchase intention.
Furthermore, to analyse whether perceived complexity and colour contrast correlate with purchase intention, a Pearson correlation test was used. There was a significant positive correlation between complexity and purchase intention, , r = 0.23, p = 0.01, showing that when a label is perceived as more complex, purchase intention for the craft beer will be higher. This is not consistent with hypothesis 1b.
4.4.2 Direct effects on taste intensity
To test the influence of the low and high colour contrast conditions and the simple vs. complex conditions on perceived taste intensity in general, a two-way ANOVA was
conducted. The test indicated that there were no significant differences between the conditions and taste intensity. There was also no significant interaction between the manipulations of both conditions. However, significant results were found for some separate taste dimensions. A two-way ANOVA was conducted to test the influence of low vs. high colour contrast and simple vs. complex design on perceived bitterness. There was a significant interaction effect between colour contrast and complexity, F(1, 126) = 3.97, p = .05, meaning that in this case the effect of the conditions depend on each other. Thus the main effects cannot be tested meaningfully. Therefore, both independent variables were tested separately using a one-way ANOVA. Both tests did not yield significant results.
29 Significant results were found between the simple vs. complex condition, F(1, 126) = 4.47, p = .04, meaning that complexity in craft beer labels influences perceived mildness. Compared to a simple craft beer label, a complex label will be perceived as having a more mild taste.
Hypothesis 2a: Higher colour contrast in craft beer labels is positively related to perceived taste intensity.
Furthermore, a correlation analysis showed that perceived colour contrast and taste intensity did significantly correlate, r = .24, p = .01, showing that higher perceived contrast in craft beer labels seems to be positively related to perceived taste intensity. More specifically, colour contrast significantly correlated positively with perceived mildness, r = .21, p = .02, and bitterness, r = .22, p = .01. Thus when colour contrast in a craft beer label is high, the taste of that craft beer label will be perceived as being stronger. These result are in line with hypothesis 2a.
Hypothesis 2b: Complexity in craft beer labels is positively related to perceived taste intensity.
The results of the two-way ANOVA showed previously that complexity in craft beer labels can enhance perceived mildness. In addition, a correlation analysis with perceived complexity and the different taste dimensions was conducted. The analysis showed some marginally significant results. There were marginally positive correlations between complexity, perceived sharpness, r = .17, p = .07, and sweetness, r = .15, p = .09. Showing that beer labels with more complex design are perceived as being more sharp in taste and more sweet. Though these effects are not truly significant and therefore they do not fully support hypothesis 2b.
4.4.3 Effects involving attractiveness
30
Hypothesis 3a: Higher colour contrast in craft beer labels is positively related to attractiveness. Hypothesis 3b: Complexity in craft beer labels is negatively related to attractiveness.
To analyse whether perceived complexity and perceived colour contrast correlate with
attractiveness, a Pearson correlation test was used. There was a significant positive correlation between colour contrast and attractiveness, r = .18, p = .04. This means that labels with higher colour contrasts are perceived as being more attractive. This is in line with hypothesis 3a. There was also a significant positive correlation between label complexity and attractiveness,
r = .25, p = .01, showing that when the design of a label is perceived as more complex, it will
be perceived as being more attractive. This is not in line with hypothesis 3b.
Hypothesis 3c: A beer label’s attractiveness is positively related to purchase intention.
To test whether the attractiveness of beer labels is associated with purchase intention, a Pearson correlation test was conducted. There was a significant positive correlation between attractiveness and purchase intention, r = .59, p = .00, which means that participants were more willing to buy a craft beer when the label of a craft beer is more attractive. Thus supporting hypothesis 3c.
Hypothesis 3d: A beer label’s attractiveness is positively related to perceived taste intensity.
To test whether attractiveness is positively associated with perceived taste intensity, a Pearson correlation test was conducted. There was no significant correlation between attractiveness and perceived taste intensity, r - .01, p = .88, meaning that attractiveness of craft beer labels is not related to perceived taste intensity. However, a significant positive correlation between attractiveness and perceived sweetness, a taste dimension, was found, r = .20, p = .02. Hence, when a label is more attractive, the taste is perceived to be sweeter. Thus, partially supporting hypothesis 3d.
Hypothesis 3e: Attractiveness mediates the effect of colour contrast and complexity in beer labels on purchase intention.
31 the independent variables, purchase intention as the dependent variable, and attractiveness as the proposed mediator. The analysis showed that the bias-corrected 95% confidence interval (5,000 bootstrap samples) for the indirect effect of complexity through attractiveness did not include zero (.0391 to .2603). This is consistent with the interpretation that purchase intention caused by complexity of the beer label design indeed stemmed from attractiveness. Another analysis showed that the confidence interval for colour contrast did not include zero either (.0044 to .1966). This is consistent with the interpretation that purchase intention caused by the colour contrast of the beer label design indeed stemmed from attractiveness. Both mediation effects were positive and thus supporting hypothesis 3e
Hypothesis 3f: Attractiveness mediates the effect of colour contrast and complexity in beer labels on perceived taste.
To test whether attractiveness mediated the differential effect of perceived complexity and perceived colour contrast on perceived taste intensity, the Preacher and Hayes' bootstrap method for testing mediation was used, employing the SPSS macro PROCESS (Model 4) developed by Hayes (2016). Complexity and colour contrast were entered (individually in separate tests) as the independent variables, perceived taste intensity as the dependent variable, and attractiveness as the proposed mediator. The analysis showed that the bias-corrected 95% confidence interval (5,000 bootstrap samples) for the indirect effect of
complexity through attractiveness did include zero (-.0438 to .0302). Meaning that there is no significant effect of perceived complexity on perceived taste intensity through label
attractiveness. Another analysis showed that the confidence interval for colour contrast did also include zero (-.0352 to .0135). Thus there was no significant effect of perceived colour contrast on perceived taste intensity through attractiveness either. These results are not in line with hypothesis 3f.
4.4.4 Effects involving Centrality of Visual Product Aesthetics
Hypothesis 4a: The effect of colour contrast and complexity in craft beer labels on attractiveness is greater for people with a higher CVPA.
32
Hypothesis 4b: The effect of colour contrast and complexity in craft beer labels on purchase intention is greater for people with a higher CVPA.
To test whether participants’ centrality of visual product aesthetics, a person’s sensitivity for product aesthetics, moderated the effect of colour contrast in labels on purchase intention, a linear regression analysis was conducted. The regression proved to be significant, R² = .08,
F(3, 126) = 3.63, p = .02, thus the model is valid. Although the direct effects of CVPA and
colour contrast on purchase intention were significantly positive, no moderation effect could be found, as can be seen from Table 3. Furthermore, a linear regression was used to analyse whether CVPA moderated the effect of complexity on purchase intention. However, no significant moderation effects were found. Thus hypothesis 4b cannot be supported.
B SE B t p Purchase intention .09 1.58 .06 .96 CVPA .69 .34 .204 .04 Colour contrast .78 .40 1.95 .05 Interaction CVPA*CC -.13 .08 1.52 .13 R² = .08
Table 3: Role of Centrality of Visual Product Aesthetics on the relationship between colour contrast and purchase intention.
Hypothesis 4c: The effect of colour contrast and complexity on perceived taste intensity is greater for people with a higher CVPA.
33 B SE B t p Taste intensity 4.86 1.08 4.52 .00 CVPA -.34 .23 -1.49 .14 Colour contrast -.42 .27 -1.54 .13 Interaction CVPA*CC .12 .06 2.15 .03 R² = .10
Table 4: Moderation effect of CVPA on the relationship between colour contrast and taste intensity.
Figure 3 shows a visualisation of the moderating effect of, low and high, Centrality of Visual Product Aesthetics on the relationship between, low and high, colour contrast and perceived taste intensity. As can be seen from Figure 3, when an individual’s sensitivity for visual product aesthetics is high, taste expectations are perceived to be stronger than when an individual has a low sensitivity for visual product aesthetics. Furthermore, perceived taste intensity when colour contrast is low will be less for individuals with high CVPA than for individuals with low CVPA. The significant interaction effect, as can be seen in Table 4, is also shown in Figure 3 as both lines are not parallel.
Figure 3: Graph illustrating the moderating effect of CVPA on the relationship between colour contrast and perceived taste intensity.
0 1 2 3 4 5
Low CVPA High CVPA
34
Hypothesis 4d: The effect of attractiveness on purchase intention is greater for people with a higher CVPA.
To test whether CVPA moderates the effect of attractiveness on purchase intention, a linear regression analysis was conducted. The regression proved to be significant, R² = .37, F(2, 127) = 37.42, p = .00, meaning that the model is valid. Attractiveness had a significant
positive effect on purchase intention, B = .32, t(129) = 2.13, p = .04, showing that participants were more willing to buy the craft beer when the label was more attractive. Furthermore, the effect was positively moderated by an individual’s sensitivity for visual product aesthetics (CVPA), B = .06, t(129) = 2.18, p = .03, which means that CVPA strengthens the effect of attractiveness on purchase intention. Thus, individuals with a higher sensitivity for visual products aesthetics are more willing to purchase products when they are more attractive, which is in line with hypothesis 4d.
Figure 4: Graph illustrating the moderating effect of CVPA on the relationship between colour contrast and perceived taste intensity.
Figure 4 shows the interaction effect, as described previously, between an individual’s
sensitivity for visual products aesthetics and the attractiveness of craft beer labels on purchase intention. Individuals who are more sensitive for the aesthetics of products are more intended to purchase a craft beer when they find the label attractive than individuals with low
sensitivity for product aesthetics. In addition, when aesthetic sensitive individuals rate the label low in attractiveness, they are less intended to purchase the craft beer than individuals who are less sensitive for visual product aesthetics.
0 1 2 3 4 5 6
Low CVPA High CVPA
35 4.5 Overview of the findings
Hypothesis Significant Remark
H1a. Higher colour contrast in craft beer labels is positively related to purchase intention.
Yes Supported
H1b. Complexity in craft beer labels is negatively related
to purchase intention.
Yes Not supported. Effect is
opposite of what was expected.
H2a. Higher colour contrast in craft beer labels is
positively related to perceived taste intensity.
Yes Supported
H2b. Complexity in craft beer labels is positively related to
perceived taste intensity.
Marginally Partially supported for sweetness and sharpness.
H3a. Higher colour contrast in craft beer labels is
positively related to attractiveness.
Yes Supported
H3b. Complexity in craft beer labels is negatively related
to attractiveness
Yes Not supported. Effect is
opposite of what was expected.
H3c. A beer label’s attractiveness is positively related to
purchase intention.
Yes Supported
H3d. A beer label’s attractiveness is positively related to
perceived taste intensity.
Yes Supported, but only for sweetness.
H3e. Attractiveness mediates the effect of colour contrast
and complexity in beer labels on purchase intention.
Yes Supported
H3f. Attractiveness mediates the effect of colour contrast
and complexity in beer labels on perceived taste.
No Not supported
H4a. The effect of colour contrast and complexity in craft
beer labels on attractiveness is greater for people with a higher CVPA.
No Not supported
H4b. The effect of colour contrast and complexity in craft
beer labels on purchase intention is greater for people with a higher CVPA.
No Not supported
H4c. The effect of colour contrast and complexity on
perceived taste intensity is greater for people with a higher CVPA.
Yes, for contrast
Partially supported
H4d. The effect of attractiveness on purchase intention is
greater for people with a higher CVPA.
Yes Supported
36 5. DISCUSSION
The aim of this research was to provide more clarity and insights on how visual design elements in product packaging influence consumer behaviour and taste expectations. More specifically, about how perceptual fluency has an influence on the perceived attractiveness, perceived taste and purchase intentions of consumers. Two dimensions and visual design elements that facilitate perceptual fluency were chosen to study its effects on taste perception and purchase intention for craft beer, based on their labels. The first one being visual
complexity (and the opposite end of the continuity: simplicity). The second one being colour contrast. An experiment with four labels, low/high colour contrast vs. simple/complex design, was conducted to test the effects on the dependent variables. The conditions themselves did not result in significant effects in most cases, but significant effects were often found with perceived complexity and perceived colour contrast.
5.1 Conclusions
First of all, perceived colour contrast seems to be positively related with purchase intention for craft beer. Higher colour contrast in craft beer labels is associated with higher purchase intention for craft beer. This could be the result of improved fluency since higher contrast levels enhance perceptual fluency, which can result in an increased preference for the object (Reber et al, 1998; Rebert et al. 2004). Furthermore, perceived colour contrast is also
positively related to perceived taste intensity. Thus, taste expectations are perceived as being more strong or intense when colour contrast in craft beer labels is higher. In addition,
perceived colour contrast was also positively associated with the attractiveness of craft beer labels. Craft beer labels with high colour contrast are perceived as being more attractive. Thus perceptual fluency in craft beer labels, facilitated by means of high colour contrast, is
positively associated with consumer responses and taste expectations.
37 research showed that complexity in brand names could intensify taste expectations for
hedonic products (Mantonakis et al., 2013). Furthermore, perceived complexity was also positively related to attractiveness of craft beer labels. Thus complex elements in the design of craft beer labels seem to be positively related with the attractiveness of the craft beer label. This association was also the opposite of what had been expected. One reason why
complexity seems to be positively related to purchase intention and attractiveness can be explained with the help of Berlyne’s (1971) model of arousal. This model suggests that there is an ideal point in terms of complexity in visual design and liking or arousal (i.e. inverted U-shape relationship). More evidence exists that, depending on the circumstances, a moderate level of complexity in design is more preferred (Palmer et al, 2013). Like in car design for example (Landwehr et al., 2011). According to the results of this research, this also seems to be the case for the design of craft beer labels. Thus although complexity negatively affects perceptual fluency, a moderate level of complexity in craft beer labels seems to be preferred. However, colour contrast also decreased perceived complexity making a design less complex when colour contrast is high (i.e. simpler and easier to perceive). Therefore it might be possible that, as long as colour contrast is high, complexity to a certain point does not
immediately infer that the design will be perceived as less fluent. It could also be the case that the difference between the simple, M = 2.75, SD = 1.16, and complex, M = 3.15, SD = 1.46, conditions was not large enough to make a difference in perceived fluency. Furthermore, 77.3 percent of the participants exposed to the complex condition rated the label as simple, or neither simple nor complex. Thus most participants found the design of the ‘complex’ label simple and thus easy to perceive, which enhances perceptual fluency.
As expected, consumers are more likely to purchase a craft beer when they find the label of a craft beer attractive. Attractiveness also mediated the effects of colour contrast and
complexity on purchase intention. Attractiveness did not mediate the effect of colour contrast and complexity on perceived taste intensity. A direct effect of attractiveness on one taste dimension was found. Attractive beer labels are perceived as being sweeter, but direct effects on other taste dimensions could not be found.
38 moderator in the relationships between colour contrast and perceived taste intensity, and attractiveness and purchase intention. CVPA strengthened the positive relationship between perceived colour and perceived taste intensity. Thus, individuals who are more concerned with the visual design of products have even stronger taste expectations, compared to individuals who have low CVPA, when colour contrast in a craft beer label is high. When colour contrast is low, the perceived taste intensity will be lower for individuals with high sensitivity for visual product aesthetics than for those who are less sensitive to visual aesthetics. Individuals with high CVPA are also more likely to purchase a craft beer when they find the label of the craft beer attractive, because CVPA strengthens the relationship between colour contrast and purchase intention. They are also less intended, compared to individuals with low CVPA, to purchase a craft beer when they find the label of the craft beer not or less attractive.
Overall, one can conclude that perceptual fluency plays a role in the consumer decision making process when it comes to evaluating and purchasing craft beer. Perceptual fluency, deriving from high colour contrast, in craft beer labels seems to be positively associated with stronger taste expectations, which is important in the evaluation process when somebody wants to purchase craft beer. Furthermore, consistent with literature, consumers find labels with higher colour contrasts more attractive. Perceptual fluency, facilitated by high colour contrast, also seems to be positively associated with consumer purchase behaviour directly. Furthermore, a moderate level of complexity in the design of craft beer labels seems to be preferred by consumers as it seems to be positively associated with their evaluation of the craft beer, in terms of taste expectations and attractiveness, and their purchase behaviour. Therefore, it could be that a moderate level of complexity in the design of a label does not necessarily have to mean that the label will immediately be perceived as being less fluent. At least when complexity is manipulated through the amount of information (e.g. shapes) that is displayed in the label and when colour contrast remains high.
5.2 Managerial implications
39 visual design of their products and labels of their craft beers in order to attract consumers and to stand out from the competition. Based on findings of this research, marketers are advised to design their labels in a way so that they have high colour contrasts, since this seems to be positively associated with the attractiveness of beer labels, purchase intention and the strength of taste expectations. This is especially interesting for craft beers that are known and often bought for their distinct and intense taste like pale ales, IPA’s, English bitters and strong golden ales. Furthermore, consumers seem to be attracted to labels that have a moderate level of complexity in them. So brewers are also advised to add elements that make the design to be perceived as a little more complex by consumers. However, as theory suggests, they need to keep in mind that there might be an ideal point for the preferred level of complexity. When the label becomes too complex, it might negatively influence perceptual fluency and could therefore negatively influence consumer evaluations of the product. Marketers, managers and brewers can overcome this by having consumers evaluate the design of craft beer labels before putting it into the market.
5.3 Scientific implications
This research provides insights on how visual design elements, and a consumer’s processing of those elements, in product packaging is related to consumer behaviour and taste
expectations. More specifically, about how perceptual fluency, in terms of
simplicity/complexity and colour contrast, in craft beer labels is related to the perceived attractiveness of labels and taste expectations and purchase intentions of consumers. Research by Mantonakis et al. (2013) showed that, for hedonic products, difficult to pronounce or difficult to read brand names (i.e. disfluent) strengthened expected taste sensations. This study provides evidence that this is also the case when fluency is controlled for in terms of
complexity in the design itself, although this was not entirely expected at the start of this research. It is consistent with Palmer et al. (2013) findings that consumers sometimes prefer a moderate level of complexity. And is an indication that there could indeed be an ideal point for a preferred level of complexity, as Berlyne (1971) suggested. Although this is not the case when design becomes more complex by decreasing colour contrasts, but only when
complexity is manipulated through the amount of information that is displayed in the label.
5.4 Limitations
40 implementing these findings since individuals with an average education level or older
individuals, who could be an interesting target group for brewers as well, are
underrepresented. Furthermore, the four labels that were used during the experiment did not have significant direct effects when the analyses were conducted. Thus taste expectations for the labels that should be more fluent did not significantly differ from the effect of the low fluency labels. One reason for this could be that the manipulation did not work entirely as planned. For example; The difference between the simple condition and the complex condition was only marginally and thus not truly, significant. In addition, the complex label was not perceived as being actually complex by most participants. Which could mean that the complex condition was not perceived as actually being less fluent than the simple label condition. Furthermore, most significant results were found by using perceived complexity and perceived colour contrast in the analyses. These were manipulation checks to see whether conditions actually differed in terms of complexity and colour contrast according to the participant. Although these variables can also be seen as reliable representations, since they explain how each individual exactly perceived the conditions and was affected by them. This way, individual differences in how visual stimuli and objects are processed, and how much effort one has to use to process, can be taken into account. In addition, taste intensity was also measured in terms of perceived taste intensity and thus not as in actual taste. How individuals perceive taste and actually experience real taste might differ. Therefore, one has to be
cautious with the findings of this research regarding implications about consumers’ actual taste experiences.
5.5 Suggestions for future research
For future research, it would be interesting for designers, marketers, managers and firms if a model was developed which could help in determining the ideal point of complexity and liking for different types of products. More research is also needed to find out in what
41 REFERENCES
Allison, R. I., & Uhl, K. P. (1964). Influence of beer brand indentification on taste perception.
Journal of Marketing Research, 1, 36–39.
Argo, J.J., & Main, K. J. (2008). Stigma by Association in Coupon Redemption : Looking Cheap because of Others, Journal of Consumer Research, 35(4), 559-572.
Barnett, A., & Spence, C. (2016). Sci Forschen Nutrition and Food Technology : Assessing the Effect of Changing a Bottled Beer Label on Taste Ratings, 1–4.
Batra, R., Seifert, C., & Brei, D. (2015). The psychology of design: creating consumer
appeal(1st ed.). Abingdon (UK): Routledge.
Bearden, W., Haws, K., & Netemeyer, R. (2011). Handbook of marketing scales. Los Angeles: Sage.
Becker, L., van Rompay, T. J. L., Schifferstein, H. N. J., & Galetzka, M. (2011). Tough package, strong taste: The influence of packaging design on taste impressions and product evaluations. Food Quality and Preference, 22(1), 17–23.
Berlyne, D. (1971). Aesthetics and Psychobiology. New York, N.Y.: Appleton Century Crofts.
Bloch, P. H., Brunel, F. F., & Arnold, T. J. (2003). Individual Differences in the Centrality of Visual Product Aesthetics Individual differences in the centrality of visual product aesthetics : Concept and Measurement. Journal of Consumer Research 29(4), 551-565. Bloch, P. H. (1995). Seeking the Ideal Form: Product Design and Consumer Response.
Journal of Marketing, 59(3), 16-29.
Centraal Bureau voor Statistiek. Aantal bierbrouwers meer dan verviervoudigd sinds 2007. (2018). Cbs.nl. Retrieved 7 March 2018, from
