The impact of nature’s scale invariance on consumer behavior:

The impact of nature’s scale

invariance on consumer behavior:

A new marketing trick?

The effect of spatial frequencies in visual

advertisements on consumer behavior

by

Lidwien Mentink

- 2 -

The impact of nature’s scale

invariance on consumer behavior:

A new marketing trick?

The effect of spatial frequencies in visual

advertisements on consumer behavior

by

Lidwien Mentink

University of Groningen

Faculty of Economics and Business

Master Thesis

11 January 2016

- 3 -

EXECUTIVE SUMMARY

To date, research in the fields of marketing and psychology about the concept of low-level visual properties has concentrated on variables like complexity, symmetry and contrast. Research about the properties of fractal geometry has focused up to now on the creation and liking of art and architecture. The aim of this research was to explore, for the first time, the impact of fractal-like properties in relation to marketing. We aimed to examine whether the level of spatial frequencies in a visual advertisement influences the intention to purchase and willingness to recommend the product. We have taken into account different consumer motivations, and investigate the effect of fractality when utilitarian versus hedonic motivated. For both independent variables, the relationship is tested for mediation by aesthetic liking. The moderating effect of the consumers’ construal level is tested as well.

The relationships were tested by a 3 (high, intermediate and low spatial frequency) x 2 (hedonic and utilitarian motivated) between-subjects design. A questionnaire with the 6 different conditions was distributed through Facebook among a group of 231 respondents. The questionnaire started testing the respondents’ construal level (e.g., the extent to which people’s thinking is abstract or concrete), by asking to choose between two different descriptions of an activity. Next, the respondents had to rate a T-shirt advertisement on several variables.

- 4 -

level is found reversed significant, compared to the stated hypothesis. The results show that the higher the construal level (i.e., the more abstract people think), the stronger the effect of aesthetic liking on the ITP.

In view of the still lacking research about fractal-like properties, further research should be designed to evaluate the effects of fractal-like properties in multiple ways. It appears to be necessary to carry out studies in real-life, under different conditions and with the change in fractal-like properties on the product itself.

- 5 -

PREFACE

- 6 -

TABLE OF CONTENTS

1.1 Introduction of the Topic ... 9

1.2 Contribution ... 10

1.3 Structure ... 11

-2. Theoretical Framework ... 12

2.1 Fractal Geometry ... 12

Fractal measurement techniques. ... 13

Dvalue ... 13

2.2 Aesthetic Liking ... 15

Variables ... 15

Perspectives ... 16

Complexity and symmetry ... 16

-- 7 --

Hypothesis 4 ... 23

Hypothesis 5 ... 24

-3. Research Design ... 25

3.1 Level of Fractallike Properties ... 25

3.2 Consumer Motivations ... 26 3.3 Aesthetic Liking ... 27 3.4 Consumer Behavior ... 27 Intention to purchase ... 27 Willingness to recommend ... 27 3.5 Construal Level ... 28 3.6 Questionnaire ... 28 -4. Results ... 30 4.1 Descriptive Statistics ... 30 4.2 Figureground Contrast ... 30 4.3 Main Effects ... 31 DV: Intention to purchase ... 31 DV: Willingness to recommend ... 32

4.4 Effect on Aesthetic Liking ... 33

4.5 Correlations of Dependent Variables ... 34

4.6 The Moderation Effects of Type of Motivation ... 34

4.7 The Mediating Effects of Aesthetic Liking ... 35

4.8 The Moderating Effects of the Consumers’ Construal Level ... 35

DV: Intention to purchase ... 35

DV: Willingness to recommend ... 36

4.9 Hypothesis Testing ... 38

-4 Discussion, Limitations & Implications ... 39

-- 8 --

The main effect ... 39

The mediating effects of aesthetic liking ... 40

The moderation effects of type of motivation ... 40

5.3 Limitations ... 41

5.4 Practical Implications and Future Research ... 42

References ... 43

Appendices ... 49

Appendix A – Questionnaire ... 50

-- 9 --

1.

INTRODUCTION

1.1 Introduction of the Topic

Look out your window. Your daily view will probably have little affinity with the great depths of the Grand Canyon, the overwhelming waterfalls in North America or just a beautiful sunset. Strangely enough, this is wrong. Most scenes, whether gorgeous or ordinary, display an enormous amount of similarity (Olshausen & Field, 2000). Natural scenes are often characterized by roughness, irregularity and complex structures, in contrast to the smoothness of many human-made objects.

The unique complexity of nature necessitates the use of a geometry that is radically different than the Euclidian geometry (Spehar, Clifford, Newell, & Taylor, 2003). Mandelbrot (1983) developed ‘fractal geometry’, to account for the complexity, irregularity, and fractured character of many natural phenomena, shapes and scenes. In contrast to the smooth shapes of the Euclidian geometry, fractals consist of patterns that recur on finer and finer scales (Redies, Hasenstein, & Denzler, 2007; Taylor, Spehar, van Donkelaar, & Hagerhall, 2011).

To date, marketing research on processing low-level visual properties has concentrated on visual variables like complexity, symmetry and contrast (Berlyne, 1974; Schwarz & Winkielman, 2004). Research on fractal-like properties has focused on the creation and liking of art (e.g., abstract paintings) (Taylor et al., 2011) and architecture (Joye, 2010). This research wants to further explore the impact of fractals in relation to marketing, since the evidence for an effect of fractal shapes on consumer behavior is very rare. The current research tries to examine whether fractal-like properties in advertisements will lead to more positive consumer behavior in regards to intention to purchase and willingness to recommend.

- 10 -

2004). The more fluently perceivers can process an object, the more positive their aesthetic response (Reber, Schwarz, & Winkielman, 2004). We can decide almost instantaneously what we like, and are highly consistent in our assessments, even across cultures (Faerber, 2012).

Besides looking at the relationship between fractal-like properties and consumer behavior, this study tries to examine whether this relationship is moderated by consumer motivations. Research has demonstrated that the motivation for consumers to shop can be on the one hand functional (e.g., utilitarian), whereas one the other hand the consumer desires to satisfy emotional hedonic needs (Kim, 2006; Arnold & Reynolds, 2003). In the utilitarian view, consumer behavior is directed toward satisfying a functional or economic need. The hedonic view suggests that consumption is driven by fun, amusement, fantasy and sensory stimulation (Babin, Dardin, & Griffin, 1994). The need for hedonic shopping is essentially aesthetic in nature, and therefore expected to have a greater impact on the aesthetic liking of the advertised product (Holbrook & Hirschman, 1982).

Considering the fact that individuals see and interpret the world in different ways, a second moderator in our research is the consumer’s construal level. Human actions can be represented at many different levels of abstraction, from ‘having dinner’ to the component act of ‘ordering’ or even ‘using the cutlery’. Numerous variables can influence at which level of abstraction an activity is represented. High level construal implies thinking abstract, whereas low level construal implies thinking more concretely and in detail (Schwarz, 2006; Trope & Liberman, 2010). Fajita, Trope, Liberman and Levin-Sagi (2006) found that individuals thinking at a high construal level have a greater tendency to make decisions reflecting self-control, than those at low levels. A low- or high level of self-control can be decisive on the effect of consumer behavior. Therefore, the effect of aesthetic liking on consumer behavior may be influenced by the consumer’s construal level.

1.2 Contribution

- 11 -

processes (Faerber, Leder, Gerger, & Carbon, 2010). The results will provide marketers with guidelines on how to optimize advertisements and eventually to generate more revenues. Second, this study will help developers to understand the effects of different construal levels on consumer behavior.

1.3 Structure

- 12 -

2.

THEORETICAL FRAMEWORK

The primary objective of this study is to explain fractal-like properties, aesthetic liking, consumer behavior, consumer motivations and construal level to model the relationship between these notions. The study will test the effect of fractal-like properties on consumer behavior, mediated by aesthetic liking. In other words, the study will test the effect of fractal-like properties via aesthetic liking on consumer behavior. The effect of fractal-like properties on consumer behavior will be studied during different shopping motivations, namely hedonic versus utilitarian. Furthermore, this research tries to examine the effect of aesthetic liking on consumer behavior, moderated by the consumer’s construal level.

Before starting to explain the conceptual model, some key terms will be explained, e.g., fractal geometry, consumer behavior, aesthetic liking, consumer motivations and construal level, in order to set the context.

2.1 Fractal Geometry

Circles, straight lines, squares and cubes are examples of geometrical shapes central to Euclidean geometry. Shapes within Euclidean geometry are characterized by smoothness and regularity. According to Mandelbrot (1983), Euclidean geometry is inadequate for describing and modelling the immense complex geometric characteristics of natural forms and processes. These natural forms can be plants, shapes of mountains, coastlines, clouds or even galaxies. Mandelbrot succeeded in developing a systematic description of the mathematical language capturing the visual patterns of the natural world, called ‘fractal geometry’ (Joye, 2010).

- 13 -

repetition of identical shapes, but more often quasi-self-similar shapes (Joye, 2010). This form of scale symmetry means that any portion of a shape, when magnified in scale, would appear quasi-self-similar to the whole curve (Forsythe et al., 2010).

Fractal measurement techniques. There are various statistical measurements to

measure the level of fractal-like properties in an image. The dimension value, box counting technique and 1/f statistics are briefly explained.

D-value

In the classical Euclidean geometry, lines have a dimension of 1, squares and triangles are two-dimensional, and volumes in space have a dimension of 3. A coastline is a one-dimensional fractal (i.e., its topology is one one-dimensional), however the repeating patterns in this line cause it to spread across two-dimensional space. The fractal dimension of the coastline is expected to lie between 1 and 2 (e.g., 1.42). A mountain is a two-dimensional fractal (i.e., its topology is two-dimensional), though this surface spread across three-dimensional space. The fractal dimension of the mountain lies between 2 and 3 (Forsythe et al., 2010).

An important mathematical property of the fractal patterns is that D-value (dimension) is thus not an integer value, as it is the measure to which a fractal 'fills a space'. The fractal does not fill the entire plane (two-dimensional), but is more space filling than a simple line (one-dimensional). The D-value can be understood as a measure of the degree to which detail recurs on different scales of the structure (Forsythe et al., 2010; Joye, 2010). Figure 1 shows examples of fractal-based artificial visuals. The right image (Koch snowflake) demonstrates the effect of an increasing D-value. Unlike Euclidean shapes this object has detail at all levels. Figure 2 shows examples of fractals found in nature.

- 14 -

The fractal dimension D can be regarded as a preliminary indicator of complexity in a pattern. Images with a low D-value would remain smooth in appearance and are a measure of shallow complexity. High D values (those which approach the dimension of the embedding space) would however appear more coarse and complex (Braun, Amirshahi, Denzler, & Redies, 2013; Mureika & Taylor, 2013).

Box-counting technique

The fractional dimension, D-value, provides a way to measure the roughness or convolution of fractal curves. Referred to as box-counting, the technique is widely used because it can measure images that are not entirely self-similar. A digitized image of an object is covered with a computer-generated network of identical squares (boxes). The statistical scaling qualities of the pattern are then determined by calculating the proportion of squares occupied by the painted pattern and the proportions that are empty. This process is then iterated for networks with a range of square sizes (Forsythe et al., 2011; Spehar, 2003).

1/f Statistic

The scale invariance of natural scenes, a property related to the fractal-like structure, is reflected by the Fourier theory. The Fourier power falls with the spatial frequency ‘f’ by a factor of approximately ‘1/f’. Scale invariance implies that, as one zooms in and out of a natural scene, there is always an equivalent amount of “structure” (intensity variation) present (Redies, 2007; Olshausen & Field, 2000). 1/f2 characteristics have been found previously in images of natural scenes. These findings imply that natural scenes possess fractal-like properties (Koch, Denzler, & Redies, 2010). The spectral power of natural scenes falls with spatial frequencies according to a power law, 1/f p, with values of p near 2 (Redies et al., 2007). It has been shown that scale-invariant images with a Fourier slope of -2 are processed efficiently and fluently by the visual system, and that they were judged

- 15 -

as more beautiful (Menzel, Hayn-Leichsenring, Langner, Wiese, & Redies, 2015; Redies et al., 2007; Braun et al., 2013).

2.2 Aesthetic Liking

The current research will test the impact of the fractal geometry on consumer behavior. Until now, fractal geometry has established its usefulness for understanding of art and architecture (Taylor et al., 2011; Joye, 2010; Forsythe et al., 2010). Taylor (2002) found that Jackson Pollock was generating paintings with a high fractal dimension, and reported specific values with respect to aesthetic preferences for fractal dimensions. Empirical research on advertising indicates that aesthetic thoughts concerning advertising can affect attitudes towards a product or brand, and consequently influence the willingness to purchase that product or brand (Shimp, 1981; Mackenzie & Lutz, 1989). Aesthetic feelings are important predictors of general behavioral intent (Ha & Lennon, 2010). Assuming that fractality has an indirect effect on consumer behavior, this study will research the mediating effect of aesthetic liking on the relationship between the fractal-like properties on consumer behavior.

The word aesthetic is derived from the German ‘Ästhetisch’ or French ‘esthétique’, both from the Greek word ‘aisthetikos’, meaning sensitive (to feel) and perceptive (to perceive by the senses or by the mind) (Harper, 2015). The psychology of aesthetics aims to identify and describe the psychological mechanisms that allow humans to experience and appreciate a broad variety of objects in aesthetic terms (e.g., beautiful, attractive, ugly, sublime) (Leder & Nadal, 2014). The term is used to describe the perception and response to art, as well as interactions with objects or scenes that evoke an intense feeling (Chatterjee, 2011).

Variables. Research on aesthetic liking has focused on obtaining insights into

- 16 -

in Japan. Next to the many cultural differences in taste, there are also universal, cross-cultural aesthetic preferences and values. An evolutionary history of aesthetic tastes of Darwin will explain this universality. The experience of beauty is one component in a whole series of Darwinian adaptations, which people extend and intensify in the enjoyment of objects, artwork and entertainment (Dutton, 2010).

Faerber et al. (2010) conducted a meta-analysis about various factors constructing aesthetic appreciation. They assessed aesthetic appreciation through the following six key variables derived from the literature: attractiveness (also beauty and liking), arousal, interestingness, valence (as well as pleasantness), boredom and innovativeness (as well as novelty, originality and old/new). Faerber et al. demonstrated high internal consistency of the multidimensional construct of aesthetic appreciation.

Perspectives. What makes a face beautiful, a painting appealing, a design

pleasing, or a scenery charming? This question has been debated for at least 2,500 years and has been given a wide variety of answers. Through the years, theorists and researchers have tried to explain aesthetic value from several perspectives. According to an objective view, dating back at least to Plato, beauty was seen as a property of an object or image, that creates a pleasurable experience for every perceiver (Tatarkiewicz, 1970). Balance, proportion, symmetry, complexity, contrast and clarity were among the identified features (Berlyne, 1974). From the 1970’s on, the subjective view introduces itself and proposed that anything could be beautiful if it pleases the senses. According to this perspective, beauty is a function of the qualities of the perceiver. Expressions like ‘beauty is in the eye of the beholder’ reflected the historically changing and culturally relative nature of beauty (Kubovy, 2000). Modern research suggested an interactionist perspective, saying that a sense of beauty emerges from patterns in the way people and objects relate. According to this perspective, beauty is grounded in the processing experiences of the perceiver that emerge from the interaction of stimulus properties and perceivers' cognitive and affective processes (Reber et al., 2004).

Complexity and symmetry. Multiple studies found that complexity plays an

- 17 -

or the irregularity of the forms (Berlyne, 1974; Mallon, 2014). Complexity expresses a condition of numerous elements in a system and numerous forms of relationships among the elements. A high aesthetic appeal goes along with an intermediate level of complexity according to Berlyne (1974). Images of intermediate complexity are considered more preferred than highly complex images in general. Berlyne (1974) argued that emotional preference for an object would increase by the level of complexity, until a certain point from where complexity would cause over stimulation and preference would decrease. Many studies have reported the relationship between complexity and beauty, some significant and some report that the trend is entirely linear (Forsythe et al., 2010). Complexity is closely related to fractal geometry, as it can be regarded as a preliminary indicator of complexity in a pattern.

Likewise, Jacobsen, Schubotz, Hofel and Cramon (2006) find in their study that symmetry guides aesthetic judgments of beauty. Symmetry can be defined as the mathematical feature of the system that remains unchanged under some transformation. Symmetry is related to fractal geometry as it consists of scale symmetry: when magnifying in scale, each part is a reduced-size copy of the whole.

Perceptual fluency. Reber et al. (2004) described in their research the modern

interactionist perspective, where beauty is grounded in the processing experiences of the perceiver that emerge from the interaction of stimulus properties and perceivers' cognitive and affective processes. This cognitive and affective processing can lead to a common experience of processing ease, which is called ‘fluency’. It is proposed that aesthetic pleasure is a function of the perceiver's processing dynamics. The more fluently perceivers can process an object, the more positive their aesthetic response (Reber et al., 2004).

- 18 -

have been proposed to account for the fluent perceiver’s processing of visual fractal-based images. First of all, the human visual system has evolved in a natural world and has – as a result – become optimally adapted and specialized in geometric characteristics of the natural environment (Braun et al., 2013; Redies, 2007). Empirical research about logos shows that naturalness is an essential logo design element which significantly influences consumer affective responses, and that natural logos are clearly preferred to abstract logos (Machado, Vacas de Carvalho, Torres, & Costa (2015). Furthermore, highly complex fractal structures can be generated using a simple recursive rule (Peitgen, Jürgens, & Saupe, 1992). Fractal structures might easily be analyzed, possibly translating into relatively low processing demand (Joye et al., 2015).

2.3 Construal Level

The relationship between aesthetic liking and consumer behavior, may be strengthened by the way individuals see and interpret the world. Social psychological research on human behavior emphasizes that people do not respond to the situation per se, but to the situation as they see it (Schwarz, 2006). The way individuals see, perceive, comprehend and interpret the world, and in particular the behavior or actions of others, are known as construals. The Construal Level Theory describes the relation between the level of construal (extent to which people’s thinking is abstract or concrete) and the psychologically distance of the event. CLT proposes that psychologically distant events are represented by their essential, abstract, and global features (high-level construals), whereas psychologically near events are represented by their incidental, concrete, and local features (low-level construals) (Wakslak & Trope, 2006). The general principle ‘psychologically distance’ is defined on other distance dimensions, namely temporal distance (time perspective), social distance (self vs. other, in-group vs. out-group), hypothetically distance (likeliness of event) and spatial distance (physical space). Each various dimension seems independently relevant to consumer choice (Trope & Liberman, 2003; Liberman, Trope, & Wakslak, 2007).

- 19 -

and higher level criteria from a choice that has to be made (Gasper & Clore, 2002). Thus in general, people in negative moods tend to use detail-oriented, analytical and effortful processing strategies, whereas people in positive moods tend to use simple heuristics in processing information (Schwarz, 1990; Schwarz, Bless, & Bohner, 1991). Fujita and Trope (2006) investigated the relationship between construal level and self-control. Their research provides initial evidence that high-level lead to greater self-control than activation of low-level construals. Levels of construal can moderate how people make decisions when faced with a self-control dilemma. Consumers at a high construal level demonstrated a greater tendency to make decisions reflecting self-control (Fajita et al., 2006). The effect from aesthetic liking on consumer behavior, may be strengthened or weakened by the ability to control yourself. Therefore, we assume that the effect of aesthetic liking on consumer behavior is greater, if consumers are acting at a low construal level.

2.4 Consumer Motivations

Research has demonstrated that consumers on a shopping trip are motivated to shop for several reasons. On the one hand the functional need, one the other hand the desire to satisfy emotional needs. (Kim, 2006; Arnold & Reynolds, 2003). In this research we want to account for different consumer motivations, and investigate the effect of fractality from an utilitarian versus hedonic perspective.

One of the most essential questions for marketers, when creating the perfect marketing strategy, is: ‘who is my customer?’. Perhaps even more important, is the follow-up question: ‘why does he or she go shopping?’. Through the years, researchers have directed attention to the need to understand the shopping experience and shopping motivations. From traditional research, with a task-oriented or utilitarian approach, towards the emotional aspects of shopping, seen from a hedonic perspective.

Utilitarian perspective. Utilitarian consumer behavior has been described as

- 20 -

Value can also result from an involved costumer collecting information, arguing that it was not a waste of time. Interest in the product class, information search in the product class, product knowledge, and word-of-mouth activity are all positively related to browsing behavior. Concluding, efficiency and achievement can be seen as two main dimensions of utilitarian motivation (Kim, 2006).

Hedonic perspective. As was mentioned by Bloch and Richins (1983), product

acquisition explanations may not fully reflect the totality of the shopping experience. Therefore, in the 1990’s, research has directed attention to shopping’s hedonic aspects; stimulated by its potential entertainment and emotional worth (Arnold & Reynolds, 2003). From the hedonic perspective, the shopping trip is viewed as a positive and emotionally satisfying experience, whether or not a purchase was made (Kim, 2006). Similar to the utilitarian perspective, value is created by completing successfully the ‘task’. Only the ‘task’ is concerned with hedonic fulfillment, such as experiencing excitement, arousal, fun, joy, amusement, fantasy, adventure and sensory stimulation (Babin et al., 1994; Bloch & Richins, 1994). Hedonic consumption has been defined as those facets of consumer behavior that relate to the multisensory, fantasy and emotive aspects of consumption. The aspect ‘multisensory’ is related to the senses or the experience of sensation (i.e., tastes, sounds, scents, tactile impressions, visual images). Individuals respond to multisensory impressions by encoding the inputs and generating multisensory images by themselves. ‘Emotive’, another aspect of hedonic consumption, is related to emotional arousal. Emotions represent motivational phenomena, including feelings such as joy, jealousy, fear, rage and rapture (Hirschman & Holbrook, 1982; Arnold & Reynolds, 2003).

- 21 -

Hirschman and Holbrook (1992) define hedonic consumption as those facets of consumer behavior that relate to the multisensory, fantasy and emotive aspects of consumption. Therefore, this hedonic side of consumption involves aspects of aesthetic liking.

2.5 Conceptual Model

In the current research we are especially interested in the effect of fractal-like properties in visual advertisements on consumer behavior. Research has pointed out that there is a close relation between art and the beauty of the natural world. Now, for the first time, research about fractal-like properties will be conducted from a marketing perspective. The current research aims to investigate whether or not fractal-like properties might really contribute to the marketing world. Figure 3 shows the conceptual model of the current research.

2.6 Hypotheses

Hypothesis 1. The main relationship we want to investigate, is the relation

between the level of fractal-like properties in an advertisement and consumer behavior. This research will focus on scale-invariance as fractal property, based on the measure

 _{Hypotheses (with the *) assume that a specific value has a positive effect on the DV.}

- 22 -

spatial frequency ‘f’.We assume that consumer behavior is positively influenced by the level of spatial frequency in the advertisement. However, Menzel et al. (2015) found that faces in front of patterns with median slopes (i.e., -2) on the background received higher attractiveness ratings than faces in front of extreme slopes. High spatial frequencies have a Fourier slope of 0 or -1, where low spatial frequencies have slopes of -3 or -4. This leads to the following hypothesis:

H1: The more extreme (high or low) the spatial frequencies in an advertisement, the lower

the intention to purchase and the willingness to recommend.

In this research we want to account for different consumer motivations, and investigate the effect of fractality from an utilitarian versus hedonic perspective. However, the effect of type of motivation as such depends on multiple variables (i.e., type of product and store, time to purchase a product). Therefore, this study will research the main effect of type of motivation and the interaction effect on consumer behavior from an exploratory perspective.

Hypothesis 2. Empirical research on advertising indicates that aesthetic thoughts

concerning advertising can influence the willingness to purchase that product or brand (Shimp, 1981; Mackenzie & Lutz, 1989). These aesthetic feelings are important predictors of general behavioral intent (Ha & Lennon, 2010). Assuming that the level of spatial frequency has an indirect effect on consumer behavior, this study will research the mediated effect of aesthetic liking.

H2: The effect of different levels of spatial frequencies in an advertisement on intention

to purchase and willingness to recommend, is mediated by aesthetic liking.

- 23 -

willingness to recommend. Whilst utilitarian consumer behavior has been described as rational and directed towards satisfying functional or economic needs (Kim, 2006). Therefore, aesthetic liking is no underlying mechanism for this type of motivation.

Hypothesis 3. Research have shown that aesthetic pleasure is a function of the

perceiver's processing dynamics. The more fluently perceivers can process an object, the more positive their aesthetic response (Joye et al., 2015; Reber et al., 2004). Fractal images have been widely acknowledged for their considerable aesthetic appeal, for a variety of reasons. This study will focus on scale-invariance, an important fractal-like property, which is considered to be aesthetic appealing. According to the work of Menzel (2015), there is an intermediate level of spatial frequencies which is preferred. The following hypothesis can be stated:

H3: The more extreme (high or low) the spatial frequencies, the less the aesthetic liking

of the advertisement.

Besides looking at the relationship between fractal-like properties and aesthetic liking, we will also test whether there is an effect of the type of motivation on aesthetic liking. We expect that observing the advertisement according to the hedonic view, will create a higher aesthetic liking compared to the utilitarian view. Aesthetic liking is used to describe the perception and response to an object, as well as interactions with objects or scenes that evoke an intense feeling (e.g., beautiful, attractive, ugly, sublime) (Leder & Nadal, 2014; Chatterjee, 2011). Hirschman and Holbrook (1992) defined hedonic consumption as those facets of consumer behavior that relate to the multisensory, fantasy and emotive aspects of consumption. When consumers are acting from a multisensory or emotive aspect, we assume that they are more likely to evoke an intense feeling and create higher aesthetic liking.

H3b: Observing the advertisement from a hedonic perspective, will create higher

aesthetic liking compared to the utilitarian perspective.

Hypothesis 4. Researchers have found that aesthetic feelings are important

- 24 -

that pleasure and arousal are positively related to consumer satisfaction, purchase intention and approach behavior (Ha & Lennon, 2010). If advertisements cause positive affective responses, the intention to purchase and willingness to recommend the advertised products increases, leading to the following hypothesis:

H4: The higher the aesthetic liking of the product, the higher the intention to purchase

and willingness to recommend.

It is shown that aesthetics play a crucial role in the decision making process of consumers with hedonic shopping motivations (Hoyer & Stokburger-Sauer, 2011; Hirschman & Holbrook, 1982). Hoyer and Stokburger-Sauer (2011) stated in their research the following: “a consumer’s aesthetic taste is a main driver of hedonic value while his or her knowledge is a main driver of utilitarian value” (p. 173). Therefore, consumers that are hedonic motivated rely more on the aesthetic liking of the product in their consumer behavior, compared to utilitarian motivated consumers. The following hypothesis is formulated:

H4b: The effect of aesthetic liking on intention to purchase and willingness to recommend

will be greater if the advertisement is observed from a hedonic perspective.

Hypothesis 5. Another interesting question is whether or not the effect of aesthetic

liking on consumer behavior also depends on the consumer’s construal level. The effect from aesthetic liking on consumer behavior, may be strengthened or weakened by the ability to control yourself. Research indicates that a high construal level leads to greater self-control than activation of low-level construals (Fajita et al., 2006). Therefore, we assume that the effect of aesthetic liking is greater, if consumers are acting at a low construal level.

H5: The effect of aesthetic liking on the intention to purchase and willingness to buy will

- 25 -

3. RESEARCH DESIGN

In this study, we examined whether visual advertisements high on fractal characteristics scored higher on the intention to purchase and willingness to recommend, than structures low on fractal characteristics. We performed a 3 x 2 factorial between-subjects design study. The factorial design consisted of six conditions (see Figure 4).

The main study consisted of 12 different questionnaires, according to the six conditions and the distinction in the product for males and females. The participants were recruited via social media (Facebook), and they were randomly assigned to one of the six conditions. For every condition, the minimum amount of respondents has to be 30. Therefore, the total amount of respondents for this study should be at least 180. The respondents could fill out the questionnaire at the time and place preferred by themselves. The questions of the questionnaire were stated in Dutch, since the majority of the network of the author is Dutch. The English version of the questionnaire can be found in Appendix A.

3.1 Level of Fractal-like Properties

Based on the level of spatial frequencies and the distinction between males and females, six visual advertisements were manipulated with regard to their level of scale-invariance. The advertisement background was black, with in the middle a lighted round plane. This plane differed in the slope of the power spectrum (-1, -2 and -3), according to the 1/f statistic. In this plane there was a white male or female T-shirt, headlined by the statement ‘Be fearlessly authentic’ and at the right the Converse logo (see figure 5). The brand Converse was chosen, as the brand has not radical associations and is seen as unisex. All participants were exposed to the advertisement for the same amount of time (i.e., 10

- 26 -

seconds). The participants were able to look again to the advertisement, if they wanted to.

3.2 Consumer Motivations

Half of the respondents had to rate the advertisement while being utilitarian motivated, while the other half was hedonic motivated. Participants were randomly assigned to read one of the following two scenario’s about a hypothetical shopping experience (Kaltcheva & Weitz, 2006):

Utilitarian perspective: You are going on a holiday trip this weekend, and realize that you do not have enough suitable T-shirts for the trip. As a result, you decide to purchase at least one more T-shirt. You are driving to a store that sells T-shirts. All you want to do is to find one or more suitable T-shirts for your trip and leave the store. Along the way to the store you see a T-shirt advertisement on a billboard next to the road.

Hedonic perspective: It’s just past noon on a Saturday. It’s pouring rain, so you can’t do anything outdoors. You find what’s on TV too dull to watch. You feel very, very bored and decide to call some friends to visit some stores to relieve the sense of boredom. On your way, you see a T-shirt advertisement on a billboard next to the road.

- 27 -

To increase the salience of the respective motivational orientation, participants were asked to write a story of five or more sentences long about their feelings and thoughts in that specific or a comparable situation.

3.3 Aesthetic Liking

The study used the variables attractiveness, arousal, interestingness, valence, boredom and innovativeness to test aesthetic liking, as mentioned by Faerber (2010). All variables were assessed on a 7-point bi polar scale.

A reliability test was conducted to check the reliability of the items that measure the aesthetic liking. As it is widely accepted in academic research, the Cronbach’s alpha has to be higher than .6 (Malhotra, 2010). The Cronbach’s alpha is .90, which indicates a good internal consistency.

3.4 Consumer Behavior

Intention to purchase. To measure the intention to purchase (ITP), several

statements were tested according to a 7-point Likert scale (strongly disagree to strongly agree). The following statements were tested: ‘I would consider purchasing this product’, ‘I am interested in trying this product’, and ‘I plan on buying this product’ (Barber, Kuo, Bishop, & Goodman, 2012). The intention to purchase the products appeared to have a good internal consistency, α = .95.

Willingness to recommend. Again a 7-point Likert scale (‘extremely unlikely’ to

- 28 - 3.5 Construal Level

The Behavioral Identification Form (BIF) is an instrument designed to measure individual differences in how abstractly or concretely they represent action (Vallacher & Wegner, 1989). Past research (Liberman & Trope, 1998) has shown that the BIF reveal differences in construal level. Each item on the BIF presents an action followed by two alternative identities, one lower and one higher in level. Each question requires participants to describe an activity (e.g., ‘washing clothes’), by choosing an option that best describes the activity. One might describe the behavior as ‘removing odors from clothes’ (more concrete), whereas another might describe it as ‘putting clothes in the machine’ (more abstract). The participants were told that there are no wrong answers and to choose the description that they personally believed is more appropriate in each pair (Vallacher & Wegner, 1989; Fujita et al., 2006).

The BIF that was used, is composed of 15 items. The alternative identities were derived from pilot subjects, who were asked to provide as many descriptions of each of the original activities as they could within 10 minutes (Vallacher & Wegner, 1989). The most frequently mentioned higher- and lower-level descriptions for each original activity were used to construct the BIF. The Cronbach’s alpha for the construct is .69, which makes it reliable for further analysis.

3.6 Questionnaire

- 29 -

- 30 -

4.

RESULTS

In total, 306 participants opened the questionnaire. However, 75 persons did not finish it and were therefore excluded from the dataset. This makes a total of 231 participants that form the collected dataset of this research. This data is analyzed with the use of Statistical Package for Social Science (SPSS, version 22). The descriptives of all conditions are first mentioned, followed by the presentation of the results of the main analyses and subsequently the rejection or acceptation of the hypotheses.

4.1 Descriptive Statistics

The 231 participants were randomly assigned to one of the six conditions, which had therefore over 30 respondents per condition. The group of respondents consisted of 92 men (39,8%) and 139 women with an average age of 28 years (SD = 10.63). An overview of the distribution of the conditions and the descriptive statistics can be seen in Table 1.

Table 1

Condition Frequency Mean age Percentage men

1 38 (16,5%) 28 42,1% 2 40 (17,3%) 29 35% 3 40 (17,3%) 29 40% 4 40 (17,3%) 27 42,5% 5 38 (16,5%) 25 39,5% 6 35 (15,2%) 28 40% Total 231 28 39,8%

Table 1: Distribution and descriptives of the conditions

A Pearson Chi-Square test was conducted to check whether the respondents were randomly assigned among the six conditions. The test included the variables age and gender. The percentage of participants per conditions did not differ in gender, 2_{(5, N =} 231) = 0.59, p = .988. Neither did the percentage of participants per age, 2_{(185, N = 231)} = 177.58, p = .639. This indicates that the randomization was done successfully.

4.2 Figure-ground Contrast

- 31 -

0.921, p = .399. This implies that the effects of the level of spatial frequency are controlled for the effect of contrasts with the background.

4.3 Main Effects

Two general linear models (univariate), i.e., an ANOVA test, with level of spatial frequency and type of motivation as the between-subject variables, and the intention to purchase and the willingness to recommend as dependent variables were performed to test the first hypothesis. Table 2 shows the means and standard deviations of the different conditions for the intention to purchase and willingness to recommend.

Table 2

Motivation Spatial Frequency Mean (ITP) SD (ITP) Mean (WTR) SD (WTR)

1 (Utilitarian) -1 2.95 1.67 2.87 1.26 -2 3.23 1.83 3.02 1.72 -3 2.76 1.72 2.35 1.25 Total 2.98 1.74 2.75 1.45 2 (Hedonic) -1 2.78 1.61 2.58 1.24 -2 2.34 1.52 2.45 1.20 -3 2.85 1.87 2.60 1.31 Total 2.65 1.66 2.54 1.24

Table 2: Means and Standard Deviations (motivation*spatial frequency) of the ITP and WTR

DV: Intention to purchase. The two-way ANOVA revealed no statistically

- 32 - Figure 6: Visual representation of the pairwise comparisons test

DV: Willingness to recommend. The two-way ANOVA with level of spatial

frequency and type of motivation as between-subject variables, and the willingness to recommend as dependent variable revealed no statistically significant main effect of spatial frequency, F(2, 231) = 0.90, p = .406. Nor was the main effect of type of motivation statistically significant, F(1, 231) = 1.37, p = .243. The results of the interaction effect on the WTR show there was no significant effect, F(2, 231) = 1.86, p = .158. Figure 7 is a graphical representation of the results. Although there was no significant difference in means, the graph shows a drop of the blue line (utilitarian motivated) from spatial frequency level -2 (slope of -2) to level -3 (slope of -3).

- 33 -

Table 3 shows us the pairwise comparisons, which allows us to discover significant differences between specific means. As expected from Figure 7, the mean difference between the slope of -2 and -3, while utilitarian motivated, is statistically significant. The mean difference between the slope of -1 and -3 is marginally significant, while utilitarian motivated.

Table 3

Motivation Spat. F Spat. F Sig

1 (Utilitarian) -1 -2 -3 0.607 0.090 -2 -1 -3 0.607 0.026 -3 -1 -2 0.090 0.026 2 (Hedonic) -1 -2 -3 0.675 0.936 -2 -1 -3 0.675 0.628 -3 -1 -2 0.936 0.628 Table 3: Pairwise Comparisons of the WTR

4.4 Effect on Aesthetic Liking

Another ANOVA test, with level of spatial frequency and type of motivation as the between-subject variables, and aesthetic liking as dependent variable was performed to test the third hypothesis, namely: the more extreme (high or low) the spatial frequencies, the less the aesthetic liking of the advertisement.

- 34 - Figure 8: Visual representation of the pairwise comparisons test

4.5 Correlations of Dependent Variables

The correlations between the dependent variables (aesthetic liking, intention to purchase and willingness to recommend) are shown in Table 4. It can be seen that each relation correlates statistically significant. In line with the fourth hypothesis, the level of aesthetic liking has an influence on the intention to purchase and willingness to recommend.

Table 4 ITP WTR Aesthetic liking Intention to purchase 1 Recommend 0.667* 1 Aesthetic liking 0.609* 0.560* 1 *. Correlation is significant at the 0,01 level (2-tailed).

Table 4: Pearson Correlations between ITP, WTR and aesthetic liking

4.6 The Moderation Effects of Type of Motivation

- 35 - 4.7 The Mediating Effects of Aesthetic Liking

The main relations between the dependent (level of spatial frequency and type of motivation) and independent variables (ITP and WTR) revealed no significant results. Nowadays, there is no need for significant evidence of the main relation between X and Y in order to estimate and test hypotheses about indirect effects. Mediation analysis by PROCESS (Hayes, 2016) were performed to test whether the relation between level of spatial frequency on both dependent variables is mediated by aesthetic liking (H2). A follow up mediation analysis tested whether the effect of different levels of spatial frequency per type of motivation on the ITP and WTR, was mediated by aesthetic liking. Nevertheless, all analysis were statistically not significant. The interested reader can find the results in Appendix B.

4.8 The Moderating Effects of the Consumers’ Construal Level

Two moderator analysis were performed to test if the effect of aesthetic liking on both dependent variables (ITP and WTR) was influenced by the consumers’ construal level. The variables were mean centered, to give the constant a meaningful interpretation.

DV: Intention to purchase. Table 5 shows the results of the moderation analyses,

with aesthetic liking as independent variable, intention to purchase as dependent variable and the consumers’ construal level as moderator. It shows that the interaction effect of the moderation is significant, ΔR2 = .01, F(1, 227) = 4.04, p = .046.

Table 5

Coefficient P LLCI ULCI

Constant 2.8036 0.000* 2.6278 2.9794

Construal level (mc) 0.2521 0.583 -0.6517 1.1559

Aesthetic liking (mc) 1.0055 0.000* 0.8352 1.1757

Interaction 0.8970 0.047* 0.0176 1.7765

*. Significant at the 0,05 level

- 36 - Figure 9: Interaction plot of the moderation analysis

Figure 9 is a graphical representation of the output of the moderation analysis, in order to correctly interpret the interaction effect. The (significant) correlation between aesthetic liking and the intention to purchase is reflected in both this figure, as in Table 5. The graph shows that the construal level enhances this relationship. The higher the construal level (i.e., the more abstract people think), the stronger this relationship. This contradicts the stated hypotheses (H5). A possible explanation may be derived from two concepts, namely: the preference for desirability and a general description when acting at a high construal level (Liberman & Trope, 1998). The desirability concept may positively influence the intention to purchase. In addition to that, the questions about the ITP had a more general description (e.g., without naming the price), which is more preferred by people who are acting at a high construal level.

DV: Willingness to recommend. Table 6 reveals the results of the moderation

analyses, with aesthetic liking as independent variable, willingness to recommend as dependent variable and the consumers’ construal level as moderator. It shows that the interaction effect on WTR is marginally significant.

1,5 2 2,5 3 3,5 4 4,5

Low Aesthetic Liking Average Aesthetic Liking

High Aesthetic Liking

Intention to Purchase

- 37 - Table 6

Coefficient P LLCI ULCI

Constant 2.6316 0.000* 2.4861 2.7770

Construal level (mc) -0.0181 0.9620 -0.7658 0.7296

Aesthetic liking (mc) 0.7350 0.000* 0.5942 0.8758

Interaction 0.6884 0.0635 -0.0391 1.4159

*. Significant at the 0,05 level

Table 6: Moderation analyses with aesthetic liking (IV), WTR (DV) and consumers’ construal level (moderator)

Figure 10 shows the graphical representation of the results. It can be seen that construal level moderates the relation between aesthetic liking and willingness to recommend. However, this is statistically not significant, ΔR2 = .01, F(1, 227) = 3.48, p = .064.

Figure 10: Interaction plot of the moderation analysis 1,5 2 2,5 3 3,5 4

Low Aesthetic Liking Average Aesthetic Liking

High Aesthetic Liking

Willingness to Recommend

- 38 - 4.9 Hypothesis Testing

The aforementioned results lead to the rejection or acceptation of the hypotheses of this study. Table 7 gives an overview of rejection or acceptation of the hypotheses.

Table 7

Hypotheses Rejected/Accepted

H1: The more extreme the spatial frequencies in an advertisement, the lower the

intention to purchase and the willingness to recommend

Rejected

H2: The effect of different levels of spatial frequencies in an advertisement on

intention to purchase and willingness to recommend, is mediated by aesthetic liking

Rejected

H3: The more extreme (high or low) the spatial frequencies, the less the aesthetic

liking of the advertisement.

Rejected

H3b: Observing the advertisement from a hedonic perspective, will create higher

aesthetic liking compared to the utilitarian perspective.

Rejected

H4: The higher the aesthetic liking, the higher the intention to purchase and

willingness to recommend.

Rejected

H4b: The effect of aesthetic liking on intention to purchase and willingness to

recommend will be greater if the advertisement is observed from a hedonic perspective.

Rejected

H5: The effect of aesthetic liking on the intention to purchase and willingness to

buy will be greater, if consumers are acting at a low construal level.

- 39 -

4 DISCUSSION, LIMITATIONS & IMPLICATIONS

In this final chapter, the main findings are discussed. Next, the limitations of this research, followed by the managerial implications and possible directions for future research.

5.1 Discussion

There is a growing interest in the impact of like properties. Research on fractal-like properties has focused on the creation and liking of art and architecture. The goal of this research was to increase our understanding of the effect of fractal-like properties in relation to marketing. Especially, we were interested into what extent the level of spatial frequency on advertisements would influence the intention to purchase and willingness to recommend. However, the results of this research did not confirm the expectations.

The main effect. Our assumption that an intermediate level of spatial frequencies

in visual advertisements will create higher purchase intention and willingness to recommend was not confirmed. The influence of the other independent variable, type of motivation, was studied from an exploratory perspective. The results showed that there is no main effect of type of motivation or interaction effect between the level of spatial frequency and type of motivation on the ITP and WTR. A possible reason for rejecting the main hypothesis relates to the choice of the advertisement in the questionnaire. The type of brand and product in combination with the specific scenarios could not have created the right simulation of a real situation. Additionally, the use of different levels of spatial frequencies in the background, instead of in the product, could have caused that the different properties were not clearly present.

- 40 -

slopes (e.g., slopes of -1 or -3) (Menzel et al., 2015; Redies et al., 2007; Braun et al., 2013). Subsequently, Oliver, Rust and Varki (1997) found that positive affect is related to patronage intent, purchase intention, satisfaction and approach behavior (Ha & Lennon, 2010).

The mediating effects of aesthetic liking. Also the expected mediating effect of

aesthetic liking on the relation between different levels of spatial frequency and the ITP and WTR was not confirmed. The result holds when the mediator analysis is done with the other independent variable, type of motivation. Again a possible reason for rejecting the mediator hypothesis relates to the choice of the advertisement in the questionnaire.

Idem where the results from the ANOVA between level of spatial frequency and aesthetic liking. The results showed no main effect of spatial frequency and type of motivation, or the interaction effect on aesthetic liking. However, the research confirms that the higher the aesthetic liking of the product, the higher the intention to purchase and willingness to recommend. This relationship is reflected in literature, as positive affect is related to consumer satisfaction, purchase intention and approach behavior (Ha & Lennon, 2010; Oliver, Rust & Varki, 1997). The study shows that the relation between aesthetic liking and ITP and WTR is not moderated by the type of consumer’s motivation. A possible explanation for the rejection of the moderator hypothesis may be derived from the defined scenarios, which could not have created the right simulation of a real situation.

The moderation effects of type of motivation. The moderating effect of a

- 41 -

and a general description (Liberman & Trope, 1998). Desirability refers to the value of an action; it reflects the superordinate ‘why’ aspects of an action (e.g., instead of ‘how’). This desirability concept may positively influence the effect of aesthetic liking on the intention to purchase, as aesthetic liking is a value judgement. In addition to that, the questions about the ITP had a more general description (e.g., without naming the price), which is more preferred by people who are acting at a high construal level.

At the end, it can be concluded that the use of different levels of spatial frequencies in the background of advertisements do not have a direct or indirect effect on consumer behavior. This conclusion holds by taking into account the consumers’ type of motivation.

5.3 Limitations

This research contains some research limitations that have to be taken into account. First of all, the dependent variable ‘willingness to recommend’ used to be measured by rating two statements on a 7-point Likert scale. The statements were stated reversed: the first statement positive, while the second statement was stated negative. The answers on this last question where not in line with the answers on the other questions, containing a lot of outliers. To make the construct somewhat reliable, only the first statement was used to measure the willingness to recommend. Though, it is questionable whether the construct is reliable enough to test the willingness to recommend.

Additionally, to test the effect of type of motivation, the questionnaire consisted of two different scenarios to assign the respondents to a hedonic or utilitarian motivation group. In order to increase the salience of the motivation, a follow up question asked to write down a short story about their feelings and thoughts in that specific or a comparable situation. Given the comments at the end of the questionnaire, the respondents struggled with this specific question. Therefore it is unclear whether the respondents where in the particular type of motivation, while giving their opinion about the T-shirt.

- 42 -

with different spatial frequency levels, it is possible to test within a group of respondents what the most preferred T-shirt is.

5.4 Practical Implications and Future Research

This research could be interesting for companies who want to know more about the impact of low-level visual properties (e.g., fractal-like properties) for marketing purposes. Marketing managers can make use of this knowledge, to set the preconditions for advertisements.

- 43 -

REFERENCES

Arnold, M.J., & Reynolds, K.E. 2003. Hedonic shopping motivations. Journal of

Retailing, Vol: 79, 77–95.

Babin, B.J., Dardin, W.R., & Griffin, M. 1994. Work and/or Fun: Measuring Hedonic and Utilitarian Shopping Value. Journal of Consumer Research, Vol: 20, 644-657.

Barber, N., Kuo, P-J., Bishop, M., & Goodman, R. Jr. 2012. Measuring psychographics to assess purchase intention and willingness to pay. Journal of Consumer

Marketing, Vol: 29, No. 4, 280–292.

Berlyne, D.E. 1971. Aesthetics and psychobiology. New York: Appleton-Century-Crofts.

Berlyne, D.E. 1974. ‘The new experimental aesthetics’ in: Berlyne, D.E. Studies in the New Experimental Aesthetics. Washington, DC: Hemisphere Publishing, 1–25.

Biederman, I., & Vessel, E. 2006. Perceptual Pleasure and the Brain: A novel theory explains why the brain craves information and seeks it through the senses.

American Scientist, Vol: 94, No. 3, 247-253.

Bloch, P.H., & Richins, M.L. 1983. ‘Shopping Without Purchase: an Investigation of Consumer Browsing Behavior’ in: Bagozzi, R.P., & Tybout, A.M. Advances in Consumer Research. Association for Consumer Research, Vol:10, 389-393.

Braun, J., Amirshahi, S.A., Denzler, J., & Redies, C. 2013. Statistical image properties of print advertisements, visual artworks and images of architecture. Frontiers in

Human Neuroscience, Vol: 4, No. 808, 1-15.

- 44 -

Fajita, K., Trope, Y, Liberman, N., & Levin-Sagi, M. 2006. Construal Levels and Self-Control. Journal of Personality and Social Psychology, Vol: 90, No. 3, 351–367.

Forsythe, A., Nadal, M., Sheehy, N., Cela-Conde, C.J., & Sawey, M. 2010. Predicting beauty: Fractal dimension and visual complexity in art. British Journal of

Psychology, Vol: 102, 49–70.

Gasper, K., & Clore, G.L. 2002. Attending to the big picture: Mood and global versus local processing of visual information. Psychological Science, Vol: 13, 34–40.

Harper, D. 2015. Online Etymology Dictionary. Retrieved on the 24th of October from http://www.etymonline.com/index.php?allowed_in_frame=0&search=aesthetic &searchmode=none

Hayes, A. F. 2016. PROCESS macro.

Hoyer, W.D., & Stokburger-Sauer. N.E. 2011. The role of Aesthetic Taste in Consumer Behavior. Journal of the Academy of Marketing Science, Vol. 40, 167-180.

Jacobsen, T., & Hofel, L. 2002. Aesthetic judgments of novel graphic patterns: analyses of individual judgments. Perceptual and Motor Skills, Vol: 95, 755-766.

Jacobsen, T., Schubotz, R. I., Hofel, L., & Cramon, D. Y. 2006. Brain correlates of aesthetic judgment of beauty. NeuroImage¸ Vol: 29, 276-285.

Joye, Y. 2010. A review of the presence and use of fractal geometry in architectural design. Environment and Planning B: Planning and Design, Vol: 38, 814-828.

Joye, Y., Steg, L., & Ünal, A.B. & Pals, R. 2015. When Complex Is Easy on the Mind: Internal Repetition of Visual Information in Complex Objects Is a Source of Perceptual Fluency. Journal of Experimental Psychology, 1-12.

- 45 -

Kim, H. 2006. Using Hedonic and Utilitarian Shopping Motivations to Profile Inner City Consumers. Journal of Shopping Center Research, Vol: 13, No. 1, 57-79.

Koch, M., Denzler, J., & Redies, C. 2010. 1/f2 Characteristics and Isotropy in the Fourier Power Spectra of Visual Art, Cartoons, Comics, Mangas, and Different Categories of Photographs. PLoS ONE, Vol: 5, No. 8, 1-11.

Kubovy, M. 2000. ‘Visual aesthetics’ in: Kazdin, A.E. Encyclopedia of psychology, Vol: 8, 188-193. New York: Oxford University Press.

Leber, H., & Nadal, M. 2014. Ten years of a model of aesthetic appreciation and aesthetic judgments: The aesthetic episode – Developments and challenges in empirical aesthetics. British Journal of Psychology, Vol: 105, 443–464.

Liberman, N., & Trope, Y. 1998. The Role of Feasibility and Desirability Considerations in Near and Distant Future Decisions: A Test of Temporal Construal Theory.

Journal of Personality and Social Psychology, Vol: 75, No. 1, 5-18.

Liberman, N., Trope, Y., & Wakslak, C.J. 2007. Construal Level Theory and Consumer Behavior. Journal of Consumer Psychology, Vol: 17, No. 2, 113–117.

Machado, J.C., Vacas de Carvalho, L., Torres, A., & Costa, P. 2015. Brand logo design: examining consumer response to naturalness. Journal of Product & Brand

Management, Vol: 24, No. 1, 78–87.

Mackenzie, S.B., & Lutz, R.J. 1989. An empirical examination of attitude toward the ad in an advertising pretest context. Journal of Marketing, Vol: 53 No. 2, 48-65.

Malhotra, N.K. 2010. Marketing Research: An Applied Orientation, Upper Saddle River, NJ: Prentice Hall

- 46 -

Mandelbrot, B.B., 1983. The Fractal Geometry of Nature. New York: W.H. Freeman and Company.

Oliver, R.L., Rust R. T., & Varki, S. 1997. Customer Delight: Foundations, Findings, and Managerial Insight. Journal of Retailing, Vol: 73, No. 3, 311-336.

Olshausen, B.A., & Field, D.J. 2000. Vision and the Coding of Natural Images. American

Scientist, Vol: 88, 238-245.

Olshausen, B.A., & Field, D.J. 2004. Spare coding of sensory input. Current Opinion in

Neurobiology, Vol: 14, 481-487.

Peitgen, H-O., Jürgens, H., & Saupe, D. 1992. Chaos and fractals. New York: Springer-Verlag.

Reber, R., Schwarz, N., & Winkielman, P. 2004. Processing Fluency and Aesthetic Pleasure: Is Beauty in the Perceiver's Processing Experience? Personality and

Social Psychology Review, Vol: 8, No. 4, 364-382

Redies, C. 2007. A universal model of esthetic perception based on the sensory coding of natural stimuli. Institute of Anatomy, 1-18.

Redies, C. 2014. Beauty: Neglected, but alive and kicking. British Journal of

Psychology, Vol: 105, 468–470,

Redies, C., Hasenstein, J., & Denzler, J. 2007. Fractal-like image statistics in visual art: similarity to natural scenes. Spatial Vision, Vol: 21, No. 1–2, 137–148.

Schwarz, N. 1990. ‘Feelings as information: Informational and motivational functions of affective states’. In Higgins, E.T., & Sorrentino, R.M. Handbook of motivation and cognition, Vol: 2, 527-561. New York: Guilford Press.

- 47 -

Schwarz, N. 2006. Mental Construal in Social Judgment. University of Michigan, 1-22.

Schwarz, N., Bless, H., & Bohner, G. 1991. Mood and persuasion: Affective states influence the processing of persuasive communications. Advances in

Experimental Social Psychology, Vol: 24, 161–199.

Selnes, F. 1993. An Examination of the effect of Product Performance on Brand Reputation, Satisfaction & Loyalty. European Journal of Marketing, Vol: 27, No. 9, 19-35.

Shapiro, S. 1999. When an Ad's Influence Is beyond Our Conscious Control: Perceptual and Conceptual Fluency Effects Caused by Incidental Ad Exposure. Journal of

Consumer Research, Vol: 26, 16-36.

Shimp, T.A. 1981. Attitude toward the ad as a mediator of consumer brand choice.

Journal of Advertising, Vol: 10, No. 2, 9-15.

Spehar, B., Clifford, C.W.G., Newell, B.R., & Taylor, R.P. 2003. Universal aesthetic of fractals. Computers & Graphics, Vol: 27, 813-820.

Tatarkiewicz, W. 1970. History of aesthetics. The Hague, The Netherlands: Mouton.

Taylor, R.P. 2002. Order in Pollock’s chaos. Scientific American, 116–121.

Taylor, R.P., Newell, B.R., Spehar, B., & Clifford, C.W.G. 2001. ‘Fractals: A resonance between art and nature?’ In Lugosi, G., & Nagy, D. Intersections of art and science. The Proceedings of Symmetry: Art and Science. Retrieved on the 29th of October from http://members.tripod.com/vismath7/proceedings/taylor.htm

Taylor, R.P., Spehar, B., van Donkelaar, P., & Hagerhall, C.M. 2011. Perceptual and psychological responses to Jackson Pollock’s fractals. Frontiers in Human

- 48 -

Taylor, R.P., & Sprott, J.C. 2008. Biophilic fractals and the visual journey of organic screen-savers. Nonlinear Dynamics, Psychology, and Life Sciences, Vol. 12, No. 1, 117-129.

Tinio, P.P.L., & Leder, H. 2009. Natural Scenes Are Indeed Preferred, but Image Quality Might Have the Last Word. Psychology of Aesthetics, Creativity, and the Arts, Vol: 3, No. 1, 52–56.

Trope, Y., & Liberman, N. 2003. Temporal Construal. Psychological Review, Vol: 110, No. 3, 403–421.

Trope, Y., & Liberman, N. 2010. Construal-Level Theory of Psychological Distance.

Psychology Review, Vol: 117, No. 2, 440–463.

Vallacher, R.R., & Wegner, D.M. 1989. Levels of Personal Agency: Individual Variation in Action Identification. Journal of Personality and Social Psychology, Vol: 57, No. 4, 660-671.

Wakslak, C.J., & Trope, Y. 2006. Seeing the Forest When Entry Is Unlikely: Probability and the Mental Representation of Events. Journal of Experimental Psychology, Vol: 135, No. 4, 641–653.

Winkielman, P., Schwarz, N., Fazendeiro, T., & Reber, R. 2003. The hedonic marking of processing fluency: Implications for evaluative judgment. In Musch, J., & Klauer, K.C. The psychology of evaluation: Affective processes in cognition and emotion (pp. 189–217). Mahwah, NJ: Lawrence Erlbaum Associates.