Lade Inhalt...

How collages reveal your deepest thoughts

A guide to consumers' minds

©2010 Magisterarbeit 201 Seiten


A deeper understanding of how consumers think, feel and act is vital to the success of management and provides valuable information for managerial decision making in many areas of marketing. One key to this understanding is brand knowledge. Customer brand knowledge is the representation of a brand in consumers' minds. This thesis focuses on a visual-based view stating that brand knowledge is rather image-based as opposed to word-based. Unfortunately, a substantial amount of relevant knowledge within consumers' minds is unconscious and cannot be retrieved, accessed and recalled by customers. As a consequence, certain methods of retrieval are required, such as projective techniques. The method this thesis works with is the collage technique, an expressive projective method.

This diploma thesis creates a multi-layered approach that facilitates the interpretation of collages without the need of any additional information given by the participants, based on metaphor analysis, color theory, a communication model and structural analysis.

The thesis answers following research questions:
- Is it possible to interpret collages without any additional information from respondents?
- What information can be gained from this interpretation?
- Are there any differences between the results of visual and verbal analysis?


Table of contents

List of tables

List of figures

1 Introduction
1.1 Problem statement
1.2 Aim of the thesis
1.3 Outlook

2 Accessing and retrieving brand knowledge
2.1 Assumptions on brand knowledge representation
2.1.1 It is all about representations.
2.1.2 Consumer memory
2.1.3 Brand knowledge structures
2.1.4 Consumers think visually
2.2 Challenges in retrieving brand knowledge
2.2.1 Overview
2.2.2 Heuristics and biases
2.3 Brand knowledge retrieval
2.3.1 Qualitative research
2.3.2 Projective techniques Verbal methods Nonverbal methods
2.4 Characteristics and applications of collages in marketing
2.4.1 History of collages
2.4.2 Characteristics, advantages and limitations of collages
2.4.3 Collages applied in marketing

3 Analyzing visually retrieved brand knowledge
3.1 Understanding the meaning of signs: semiotics and hermeneutics
3.1.1 Semiotics
3.1.2 Hermeneutics
3.2 Metaphor analysis in interpreting collages
3.2.1 Metaphors as cognitive constructs
3.2.2 Metaphors versus image schemata Types of image schemata
3.2.3 Visual metaphors as sources of brand knowledge
3.3 Color theory in interpreting collages
3.3.1 What colors reveal about customers
3.3.2 Same color, different meaning White – the “godlike” color Black – the “negative” color Red – the color of fire Blue – the color of infinite dimensions Gray – the “boring” color Green – the color of life Yellow – the “sunny” color
3.4 A communication psychological view on collages
3.5 Structural analysis of collages

4 Empirical study
4.1 Data collection and sample
4.2 Data analysis
4.2.1 Empathy and mentalizing as justification for this work
4.2.2 The multi-layered interpretation process Metaphor analysis – method and procedure Color theory – method and procedure Communication model – method and procedure Structural analysis – method and procedure Cross validation procedure
4.2.3 Findings and results Interpretation of collages and interviews by the researcher Collage 1bm0407h Collage 1bh0506f Collage 1bc0109m Collage 1bf1205i Collage 2bd0407r Collage 2bv0809r Collage 2ba0309e Collage 2bm1005n Collage 2bk0109n Collage 2bm0907r Information gained from metaphor analysis of collages Information gained from color theory in collages Information gained from communication model Information gained from structural analysis Information gained from metaphor analysis of interviews Differences in results between collages and interviews

5 Discussion

6 Managerial implications and future research

7 References

List of tables

Table 1: Image schemata by Johnson (1990, p. 126)

Table 2: Metaphor analysis (collage 1bm0407h)

Table 3: Color theory (collage 1bm0407h)

Table 4: Communication model (collage 1bm0407h)

Table 5: Structural analysis (collage 1bm0407h)

Table 6: Interview (collage 1bm0407h)

Table 7: Aggregation of meanings (collage 1bm0407h)

Table 8: Metaphor analysis (collage 1bh0506f)

Table 9: Color theory (collage 1bh0506f)

Table 10: Communication model (collage 1bh0506f)

Table 11: Structural analysis (collage 1bh0506f)

Table 12: Interview (collage 1bh0506f)

Table 13: Aggregation of meanings (collage 1bh0506f)

Table 14: Metaphor analysis (collage 1bc0109m)

Table 15: Color theory (collage 1bc0109m)

Table 16: Communication model (collage 1bc0109m)

Table 17: Structural analysis (collage 1bc0109m)

Table 18: Interview (collage 1bc0109m)

Table 19: Aggregation of meanings (collage 1bc0109m)

Table 20: Metaphor analysis (collage 1bf1205i)

Table 21: Color theory (collage 1bf1205i)

Table 22: Communication model (collage 1bf1205i)

Table 23: Structural analysis (collage 1bf1205i)

Table 24: Interview (collage 1bf1205i)

Table 25: Aggregation of meanings (collage 1bf1205i)

Table 26: Metaphor analysis (collage 2bd0407r)

Table 27: Color theory (collage 2bd0407r)

Table 28: Communication model (collage 2bd0407r)

Table 29: Structural analysis (collage 2bd0407r)

Table 30: Interview (collage 2bd0407r)

Table 31: Aggregation of meanings (collage 2bd0407r)

Table 32: Metaphor analysis (collage 2bv0809r)

Table 33: Color theory (collage 2bv0809r)

Table 34: Communication model (collage 2bv0809r)

Table 35: Structural analysis (collage 2bv0809r)

Table 36: Interview collage (2bv0809r)

Table 37: Aggregation of meanings (collage 2bv0809r)

Table 38: Metaphor analysis (collage 2ba0309e)

Table 39: Color theory (collage 2ba0309e)

Table 40: Communication model (collage 2ba0309e)

Table 41: Structural analysis (collage 2ba0309e)

Table 42: Interview (collage 2ba0309e)

Table 43: Aggregation of meanings (collage 2ba0309e)

Table 44: Metaphor analysis (collage 2bm1005n)

Table 45: Color theory (collage 2bm1005n)

Table 46: Communication model (collage 2bm1005n)

Table 47: Structural analysis (collage 2bm1005n)

Table 48: Interview (collage 2bm1005n)

Table 49: Aggregation of meanings (collage 2bm1005n)

Table 50: Metaphor analysis (collage 2bk0109n)

Table 51: Color theory (collage 2bk0109n)

Table 52: Communication model (collage 2bk0109n)

Table 53: Structural analysis (collage 2bk0109n)

Table 54: Interview (collage 2bk0109n)

Table 55: Aggregation of meanings (collage 2bk0109n)

Table 56: Metaphor analysis (collage 2bm0907r)

Table 57: Color theory (collage 2bm0907r)

Table 58: Communication model (collage 2bm0907r)

Table 59: Structural analysis (collage 2bm0907r)

Table 60: Interview (collage 2bm0907r)

Table 61: Aggregation of meanings (collage 2bm0907r)

Table 62: Most common image schemata (visual metaphor analysis)

Table 63: Key brand meanings (visual metaphor analysis)

Table 64: Most common colors (color theory)

Table 65: Key brand meanings (color theory)

Table 66: Key brand meanings (communication model)

Table 67: Structural characteristics (structural analysis)

Table 68: Most common image schemata (verbal metaphor analysis)

Table 69: Key brand meanings (verbal metaphor analysis)

Table 70: Characteristics (collages & interviews)

Table 71: Most common image schemata (collages & interviews)

Table 72: Key brand meanings (collages & interviews)

List of figures

Figure 1: Collage 1bm0407h

Figure 2: Collage 1bh0506f

Figure 3: Collage 1bc0109m

Figure 4: Collage 1bf1205i

Figure 5: Collage 2bd0407r

Figure 6: Collage 2bv0809r

Figure 7: Collage 2ba0309e

Figure 8: Collage 2bm1005n

Figure 9: Collage 2bk0109n

Figure 10: Collage 2bm0907r

1 Introduction

1.1 Problem statement

A deeper understanding of how consumers think, feel and act is vital to the success of management and provides valuable information for managerial decision making in many areas of marketing (Christensen & Olson, 2002; Keller, 2003). One key to this understanding is brand knowledge. Keller (2003) stated that customer brand knowledge was the representation of a brand in consumers’ minds. A brand is more than a logo; it is a complex system of ideas, attributes and associations (Gardner & Levy, 1955). Mental representations are experiences and knowledge transformed and retained in mind for later retrieval (Tulving, 2007). All representations in the brain build up human memory (Dudai, 2007), which consists of multiple systems (Squire, 1987). In memory, experiences and knowledge can be represented verbally and nonverbally (Wyer & Radvansky, 1999; Woodside, 2004) and can function at a conscious and unconscious level (Paivio, 1986); consequently, different memory systems contain different forms of knowledge (Rolls, 2007). Most of the earlier research on brand knowledge (Anderson & Bower, 1973; Gutman, 1982; Aaker, 1991; Keller, 1993) assumed that mental representations were mainly verbal in nature (Costa, Schoolmester, Dekker & Jongen, 2003). However, recent research in psychology and neurobiology has found that images play a major role in consumers’ minds (Costa et al., 2003), as most thoughts originate from images (Damasio, 1994; Pinker, 1994; Zaltman & Coulter, 1995; Zaltman, 1997). Therefore this thesis focuses on a more visual-based view and takes Christensen and Olson’s (2002) perspective of brand knowledge into account, which stated that mental representations were so-called mental models and rather image-based as opposed to word-based. This thesis uses the term “image schema” coined by Johnson (1990) rather than mental models. An image schema is a chief mental structure in the human mind based on experiences (Johnson, 1990). Unfortunately, a substantial amount of relevant knowledge within consumers’ minds, including image schemata, is unconscious and cannot be retrieved, accessed and recalled by customers (Woodside, 2006). As a consequence, certain methods of retrieval are required (Woodside 2004, 2006), such as projective techniques. As a part of qualitative research, projective techniques attempt to circumvent this problem by disguising the true purpose of a study and using methods that do not necessarily require verbal communication (Donoghue, 2000; Steinman, 2009). Such techniques are able to uncover unconscious personal information about respondents by removing social and psychological barriers (Will, Eadie, & MacAskill, 1996; Donoghue, 2000). The method this thesis works with is the collage technique, an expressive projective method (Rook, 2006). The collage technique requires participants to cut out pictures from magazines and to assemble them on a piece of paper (Moisander & Valtonen, 2006). The collage gained popularity during the last century and became an important medium of expression (Davis & Butler-Kisber, 1999). It was first used as a therapeutic form of psychotherapy in the 70s (Buck & Provancher, 1972; Olson, 2000) and gained popularity from then on (Moritani, 1993; Beck, 1998; Takata, 2002; Johnson & Sullivan-Marx, 2006; Meguro, Ishizaki, & Meguro, 2009). However, it took the collage two decades to find its way into the field of marketing (Zaltman & Coulter, 1995; Havlena & Holak, 1996). One of the main benefits of collages is their ability to enable participants to communicate unconscious feelings, ideas and perceptions (Blümelhuber, 2004). Zaltman and Coulter (1995) and Havlena and Holak (1996) were the earliest to utilize the collage technique in marketing research (Costa et al., 2003; Moisander & Valtonen, 2006; Saunders, 2006). Zaltman & Coulter (1995) developed the Zaltman Metaphor Elicitation Technique (ZMET), a tool that assisted in the understanding of consumers’ mental models by creating collages and eliciting metaphors. Havlena and Holak (1996) explored nostalgia with the help of the collage technique. Although the collage technique is used as a popular marketing tool in several studies (Belk, Ger, & Askegaard, 1997, 2003; Costa et al., 2003; Chaplin & John, 2005; Saunders, 2009), two major shortcomings are encountered across the board. Firstly, most studies base their analysis and interpretation of collages on participants’ statements, focusing primarily on verbal data. Secondly, few studies provide a comprehensible guideline to assist in the interpretation of collages. Although visuals are central to meaning, understanding, and reasoning, none of the theories of meaning widely used today offer any serious theory of interpretation (Johnson, 1990). The problem is obvious; there is hardly any method for the analysis and interpretation of collages without the use of any additional information given by respondents.

1.2 Aim of the thesis

The aim of this diploma thesis is to create a multi-layered approach that facilitates the interpretation of collages without the need of any additional information given by the participants, based on metaphor analysis (Johnson, 1990; Seitz, 1998; Niedermair, 2001; Schmitt, 1997, 2003; Schmieder, 2006; Lakoff & Johnson, 2008), color theory (Frieling & Auer, 1961; Itten, 1962; Riedel, 1986; Braem, 1998; Heller, 1998), a communication model (Schulz von Thun, 1992; Holzbrecher & Tell, 2006) and structural analysis. A standardized multi-layered approach to collage interpretation independent from additional information could be a very valuable resource in marketing research.

This thesis attempts answer the following research questions:

- Is it possible to interpret collages without any additional information from respondents?
- What information can be gained from this interpretation?
- Are there any differences between the results of visual and verbal analysis?

The research questions are answered by a three-step procedure consisting of: a visual interpretation of collages, a verbal interpretation of additional interviews given by participants, and a comparison of the results of both interpretations. Note that the researcher strictly followed this sequence in order to avoid any potential bias from interviews. Step three fulfills two purposes. Firstly, it functions as a cross-validation procedure in order to assure validity, which is the extent to which the results of the interpretation of collages coincide with the intended meaning of their creators. This thesis assumes that the intended meanings of participants are stated in the interviews. Beyond that, it determines the value of the multi-layered approach in direct comparison to other procedures. Secondly, valuable information can be gathered by the comparison of visual and verbal interpretation results, as the differences identified may provide interesting information on expressive capabilities of consumers. Differences could indicate that either customers are not able to express their intended meanings, or that visual and verbal analysis focuses on different kinds of information.

1.3 Outlook

Following the preface, chapter 2 introduces the reader to brand knowledge and how it is accessed as well as retrieved. The first section explains how consumers’ minds work and demonstrates the decisive role of mental representations. Different forms of representations are stored in distinct systems of consumer memory forming diverse structures of brand knowledge. The major distinctions are made between implicit/explicit (Schacter, 1987), declarative/procedural (Cohen & Squire, 1980) and semantic/episodic memory (Tulving 1972). The earlier perspectives of brand knowledge evolved from a verbal view-point, such as associative network theory (Anderson & Bower, 1973) and means-end chain theory (Gutman, 1982), to a more visual view, including mental models (Christensen & Olson, 2002). The following section discusses the challenges faced in the retrieval of brand knowledge and explains when and why they occur. Following on from this, techniques for brand knowledge retrieval, particularly verbal and nonverbal projective techniques are outlined. In the final section there is a thorough discussion of history, characteristics, advantages, limitations and applications of collages.

From accessing and retrieving brand knowledge in chapter 2, the thesis continues with the analysis of brand knowledge in chapter 3. The chapter opens with an explanation of the meaning of signs based on semiotics and hermeneutics. Semiotics is the study of signs (Mick, 1986) and is largely influenced by the Saussurean (1983) and Peircean (1965-67) paradigms. Although semiotics is popular in investigating signs, the advantages are clearly outweighed by the disadvantages, making it a rather unsuitable approach for analyzing visuals (Mick, Burroughs, Hetzel, & Brannen, 2004; Rose, 2007). Hermeneutics, the theory of interpretation, offers more opportunities in terms of examining meaning of signs (Madison, 1990). An important part of hermeneutics is the hermeneutic circle, which relates part and whole in a circular way (Hoy, 1982) and analyzes signs on three levels; namely, semantics, the meaning of signs; pragmatics, the relationship of signs; and syntactics, the structure of signs (Niedermair, 2001).

Through the application of certain approaches a three-level analysis of signs is possible. The first is a metaphor analysis in order to interpret semantics of collages. The dominant paradigm of metaphor analysis (el Refaie, 2003), the cognitive metaphor theory (Lakoff & Johnson, 2008) is introduced in this section. A metaphor is the expression of one kind of thing in terms of another (Lakoff & Johnson, 2008). The term “cognitive” stands for mental processes. Metaphor analysis builds upon the concept of image schemata, as metaphors provide vehicles for customers to express image schemata (Johnson, 1990). The aim of this procedure is to identify these underlying schemata and to gain brand knowledge by examining existing metaphors. Explaining the distinction between image schemata and metaphors is crucial and done so in the continuation of this section. Examples are provided, characteristics and relationships are discussed, and the most common types of image schemata are outlined. Finally a discussion on the singularities of visual metaphors concludes this section. The second approach is color theory, which provides additional information on semantics. In this section the impacts of colors on customers and their behaviors are addressed (Riedel, 1986; Heller, 1998) in addition to color theory’s ability to reveal experiences, perceptions and meanings of brands (Aslam, 2006). The section concludes with an illustration of the various meanings of the colors white, black, red, blue, gray, green and yellow. The third approach is Schulz von Thun’s (1992) four sides model used to analyze the collage’s pragmatics. The model assumes that each piece of information possesses four different aspects of information: a matter, self-revelation, relationship and appeal layer. The last approach in the interpretation of collages is the identification of syntactics, based on the analysis of structural characteristics of collages by the researcher. Chapter 4 covers the empirical element of this diploma thesis. In this chapter the interpretation process is explained in greater detail and subsequently applied to ten collages created as part of an empirical study by Koll, von Wallpach and Kreuzer (2010). Firstly, the ten collages are interpreted visually by using the approaches outlined above and secondly, the interpretation process is applied to the ten interviews relating to the collages. Findings and results of the visual and verbal interpretations for each approach are subsequently illustrated. Beyond that, the results and meanings of visual and verbal analyses are compared in order to gain further information.

Chapter 5 summarizes general results and findings and discusses limitations of the single approaches in the interpretation process, whereas chapter 6 outlines managerial implications and future research.

2 Accessing and retrieving brand knowledge

2.1 Assumptions on brand knowledge representation

2.1.1 It is all about representations.

Accessing and retrieving brand knowledge provides valuable information for understanding consumer behavior in marketing (Christensen & Olson, 2002; Keller, 2003). To be able to access and retrieve brand knowledge, it is important to find out how consumers process and store information. When people experience their environment, neurons (nerve cells) in the brain change (Tulving, 2007). This change of neurons forms mental representations, which retain the experience and make it available for later retrieval (Tulving, 2007); this process of forming permanent representations is called encoding (Hasselmo, 2007). A representation is a type of code that determines how elements are expressed in a certain system (Kosslyn, 1994). Representations are unique to every situation and individual and structure people’s behaviors (Rose, 2007). Knowledge and experiences are encoded and stored as experience-dependent representations in human brains, constituting memory (Dudai, 2007). Memory is the representation of past experiences lasting over time (Schacter, 2007), or in a broader sense the capacity to encode, store and retrieve information (Tulving, 2000). Representations make sense of people’s everyday lives and are perceived as truth or fantasy, science or commonsense and conveyed via speech, art, television, dreams and so on (Rose, 2007). In short, mental representations display how brand knowledge is organized in memory (Hutchinson & Eisenstein, 2008). The question of how consumers represent information mentally and how behavior is affected is not completely solved yet and still considered a problem in neuroscience (Paivio, 1986; Kosslyn, 1994; Dudai, 2007). To provide a better understanding of representations and information processing, the next section elaborates on different forms of consumer memory.

2.1.2 Consumer memory

Memory is more than simply a record of our past; it is the controller of all human behavior, including speech, cognition and knowledge (Mantonakis, Whittlesea, & Yoon, 2008). Memory consists of multiple dissociable brain systems capable of changing behavior, emotion and thought based on experiences (Squire, 1987; Johnson, 2007). The human memory system can be distinguished in terms of conceptual structures, types of content, types of processes and brain structures (Johnson, 2007). The long-term memory is divided into declarative and procedural memory (Cohen & Squire, 1980). The declarative memory contains memories consciously acquired and stores all that we have learned about the world. The procedural memory contains skills and abilities acquired unconsciously (Cohen & Squire, 1980). Declarative knowledge is reflected in information that is retrieved consciously, whereas procedural knowledge is reflected in cognitive acts, which are retrieved unconsciously (Wyer, 2008; Cohen & Squire, 1980). The explicit and implicit memory is similar to the declarative and procedural memory, and is about the role of consciousness in performance (Mantonakis, Whittlesea, & Yoon, 2008). Schacter (1987) explained that explicit memory was the conscious acquisition and recall of information in the past, while implicit memory referred to performing tasks based on past experiences not consciously recalled. The declarative memory is divided further into semantic and episodic memory (Tulving, 2002). Semantic memory receives and stores context-free and abstract knowledge such as words and verbal symbols, and their meanings and relations and does not require effort to recall (Tulving, 1972, 2002). Episodic memory contains context-related and event-related information about past experiences and their relations and requires conscious recollection (Tulving, 1972, 2002; Mantonakis et al., 2008). Representations in episodic memory are mainly stored in terms of interrelated stories (Wyer, 2008) that can be described by spatial and temporal attributes (Tulving, 1972). Individuals have the potential for unlimited storage of episodic memory that can additionally be easily retrieved, since everyday experiences are immediately and temporarily stored in the hippocampus and then transferred to a long-term storage in the neocortex (Swap, Leonard, Shields, & Abrams, 1991). Knowledge and experiences people possess are represented in different modalities in their minds; much knowledge is coded verbally, but a large portion is nonverbal (Wyer & Radvansky, 1999; Woodside, 2004). Beyond that, verbal representations, such as words, phrases, numbers and facts, as well as nonverbal representations, such as images, maps and feelings, both function at a conscious and unconscious level (Paivio, 1986). This is due to the fact that memory is based upon two different thinking styles: rational thinking including analytical, deliberate and verbal processes, and experiential thinking involving intuitive, narrative and nonverbal processes (Epstein, 1994). Experiential thinking forms images, metaphors and narratives, while rational thinking encodes reality in symbols, words and numbers (Epstein, 1994). The specific thinking styles lead to different information processing (Epstein, 1994), which results in different aspects of brand knowledge in each memory system (Rolls, 2007). Some contain more verbal and conscious information, such as semantic, explicit and declarative memory and others more nonverbal and unconscious data, such as episodic, implicit and procedural memory (Cohen & Squire, 1980; Schacter, 1987; Tulving, 2002). As a consequence, different methods for retrieval are required (Woodside, 2004, 2006) which are discussed in section 2.3. But first different theories of brand knowledge representations are outlined.

2.1.3 Brand knowledge structures

A major theory of brand knowledge in consumer research is the associative network theory also known as the associative memory model (Anderson & Bower, 1973). This model assumes that knowledge is organized in networks of nodes which create certain knowledge structures (Henderson, Iacobucci & Calder, 2002). Nodes are units of information that are activated by processing data or by activation of linked nodes (Keller, 1993) and can include verbal descriptions, visual representations, sensory or emotional impressions (Supphellen, 2000). This theory has especially been adopted by the semantic/episodic view of memory (Mantonakis et al., 2008). Keller (1993) based his model of brand knowledge on associative network theory and conceptualized brand knowledge as “a brand node in memory to which a variety of associations are linked” (Keller, 1993, p. 3). In general, Keller (2003) defined brand knowledge in terms of “the personal meaning about a brand stored in consumer memory, that is, all descriptive and evaluative brand-related information” (p. 596). Keller (1993) determined two major components of brand knowledge: brand awareness and brand image. Brand awareness is the ability “to identify a brand under different conditions” (Keller, 1993, p. 3) and consists of brand recognition and brand recall. Brand image is the “perception about a brand as reflected by the brand associations held in consumer memory” (Keller, 1993, p. 3). The types of brand associations (attributes, benefits or attitudes) differ along the dimensions of favorability, strength and uniqueness. The content and structure of brand knowledge influence what comes to mind when customers think about a brand, and therefore impacts the power of a brand and the effectiveness of marketing strategies (Keller, 1993, 2009). Aaker (1991) based his model of brand equity on the associative memory model. Brand knowledge forms the basis of brand equity (Keller, 2003) and is the value added or subtracted by assets and liabilities linked to a brand (Aaker, 1991). In other words, brand equity is “the differential effect of brand knowledge on consumer response to the marketing of the brand” (Keller, 1993, p. 8). Aaker (1991) determined brand awareness, brand associations, perceived quality, and brand loyalty as the dimensions of brand equity. Earlier research (Aaker, 1991; Keller, 1993) focused more on verbal (as opposed to nonverbal) representations in associative networks and emphasized brand knowledge largely consisting of semantic memory. It neglects nonverbal and unconscious information found in other memory systems, such as episodic memory (Tulving, 1972). In recent years, there has been a shift from tangible and product-related information to more abstract, intangible and product-unrelated aspects (Keller, 2009). It is no longer only about facts linked to a brand, but about thoughts, feelings, perceptions, images and experiences that form associative networks in consumer memory (Keller, 2009). Despite this shift, these models still rely on the associative network theory, which emphasizes information that is conscious and readily available, neglecting most unconscious information (Anderson & Bower, 1973). Moreover this model focuses on information in nodes per se and ignores structure and links between them (Christensen & Olson, 2002). A second theory of representation of brand knowledge is the means-end chain theory (Gutman, 1982). This model links attributes of products and brands to consequences which hold particular personal value. The consequences provide valuable benefits for consumers leading to a choice of products and brands based upon certain attributes (Gutman, 1982). Although this approach focuses more on structure, it does not provide much information on content, as structure only determines how the content is organized and not its implicit meaning (Christensen & Olson, 2002).

2.1.4 Consumers think visually

A third perspective of brand knowledge is Christensen and Olson’s (2002) focus on structure and content. Christensen and Olson (2002) shifted further towards nonverbal brand knowledge defining it as mental models, namely mental representations about brands and products in memory. Product perceptions, brand attitudes and attributes as well as brand personalities interrelate and form mental models defining the meanings of a brand. This in turn guides consumers’ thinking, reactions and behavior relating to that brand (Christensen & Olson, 2002). The advantage of this perspective is that mental models incorporate not only attitudes, symbols and attributes, but also emotions, feelings, values, images, events and sensory experiences (Christensen & Olson, 2002). Wyer and Radvansky (1999) supported this view and stated that nonverbal brand knowledge consisted of mental models involving visual and acoustic components. This definition of brand knowledge is consistent with recent developments in psychology and neurobiology, which find that the role of visuals in the context of consumer behavior is more significant than first assumed (Costa et al., 2003). According to Blümelhuber (2004) we live in a visual age and most thoughts occur from images (Damasio, 1994; Pinker, 1994; Zaltman & Coulter, 1995; Zaltman, 1997). The various structures of language are often strongly linked to visual imagery (Davis & Butler-Kisber, 1999) and all kinds of visual objects and arrangements carry certain meanings (Moisander & Valtonen, 2006). Images can be sound and olfactory images, but more often they are visual (Damasio, 1994), as a major part of stimuli (two-thirds) reach the brain through the visual system (Kosslyn, Seger, Pani, & Hillger, 1990). Furthermore, humans primarily communicate nonverbally (Weiser, 1988). This is shown in studies that have found that only 7 to 30 per cent of meaning in messages is conveyed by verbal elements (Birdwhistell, 1970; Mehrabian, 1972). “Most thought, emotion, and learning occur without awareness… [and] most mental life is tacit” (Zaltman, 1997, p. 426). Christensen and Olson (2002) took these assumptions into account stating that the content of mental models is image-based as opposed to word-based. Mental models are also referred to as: mental images (Wyer & Radvansky, 1999), cognitive structures (Christensen & Olson, 2002), deep metaphors (Zaltman & Zaltman, 2008) or image schemata (Johnson, 1990). This thesis will use the term “image schema” to replace mental model henceforth. Image schemata are recurring, dynamic and mental patterns that give coherence and structure to experiences found in bodily movement, manipulation of objects and perceptual interactions, helping consumers to reason and make sense of their experiences and environment (Johnson, 1990). Image schemata are one key to brand knowledge, as they guide consumer behavior and include valuable information about brands (Johnson, 1990; Christensen & Olson, 2002). Unfortunately, image schemata and many other relevant forms of knowledge are unconscious (Woodside, 2006) and customers are not able or willing to interpret and report certain aspects of brand knowledge by words alone (Costa et al., 2003; Rook, 2006). The following section elaborates on challenges in retrieving brand knowledge and explains when and why they occur.

2.2 Challenges in retrieving brand knowledge

2.2.1 Overview

The expression of thought is a critical element in retrieving brand knowledge and is often restricted by various barriers (Barner, 2008). Only a limited amount of relevant knowledge and insight can be retrieved, reported and interpreted by customers, since individuals have limited access to the unconscious, where a substantial amount of relevant thinking processes occur (Woodside, 2006). Therefore customers can only retrieve and give reason for actions that are verbalizable, accessible und plausible (Woodside, 2004, 2006). Methods of direct questioning, which are techniques that do not disguise the true purpose of a study, only work if consumers have extraordinary memories, profound self-knowledge, the capacity for introspection, self-disclosure and excellent verbal skills (Rook, 2006). However consumers are not always able to accurately retrieve, access and recall memory, because they simply do not remember situations, are not aware of them or cannot access them (Rook, 2006). One major reason is the transience of memory that is an ongoing decrease in memories of an individual (Schacter, 2001). This effect is enhanced if information is not properly encoded or overlooked due to absent-mindedness. In contrast to forgetting information, traumata affect memory by being remembered constantly (Schacter, 2001). Moreover, consumers differ in their abilities to express and articulate answers, as individuals possess varying levels and abilities in regards to vocabulary and expressiveness (Rook, 2006). Sometimes relevant information is blocked by other memory systems, demonstrated by the tip of the tongue phenomenon (Schacter, 2001). Another issue is the misattribution of current information onto past events or vice versa, resulting in the déjà-vu phenomenon (Schacter, 2001). Finally consumers are often unwilling to provide truthful information. If topics are too sensitive or privacy is infringed, consumers feel uncomfortable providing answers and are inclined to lie, this is also known as the social desirability bias (Rook, 2006). Heuristics and biases can be common causes for memory errors; therefore the next section investigates these phenomena more closely.

2.2.2 Heuristics and biases

Heuristics are cognitive shortcuts that generate judgments based on incomplete or vague information and are the cause for many cognitive biases (Tversky & Kahneman, 1974). A cognitive bias is an error in memory that leads to a distortion in perception, recall and judgment (Wilson, 2002). The collage technique can assist in reducing the influence of heuristics and biases (Rook, 2006). Tversky and Kahneman (1974) found three major heuristics. Firstly, the representativeness heuristic; if a new product is similar to an old one that performed very well, the consumer automatically believes that the new product will also perform well. Secondly, the availability heuristic, which is based on the ease with which information can be retrieved (Tversky & Kahneman, 1974). If many high-quality products of the same brand can be recalled, the consumer automatically assumes that new products of this brand are also high-quality. The last is the anchoring heuristic, causing consumers to form an initial judgment about a product and later adjust the judgment from that point forward (Tversky & Kahneman, 1974). Memory biases are one type of cognitive bias and influence memory-related processes such as the content and recall of memories (Schacter, 2001). Consistency and change biases are two of the most common biases and distort our view of past events, making the past seem more different or similar to present than it actually was. Hindsight biases are also common and cause the change of an early opinion once the actual outcome is known. Last but not least an egocentric bias makes people see their past actions in a more positive light than in reality, and stereotypical biases make people judge their environment based on prejudices (Schacter, 2001). Another group of cognitive biases are attributional biases, which influence people’s understanding of what was responsible for a certain event (Woodside, 2006). The fundamental attribution error for example, refers to underestimating external and situational influences and overestimating personal influences. The self-serving biases refer to attributing failures to situational influences and successes to personal factors. The overconfidence bias consists of the overestimating the validity of one’s own answers (Wilson, 2002). Finally, the confirmation bias influences an individual’s preference towards information that coincides with their beliefs (Tversky & Kahneman, 1974). Most subjective data gathering techniques are susceptible to cognitive biases (Woodside, 2004), which lead to significant differences between explicit and implicit attitudes of consumers, since people are not able or willing to report certain information (Woodside, 2006). Rook (2006) suggested the use of projective techniques to eliminate or reduce cognitive biases and memory failures.

2.3 Brand knowledge retrieval

2.3.1 Qualitative research

In contrast to quantitative research, which quantifies data, qualitative research explores, describes and interprets psychological and sociological contexts by providing insights and understanding of the problem setting (Malhotra & Peterson, 2006). What is distinctive about qualitative research methods is their unstructured and open method of data collection, which does not constrain customers in their answers and is appropriate for developing a richer understanding of a certain context when facing a situation of uncertainty or, a situation that is not yet fully exploited (Malhotra & Peterson, 2006). The procedures can be direct or indirect depending on whether or not respondents are aware of the true purpose of the project (Rook, 2006). Direct measures, such as focus groups and in-depth interviews, question the respondents straightforwardly on a topic, whereas indirect measures use disguised questions (Rook, 2006). Indirect research uses mainly projective approaches such as: association, completion, construction or expressive techniques (Malhotra & Peterson, 2006; Rook, 2006).

2.3.2 Projective techniques

Projective techniques are types of qualitative research methods which are unstructured and disguised. Vague, unstructured and ambiguous stimuli are used (Donoghue, 2000) to elicit unconscious personal information from consumers without their awareness (Donoghue, 2000; Steinman, 2009). Projection, is a defense mechanism and is regarded as the process of attributing characteristics to others in order to protect oneself from revealing unpleasant personal characteristics and experiences (McGrath, Sherry, & Levy, 1993; Donoghue, 2000). In Levy’s projective hypothesis (1985) he stated that “all behavioral manifestations of the human being, including the least and the most significant, are revealing and expressive of his personality” (p. 68). Consequently, projective techniques are able to uncover feelings, beliefs, attitudes and motivations about respondents within a comfortable environment (Donoghue, 2000). Projective techniques assist in overcoming social and psychological barriers such as awareness, social desirability issues and irrationality (Will et al., 1996). They are especially useful in situations in which people are unable or unwilling to verbalize their feelings and experiences, particularly if the subject addressed is sensitive (Khoo-Lattimore, Thyne, & Robertson, 2009). Projective techniques are also used when consumers are unaware of their underlying motives and attitudes (Donoghue, 2000) or are not particularly involved in a brand or product (Rook, 2006). Originally, such techniques were used in clinical and developmental psychology in the 40s and 50s to assist in the understanding of patients’ underlying problems (Donoghue, 2000; Rook, 2006). After World War II marketers, inspired by these methods and convinced of their usefulness in marketing research, started adopting psychological practices, such as association tests, completion tasks and expressive techniques (Khoo-Lattimore at al., 2009). Within the field of projective techniques there are several methods of tapping different forms of brand knowledge. Levy (1985) employed projective techniques to elicit positive, negative, conscious and unconscious responses in order to get information about the perception of products and brands. Some of the most frequently used methods were: word association, sentence completion, cartoon tests, storytelling and collage construction (Rook, 2006). Verbal methods

In order to gain valuable brand knowledge, researchers must engage customers more actively in the research process by enabling them to represent fully their thinking (Zaltman, 1997). As written above, memory systems contain specific brand knowledge, which in turn requires specific retrieval methods (Woodside, 2004, 2006). Projective techniques can be differentiated based on the modality and quantity of the data elicited (Rook, 2006).

Firstly, association tasks, such as word or picture association, require the respondent to answer immediately to a presented stimulus (Will et al., 1996). Though these methods elicit little information in terms of quantity, the information presented is indeed valuable in regards to brand image and product attributes (Donoghue, 2000; Rook, 2006). Participants provide oral or written responses, mainly conscious and verbal in nature (Schacter, 1987; Rook, 2006) and stored in semantic memory (Tulving, 1972). Next, completion tasks require the respondent to respond to a given stimulus, such as to a sentence or story (Will et al., 1996). The data from sentence completion is similar to the data from association tasks (Rook, 2006). Story completion elicits mainly feelings and attitudes (Donoghue, 2000) and goes deeper into episodic memory (Tulving, 1972). Thirdly, construction tasks include indirect questioning, such as asking respondents to project their thoughts into a fictitious situation, or storytelling, such as telling a story based on a visual stimulus (Will et al., 1996). It is assumed that respondents attribute their feelings, attitudes and opinions to other individuals or situations during those tasks (Donoghue, 2000). A considerable amount of data is provided and the brand knowledge retrieved is verbal as well as nonverbal, depending on whether the story is written or spoken (Rook, 2006). Storytelling is a useful method, as consumers organize and express much of their experiences, information and knowledge in the form of narratives (Bruner, 1991; Padgett & Allen, 1997; Adaval & Wyer, 1998). This is very effective, because listening to narratives impacts decision-making and judgment, as it connects to one’s own experiences (Swap et al., 1991). Moreover during purchase decisions consumers often imagine situations of usage and consequence, regarding the product in terms of stories or episodes (Adaval & Wyer, 1998). Special types of storytelling are the Thematic Apperception Test (Morgan & Murray, 1935), the photo-elicitation technique called autodriving (Heisley & Levy, 1991) and visual narratives (Moisander & Valtonen, 2006). The TAT is a powerful tool to assess customers’ attitudes towards products, brands or advertisements by asking the respondent to devise a story about a picture which usually shows a consumer in a usage situation (Schlackman, 1989). Autodriving involves respondents telling a story about photographs showing themselves in specific situations. This enables respondents to recall particular situations immediately (Blümelhuber, 2004). Visual narratives are similar to photo-elicitation techniques and require participants to take photos, to bring them in, and to offer an explanation of the given photographic scenario (Moisander & Valtonen, 2006). To summarize, much brand knowledge is stored in terms of narratives and storytelling and these methods can be a useful tool to retrieve those aspects of brand knowledge (Blümelhuber, 2004). Nonverbal methods

In marketing research most tools are verbocentric and for the most part rely on literal language to collect, synthesize, and report responses, ideas and other data (Zaltman, 1997). However, often consumers have thoughts, emotions and experiences they are unable to articulate because they are “too vague, too complex or too intense for ordinary speech” (Siegelmann, 1990, p. 7). Therefore methods that “engage people in ways that enable them to bring unconscious states to a level of awareness” are needed (Zaltman, 1997, p. 427). Thought is embedded in images, metaphors and emotions and research methods that align with those elements, such as storytelling, role playing and collage construction, facilitate elicitation (Rook, 2006). The value of a brand is determined by how that brand is perceived and by related knowledge based on stories and experiences mainly stored nonverbally in episodic memory (Blümelhuber, 2004). But as people are often unable to access and retrieve such knowledge without other means, images can be used as stimuli for retrieval and as vehicles for communication of emotional and intuitive aspects (Blümelhuber, 2004). This leads over to the last category of projective techniques, namely expressive tasks, such as role-playing, psycho-drawing, objects-drawing and collage construction. In role-playing the respondents are asked to adopt the role of a brand or product, where the content and performance are analyzed and provide information about how consumers see a certain brand or product (Steinman, 2009). Another visual projective technique is psycho-drawing, which is especially useful for expressing moods and feelings (Schlackman, 1989). Respondents are asked to draw a picture about a topic to express feelings and ideas (Schlackman, 1989). A similar drawing task is the simple objects drawing method, where respondents are asked to draw objects that are important to them in a specific situation or location. By doing so the respondent reveals important information about his or her preferences (Schlackman, 1989). The collage technique, which is the technique this thesis deals with, requires the respondents to construct a collage with images out of magazines on a piece of paper based on a certain topic (Moisander & Valtonen, 2006). The images used symbolize cultural meanings and narratives to represent customers’ views of a topic and can be used subsequently as a stimulus for further probing (Moisander & Valtonen, 2006). The collage became an essential innovation in the 20th century, when researchers started appreciating their value as a form of creative visualization and as an important medium of expression (Davis & Butler-Kisber, 1999). The collage technique is a valid alternative to the TAT, autodriving and psycho-drawing (Havlena & Holak, 1996).

2.4 Characteristics and applications of collages in marketing

A picture is worth a thousand words; there is far less linguistic information in pictures than verbal texts, but pictures contain much more of that information, which “is assimilatable to the kind of information present in the perceptual world” (Sonesson, 1989, p. 10). The main premise of visual research is “everything is text”, implying that images can be read like words and that meaning is derived from a medium by putting meaning into it (Blümelhuber, 2004). These statements indicate that the collage technique may be an effective method for retrieving brand knowledge.

2.4.1 History of collages

Eisner (1991) described collage as a flexible composition of pictures assembled gradually and additively until an overall pattern is achieved. The collage originated in arts at the beginning of the 20th century. It was first used by Pablo Picasso in 1912, and it soon became one of the most essential twentieth-century art forms (Leland & Williams, 1994). Today, collages are a unique form of modern art (Blümelhuber, 2004), although Davis (2008) described them as an organizing principle, conceptual strategy or method rather than an art form. When projective techniques became popular, psychologists became interested in the use of collages for diagnostic and therapeutic means (Ikemi, Yano, Miyake, & Matsuoka, 2007) in order to better understand patients (Will et al. 1996). The collage was first introduced by Buck and Provancher (1972) as a therapeutic technique in an occupational therapy journal in the USA (Takata, 2002). It eventually became known as the magazine picture collage technique and steadily gained in popularity from then on (Takata, 2002). Buck and Provancher (1972) employed collages as an evaluative technique in assessing symptoms of male and female adult psychiatric patients. They revealed that collages reflected the patient's psychodynamics, such as self-image and the patient's mental organization (Buck & Provancher, 1972). At about the same time Olson (2000) developed his “Collage Method”, a psychotherapeutic technique that helped creators to reconstruct the evolution of their psyche over time by enabling them to tell personal stories via images (Olson, 2000). In Japan, collage technique became known as collage therapy and was used as a method of psychological assessment (Moritani, 1993) and psychotherapy in order to eliminate mental health problems by reconstructing an image of the self (Takata, 2002). One of the major challenges was that patients often need help to perform the task (Meguro et al., 2009). It has also been applied in art therapy as a means to address emotional needs (Johnson & Sullivan-Marx, 2006) and in recreational activities for dementia patients (Beck, 1998). Besides several fundamental researches and case studies on collages in psychology (Buck & Provancher, 1972; Moritani, 1993; Beck, 1998; Olson, 2000; Takata, 2002; Johnson & Sullivan-Marx, 2006; Ikemi et al., 2007; Meguro et al., 2009), collage technique has also become a well-known tool in marketing research. The remainder of this section illustrates features and applications of collages.

2.4.2 Characteristics, advantages and limitations of collages

As an essential innovation of the 20th century, researchers started appreciating collages as a form of creative visualization and as an important medium of expression (Davis & Butler-Kisber, 1999). “Collage is a heterogeneous, multivalent, multidimensional medium… [that] readily produces effects of spontaneity, simultaneity, ephemerality, fantasy, and disorientation” (Davis & Butler-Kisber, 1999, p. 4). Collages are independent from articulateness as well as assertiveness and are multidimensional, as they incorporate plurality, diversity and richness (Blümelhuber, 2004). Collages promote creative, metaphorical thinking and trigger tacit content as well as unexpected new associations (Davis & Butler-Kisber, 1999). Collages are never complete and there is no right or wrong, although it is important to base them on an idea (Blümelhuber, 2004). The interaction of colors, shapes, internal visual rhythms and balances produce integrated responses (Davis & Butler-Kisber, 1999). Creating a collage involves a process of selection (Blümelhuber, 2004) and mechanisms of metaphor, analogy, and allusion that bring hidden relationships and patterns to a level of awareness (Davis & Butler-Kisber, 1999). Collage has become an exciting strategy for the exploration of memory, imagination, and experiential reflection (Davis, 2008). As no artistic skills are needed collages are appropriate for all ages and mental states (Blümelhuber, 2004). Collages help to uncover thoughts people are not aware of, reduce the blocking of undesired thoughts, stimulate people to express different kinds of thoughts (Woodside, 2004) and are more immune from social-desirability effects and conscious filtering (Barner, 2008).

A major limitation of collages and visual research in general is that collage research cannot be standardized, as visuals are always subjective, interpretative and abstract (Sykes, 1990). Moreover, the difficulty and subjectivity of interpretation is enhanced, as collage construction can be open-ended and almost anything is allowed (Malhotra & Peterson, 2006). In order to control the construction process and facilitate subsequent analysis, a time limit should be set and instructions should be given; for example the use of text passages and writings or drawings on collages unnecessarily complicates them. However, if the respondents have insufficient visual materials or too little time, creativity is constrained and they are not able to express themselves as they would like to. The availability of enough pictures is especially important, as it determines the look and content of a collage. In terms of analysis, there is the risk of analysis bias (Malhotra & Peterson, 2006) due to differences in the conditions of the researcher, as well as situations in which the interpretation processes are performed. Differences may occur depending on the time of the day, physical condition of the researcher and external distractions. As a consequence the interpretation results can vary slightly for different data and samples (Sykes, 1990). In general, it is important to devise an interpretation procedure that is not too complicated and can easily be repeated for each collage, allowing interpretation based on specific criteria. This thesis follows a multi-layered interpretation process that is applied identically on each collage.

2.4.3 Collages applied in marketing

Various studies (Costa et al., 2003; Moisander & Valtonen, 2006; Saunders, 2006) referred to Zaltman and Coulter (1995) and Havlena and Holak (1996) as the first to use the collage technique in marketing research. Zaltman and Coulter (1995) developed the Zaltman Metaphor Elicitation Technique (ZMET), a tool designed to produce insights into consumers’ mental models by using nonverbal channels and metaphors (Zaltman & Coulter, 1995; Khoo-Lattimore at al., 2009) that represent thoughts and feelings about the personal relevance of a topic (Christensen & Olson, 2002). The first stage of the method starts with storytelling based on images collected beforehand by participants. In the second stage, missed images that people would have liked but could not find are discussed. Stage three divides the images into meaningful piles, whereas stage four elicits constructs based on the means-end chain theory (Gutman, 1982). Fifth, the most and least representative pictures are chosen and discussed. Sixth, a summary of the topic discussed is provided and the chosen pictures are assembled into a collage (Zaltman & Coulter, 1995; Zaltman, 1997). Next, key themes are identified and tangible and intangible data is categorized and coded (Zaltman & Coulter, 1995). Finally, a consensus map is constructed. The ZMET is useful for gathering information concerning product-purchase and product-usage as well as perceptions of brands, products and advertisements (Zaltman & Coulter, 1995). One of the major limitations of the ZMET is its labor and time-insensitivity and its need for special training in interviewing and data analysis (Catching-Castello, 2000). Moreover, Zaltman (1997) argued that four or five participants were enough to be representative, a rather doubtful assumption (Catching-Castello, 2000). Furthermore customers are forced to focus on a few images only, which can potentially restrict creativity and mislead findings. Participant fatigue could augment this effect (Catching-Castello, 2000). Beyond this the ZMET bases its interpretation completely on the statements given by participants and does not offer a guideline in regard to how the analysis of data is undertaken. Havlena and Holak (1996) applied collages to explore the nature and structure of nostalgia. Twenty graduate students were assigned to groups of four and asked to create a collage about nostalgia. The students had thirty minutes and were observed and videotaped. At the end of the thirty minutes the groups explained and presented their collages to a panel of researchers (Havlena & Holak, 1996). Certain types of images, such as black and white pictures, and different classes of nostalgia, such as personal and cultural nostalgia were found. Furthermore, persons, products and activities familiar from childhood were frequently related to nostalgia (Havlena & Holak, 1996). Rickard (1994) applied collages in combination with focus groups to obtain better qualitative data to improve car advertisements. The intention was that creating a collage would support the focus group participants’ discussion by providing a vehicle to talk through. The results showed that emotions were a major determinant in car purchases. This knowledge helped car producers to better adapt their advertisements to customers (Rickard, 1994). Costa et al. (2003) used the same approach and combined collages with focus groups to investigate feelings, emotions and experiences related to eating habits. The 29 Dutch respondents were assigned to groups of three and asked to construct a collage concerning the topic of either cooking a meal or having a ready meal. The respondents were asked to put the more important images in the centre. Cooking magazines were intentionally excluded to ensure that the participants did not only choose images from that particular resource (Costa et al., 2003). Between the collage task and the focus group, each group presented its collage briefly. Following that, the collages, presentations and discussions were content-analyzed and patterns were identified. The key findings were categorized based on their relative importance. The collages served as a stimulus for the discussion and complemented the focus groups very well. The collages enabled the expression of unconscious aspects and their value to product and advertising research was confirmed (Costa et al., 2003). Belk et al. (1997, 2003) used collages to investigate consumption desire among customers. The participants were students from three different countries: USA, Denmark and Turkey. The researchers came from the same three countries, consequently avoiding any issues with translation. The collage creation was one part of several projective techniques, including written storytelling, association tasks and drawing (Belk et al., 1997, 2003). The respondents were encouraged to create collages expressing feelings and dreams about desire and then subsequently record their interpretations. As a result consumer desire was identified as a powerful and mainly positive emotion overshadowing other emotions. There were larger differences between men and women than between cultures and a cycle of desire was identified (Belk et al., 1997, 2003). Blümelhuber (2004) assumed that brand knowledge was represented in form of a collage in consumers’ minds and encouraged the creation of physical collages to recreate this knowledge of brands. Chaplin and John (2005) used collages in studies with children in order to find out whether self-brand connections (a brand/product used to represent oneself) developed over time. The method is especially useful in research with children, as creating a collage is a fun and easy task for a child. However, children are easily influenced by peers or parents, so it is important to ensure that they are able to fully focus on their task without any distractions. The general findings showed that the older the children the more self-brand connections existed and the more thorough they were (Chaplin & John, 2005). Saunders (2009) applied the collage technique to collect and evaluate scenario planning information. This was in order to overcome any potential communication, topic sensitivity, memory and qualitative richness problems. Saunders (2009) found out that collage construction was well suited to the objectives of scenario planning, as long as the participants were well informed and willing and able to perform the task. Kriechbaum-Vitellozzi and Kreuzbauer (2009) investigated consumption behavior of refugees in Western countries. The study was conducted in a refugee shelter in Austria. It showed that the collage technique was well suited for overcoming communication problems, as the refugees had different cultural and social backgrounds and were not all proficient in German. The most difficult tasks were getting access to the refugees and building trust in order to ensure their cooperation. The collage technique provided deep insights into the participants’ lives (Kriechbaum-Vitellozzi & Kreuzbauer, 2009).

This section shows that several studies in various fields of marketing research use collage construction as a data collection tool. Nevertheless two major shortcomings are encountered regularly in those works. Firstly, almost every analysis depends for the most part on participants’ statements leading to a focus on verbal data. If there is an analysis of visuals, researchers have a large dependence on statements given by the participants in the form of interviews or written explanations. This focus on verbal data impedes the identification of nonverbal knowledge. It would be more interesting to find out whether it is possible to gather knowledge with an analysis of visuals only, which implies a larger shift from verbal to nonverbal data. Secondly, in most studies the interpretation criteria applied is vague, untransparent or even missing. Hardly any research actually shows how the analysis is done, or provides a detailed guideline for the interpretation of a collage. A standardized approach to collage interpretation could be very useful in marketing research and would facilitate interpretation procedures.

3 Analyzing visually retrieved brand knowledge

After discussing how to access and retrieve brand knowledge, this chapter deals with the approaches used for the interpretation of brand knowledge retrieved via collages. It begins with an introduction to signs and their meanings and continues with a detailed theoretical explanation of the approaches applied.

3.1 Understanding the meaning of signs: semiotics and hermeneutics

Since all knowledge is based on signs and semiotics is the study of signs (Mick, 1986), knowledge is exposed to semiotic interpretation (Rose, 2007). As such, this thesis attempts to deduce knowledge in terms of images in collages, and an elaboration on the concepts of signs and their meanings is provided at this point.

3.1.1 Semiotics

Semiotics is one of the richest and oldest theories for understanding meaning and investigates the nature of meaning and how it is acquired by reality (Mick, 1986; Mick & Oswald, 2006). Mick (1986) stated that semiotics “analyzes the structures of meaning-producing events, both verbal and nonverbal” (p. 197). Generally speaking semiotics is the study of signs and their meanings (Chandler, 2006; Mick, 1986). At the beginning of the 20th century new developments in semiotics appeared. Saussure (1983) and Peirce (1965-67), two major contributors, found two dominant paradigms (Mick et al., 2004). The Saussurian paradigm (1983) was mostly used in structural and text-interpretive analyses (Mick et al., 2004), as it focused on the linguistic sign and privileged the spoken word (Mick, 1986). Saussure (1983) assumed that meaning was formed only through the relationships and interactions of words and not through words per se (Mick, 1986). Saussure’s paradigm (1983) was a two-part model consisting of the signifier, the form of the sign, such as a sound or image, and the signified, the concept or object referred to (Rose, 2007). The sign itself, which is conventional and arbitrary, results from the relationship between signifier and signified (Chandler, 2006). A sign is a fundamental unit of meaning (Rose, 2007) and can be visual or textual and ranges from pictures and drawings to words and texts (Chandler, 2006). Saussure (1983) mainly emphasized the structure, symbolism and cultural impact of communication and meaning (Mick et al., 2004). Fields of application of this paradigm are brand strategy research and consumer behavior. In brand strategy, brands are seen as systems of signs relating brand attributes to consumer needs. In consumer behavior however, it is applied to diagnosing and developing communication strategies (Mick & Oswald, 2006). Unfortunately there are several drawbacks related to this paradigm. First of all, it is not clear whether this theory of language is even able to deal with visuals (Rose, 2007), as the relationships between signifier and signified differ between written/spoken and visual signs (Iversen, 1986). Secondly, Saussure’s (1983) theory provided a static notion of how signs work and neglected changes in meanings (Rose, 2007). Thirdly, while Saussure (1983) only explained how systems of arbitrary signs worked, Peirce (1965-67) provided a much richer typology of signs, explaining how different modes of signification worked (Iversen, 1986). Another advantage of Peirce’s paradigm (1965-67) is that it was more appropriate for conceptual treatises and qualitative and quantitative analyses (Mick et al., 2004). This leads to the second important paradigm, a three-part model, consisting of: the representamen, the form of the sign; the interpretant, the sense of the sign; and the object, the concept or object referred to (Mick & Oswald, 2006). Again the sign itself arises from the relationship between those three parts (Mick, 1986). Peirce (1965-67) identified three categories of signs in his analyses, namely: icon, index and symbol. These categories are distinguished by the different relations between the signifier and signified (Rose, 2007). Iconic signs imitate a concept or object, indexical signs are inherently related to a concept or object, and symbolic signs are related to a concept or object in a conventional and arbitrary manner (Chandler, 2006; Rose, 2007). Fields of application are product and logo design, advertising, and consumer behavior, where it is used to develop meanings and functions in design, to deduce meaning from ads and to help understanding consumption experiences (Mick & Oswald, 2006). This demonstrates that semiotics is a popular method that offers vocabulary and tools for decompounding images and exploring how they work (Moisander & Valtonen, 2006). However, although semiotics can be useful in investigating signs and their meanings in several fields of marketing, the disadvantages outweigh the advantages. Firstly, existing research is eclectic and fragmented due to the magnitude and complexity of works (Mick et al., 2004) which causes concerns about representativeness and replicability of analyses (Rose, 2007). Also, there is a density of terminology in semiotics that is confusing, unnecessary and difficult to understand, making the simple complex, due to the fact that most semiotic studies invent their own analytical terms (Rose, 2007). Lastly, many descriptions and applications are cryptic and do not go beyond lexicographic or taxonomic stages of knowledge development (Mick et al., 2004). Moreover, semiotics neglects the social environment within which visual images are produced and interpreted (Rose, 2007) and many works are unknown, outdated, constrained in length, or focus on only one country or topic (Mick et al., 2004). Finally, research in semiotics has never produced systematic theorems and will remain only an “integrator of previously compartmentalized, incongruent, or unjuxtaposed theories, models, methods, concepts and so on” (Mick, 1986, p. 209).

3.1.2 Hermeneutics

Semiotics is the basis for any interpretative work, as it deals with an analysis of signs and their meanings; however the drawbacks make it hardly suitable for analyzing visuals. Hermeneutics, the “theory of understanding and interpretation of linguistic and non-linguistic expressions” (Stanford encyclopedia of philosophy, 2005), is related to the science of semiotics and offers more opportunities (Madison, 1990). Hermeneutics concentrates only on certain aspects of semiotics, namely interpretation and understanding of signs, in order to grasp the creator’s intended meaning of a text or in this case of a collage (Madison, 1990). The interpretation procedure applied in this thesis follows the concept of the hermeneutic circle. The hermeneutic circle is a process by which understandings are formed and understood as a continuous part-to-whole and whole-to-part movement (Thompson, Pollio, & Locander, 1994). There are various definitions of the hermeneutic circle, depending on the use, purpose and perspective of this approach (Hoy, 1982; Thompson et al., 1994). “In order to understand the whole, it is necessary to understand the parts, while to understand the parts, it is necessary to have some comprehension of the whole” (Hoy, 1982, p. vii). In other words, single elements aid in tracing meaning of the overall context and meaning of the overall context helps to gain understanding of single elements (Holzbrecher & Tell, 2006). In general, part and whole are related in a circular way when it comes to the process of understanding and interpretation (Hoy, 1982). The circle developed from the principle of understanding a text, to the principle of understanding one’s own nature and situation (Hoy, 1982; Thompson et al., 1994). It can be applied to many fields of study dealing with various subjects (Holzbrecher & Tell, 2006). Two examples should make this concept clearer. In the hermeneutics of paintings or photo analysis the context refers to the total stock of paintings from a certain era, whereas the part is a single painting from this era. The aim is to establish the extent to which the single picture represents the era and by what characteristics. As a consequence the understanding of this era and of each single picture itself is enhanced (Holzbrecher & Tell, 2006). In educational studies the context may refer to the work with children as a whole. The parts would be single processes within the context, such as group dynamics, learning processes and the usage of tools. The objective is to establish how the parts are related to the context and vice versa (Holzbrecher & Tell, 2006). Researchers are not able to interpret a collage by only looking at its single images in isolation; it is necessary to relate the images to each other as well as the overall context (Moisander & Valtonen, 2006). In this thesis parts refer to the single elements and objects in a collage, whereas the context refers to the collage in its entirety and the context within which it was created. An understanding of single pictures within a collage is crucial in the understading of the entire collage. By the same token, the whole collage and its context are important in the understanding of the single pictures chosen. The hermeneutic circle involves analyzing a text or picture on three levels and links the concrete to the overall context or situation. The three levels of language, developed by Morris (1972), are semantics, pragmatics, and syntactics and make up the structure and meaning of a text (Niedermair, 2001). Semantics is the meaning and signification of signs not related to the context; pragmatics is the meaning of signs based on context, sender, receiver and their relationships; and syntactics is the structure of signs (Niedermair, 2001).

However, the concept of the hermeneutic circle is not undisputed. Firstly, it is difficult to “enter” the circle, as this is only possible with prior knowledge (Bardzell, 2009). “You can’t understand Shakespeare without being in a position to understand Shakespeare; how can you get in such a position without reading Shakespeare?” (Bardzell, 2009). Even more important is the fact that the circle interferes with objective knowledge, as reasoning is circular and does not allow for justification (Bardzell, 2009). Therefore, opponents encourage researchers to break out of the circle as it is only possible through the aid of empirical science, which requires going beyond personal interpretation and subjectivity (Taylor, 1971). But pure empiricism does not work, as it is not possible to break out of the circle at all (Bardzell, 2009). Personal understandings are always subjective and never free of individual assumptions (Thompson et al., 1994), as “a person’s understanding of a situation results from an interpretation that is based on cultural traditions of meaning” (Thompson et al., 1994, p. 434). However, proponents state that it is possible to achieve scientific progress even within the hermeneutic circle (Bardzell, 2009). The major benefits are that the circle enhances perception and makes scientific works susceptible to reflection and critique. It also allows participation in the development of a certain field of study; it enables researchers to use contributions of others in their works, to produce knowledge, and to critique and to improve others’ works (Bardzell, 2009).

In order to be able to apply the concepts of the hermeneutic circle and three-level analysis on collages, certain methods are required. Metaphor analysis and color theory are used to analyze semantics of the collage by identifying and interpreting image schemata and major colors used throughout. The communication model by Schulz von Thun (1992) analyzes the collage’s pragmatics by investigating the context and relationships of the collage and its creator. To conclude, structural analysis identifies the syntactics of the collage. The remainder of this chapter explains the relevant concepts in detail.

3.2 Metaphor analysis in interpreting collages

Collage creation, the technique this thesis examines, is a method that reproduces mental images and generates visual metaphors (Rook, 2006). Metaphors offer ways to penetrate consumers’ thoughts and serve as windows into people’s minds (Joy, Sherry, Venkatesh, & Deschenes, 2009). Beyond that, much brand knowledge is embedded in mental images, metaphors and emotions, as people mainly process information metaphorically (Zaltman, 1997). Metaphor analysis is the first concept used in the interpretation process of collages, as systematic elicitation and analysis of metaphors significantly augment the knowledge gained (Zaltman, 1997; Woodside, 2004; Rook, 2006). In general, metaphor analysis is an attempt to understand, to clarify, to unfold and to interpret metaphors (Niedermair, 2001). More specifically metaphor analysis is known as a qualitative research method based on the cognitive metaphor theory by Lakoff and Johnson (2008) (Niedermair, 2001).

3.2.1 Metaphors as cognitive constructs

Metaphors play an essential role in human communication, as they reduce complexity by facilitating expression and understanding (Schmieder, 2006). Metaphors highlight certain aspects and minimize others via projecting certain features of one object onto another (Barner, 2008; Schmieder, 2006). The novelty and abstractness of certain situations may not enable customers to express genuine feelings by words alone (Joy et al., 2009) and may require vehicles for expressing situations that are difficult to communicate (Barner, 2008). Metaphors possess such qualities (Barner, 2008) and provide a simple method to make sense of everyday experiences (Lakoff & Johnson, 2008). They enable people to convey a complex situation instantly with few words (Barner, 2008; Joy et al., 2009) by connecting a broad array of interrelated thoughts, feelings and beliefs (Barner, 2008).

Aristotle laid the foundation for today’s definitions of metaphors by stating that a metaphor was the transfer of one word to another word (Niedermair, 2001). In the 1980s researchers (Johnson, 1990; Lakoff and Johnson, 2008) started recognizing metaphors as cognitive phenomena and structures, which were part of thought rather than language (el Refaie, 2003). The cognitive metaphor theory by Lakoff and Johnson (2008) connected linguistics and social sciences and provided new opportunities for operationalizing metaphor analysis (Niedermair, 2001). Today, most approaches to metaphor analysis are based on this concept, including Schmitt’s (1997, 2003), which is applied in this thesis. In a cognitive sense, metaphor analysis describes the understanding and experiencing of a concept by transferring one concept for another (Seitz, 1998; Lakoff & Johnson, 2008). Lakoff and Johnson (2008) developed their metaphor theory by defining metaphors as a matter of concept, such as an event, activity, idea, or emotion. According to this, people understand the environment through experiences that are organized in terms of metaphorical concepts (Seitz, 1998). Those metaphorical concepts are expressed systematically and metaphorically by language making metaphors an inevitable part of human language, mind and communication (Lakoff and Johnson, 2008). In short, Lakoff and Johnson (2008) identified metaphorical concepts via language. This thesis applies this theory on collages and investigates visual elements on collages instead of spoken or written words. The example “argument is war” by Lakoff and Johnson (2008) should assist in the understanding of this rather complex construct. “Argument is war” is a metaphorical concept, in which the concept of argument is transferred to and structured by the concept of war. This means that people’s actions, behaviors and understandings during an argument are structured by the “argument is war” concept (Lakoff and Johnson, 2008). This is seen in expressions used in everyday life such as “attack the weak point in an argument”, “win an argument” or “shoot down arguments”. Note that it is not about the very words used, but about the concept behind these expressions (Zaltman & Zaltman, 2008). Other examples are “life is a journey”, “ideas are products” and “eyes are containers”. This shows that metaphors are “pervasive in everyday life, not just in language, but also in thought and action” (Lakoff & Johnson, 2008, p. 12) and that “the way we think, what we experience and what we do every day is very much a matter of metaphor” (Lakoff & Johnson, 2008, p. 11).

3.2.2 Metaphors versus image schemata

This section explains the relationship between metaphorical concepts, image schemata and metaphors. Broadly speaking metaphorical concepts are based on image schemata and expressed by metaphors. Image schemata are “a pervasive, irreducible, imaginative structure of human understanding that influences the nature of meaning and constrains our rational inferences” (Johnson, 1990, p. xii). Put another way image schemata are unconscious and abstract structures of experiences, images or perceptions (Johnson, 1990) that emerge from interactions between brain, body, and society (Zaltman & Zaltman, 2008). A metaphor is “understanding and experiencing one kind of thing in terms of another” (Lakoff & Johnson, 2008, p. 13). Metaphors provide a vehicle for expressing experiences, images or perceptions, structured as image schemata in mind, by projecting attributes of one experience, image or perception onto one of a different kind (Johnson, 1990). Metaphors give customers an opportunity to give voice to their mental representations (Johnson, 1990). The terms “image schema” and “metaphor” are used differently in literature. This thesis uses the term “image schema” to denote a chief cognitive structure in the human mind that is fundamental to our understanding of what and how we perceive (Zaltman, 1997), and the term “metaphor” to denote a verbal or visual expression of an image schema. Zaltman and Zaltman (2008) provided another definition by defining three levels of metaphorical thinking, namely: surface metaphors (metaphors), metaphor themes (metaphorical concepts) and deep metaphors (image schemata). Surface metaphors are used in everyday language and people are aware of using them, whereas deep metaphors underlie metaphor themes and are unconscious. “Money is a resource” is a deep metaphor, providing the basis for the metaphor theme “money is liquid”, which in turn offers the foundation for the surface metaphor “money runs through his fingers” (Zaltman & Zaltman, 2008). On this basis metaphors and image schemata are related phenomena that work on different levels within the human brain. As image schemata are found in memory systems which function at a more nonverbal and unconscious level, the conscious concept of metaphor is needed as a vehicle of expression (Johnson, 1990; Epstein, 1994). Therefore metaphors essentially represent our thinking processes and can be found in every form of communication (el Refaie, 2003), implying that metaphors are key to accessing image schemata, which in turn allows access to brand knowledge (Zaltman, 1997). To summarize, image schemata are the abstract concepts that make sense of our experiences, and metaphors are the instruments we use to express and react to those concepts (Johnson, 1990). Something unknown, unresolved or problematic is represented in terms of something familiar (el Refaie, 2003). The actual referent of a metaphor is an abstract concept, whereas the figurative term is related to basic human experience (Lakoff & Johnson, 2008). This thesis assumes that visual elements on collages are metaphors based on image schemata and that metaphors help to reveal image schemata and their meanings and therefore brand knowledge. The remainder of this section illustrates different types of image schemata and should assist in clarifying these concepts. Types of image schemata

Johnson (1990) provided the following list of image schemata:

illustration not visible in this excerpt

Table 1: Image schemata by Johnson (1990, p. 126)

To provide a better understanding of those concepts, the force schema is explained in greater detail. Force is one of the most common schemata and can be divided into several subcategories including: counterforce, blockage, enablement, compulsion, restraint removal, and attraction (Johnson, 1990). Each schema deals with certain experiences people encounter and structures them. Counterforce schemata refer to experiences related to forces that work against each other, such as a box fight, a car accident, or strong winds (Johnson, 1990). Blockage deals with obstacles resisting a force, such as a river, a wall, or a dead end. Enablement is the experience of being able to do something and to perform an action, such as driving a car, picking up a stone or writing a text. Compulsion is a schema referring to external forces, such as winds, water, or other crowds. Restraint removal is related to the blockage and enablement schemata and occurs when people are able to do something because a potential barrier is missing, such as entering a building, or crossing a street. Attraction represents a force that has attractive power, such as a magnet, or a vacuum cleaner (Johnson, 1990). Force is the main image schema underlying the “argument is war” concept and can be expressed by various spoken or written metaphors. Therefore the expressions used allow for identifying the underlying image schemata. “He attacked every weak point in my argument” is based upon the counterforce schema. “I demolished his argument” has the restraint removal schema as its basis. “She had no chance to resist my arguments” is related to compulsion. “He defended his arguments strongly” builds up on the blockage schema. Different metaphorical concepts have different schemata underlying: “Life is a journey” (path schema), “ideas are products” (transformation schema, process schema) and “eyes are containers” (container schema).

Another very common schema is balance and refers to experiences that involve keeping the balance and being in equilibrium (Johnson, 1990). This involves physical equilibria, such as standing straight or riding a bike; psychic equilibria, such as finding the balance between work and free time; and perceptual equilibria, such as looking at an asymmetrical picture. The third schema illustrated is the path schema structuring people’s spatial and temporal experiences (Johnson, 1990). The path schema consists of three parts: a starting point, an end point, and a route connecting those points. There are physical paths such as from your bed to your bathroom, from your apartment to your garage, or from your block to your work place; as well as imaginative ones such as from your balcony to the stars above, or from this moment to a moment in the future (Johnson, 1990). The container schema represents containment consisting of an inside, an outside, and boundaries such as a building or a purse (Raubal, Egenhofer, Pfoser, & Tryfona, 1997). In addition to this bodies and minds are containers as well (Johnson, 1990). The part-whole schema deals with the decomposability of entities, such as a tree consists of leafs, branches and roots, or a nation and its states (Rosa, 2001). The center-periphery schema is a space schema and defines distances, both physical and mental. People experience their world from an individual perspective and therefore some experiences and persons are more important or central than others (Johnson, 1990). The transformation schema is about changes. If a person changes in shape, time passes by, or a group of people alters their attitudes, the transformation schema is at work. The scale schema is best described by the “more is up” metaphor (Lakoff & Johnson, 2008). The schema is about increasing and decreasing substances, such as filling up a glass of water; or degrees of intensity, such as temperature (Johnson, 1990). The more water or temperature the larger the scale becomes. Furthermore the cycle schema represents all kinds of circles; temporal ones, such as time; natural ones, such as the tides; or conventional ones, such as politics and culture (Johnson, 1990). The link schema is omnipresent and a very dominant schema. Links represent various kinds of connections, including: temporal, causal, spatial or functional connections. Without links mankind would not exist (Johnson, 1990). The identity/matching schema consists of two single schemata found by Johnson (1990) that are combined for the purposes of this thesis. This schema is about the similarity and fit of two or more subjects (Johnson, 1990). For example, two persons are similar in their appearance/behavior or a man and a woman are well suited. Moreover, the splitting schema shows the divisibility of entities (Johnson, 1990). For example a crowd can be divided into men and women, or into brunette, blond, black and red haired. However, the list of image schemata by Johnson (1990) can be extended freely in order to better fit the working method or subject treated. Raubal et al. (1997) added a front-back and verticality schema, whereas Rosa (2001) modified the path schema and added the agency schema. Amant, Morrison, Chang, Cohen and Beal (2006) identified eight key-schemata, namely space, force, containment, locomotion, balance, identity, multiplicity, and existence.

According to collages, it is assumed that participants use visual metaphors in the form of images as vehicles to express their underlying image schemata. For example, using a picture of sneakers in a collage can have various schemata underlying, from container to enablement to merging to transformation. The number and dominance of schemata related to a picture depends on the context, and as such the examination of an element in its appearance and context in order to identify the relevant image schemata is required. Sneakers have the form of containers that people can step in; beyond that, wearing sneakers may enable people to perform better by merging and being transformed. A detailed description of the application of metaphor analysis on collages is provided in chapter 4.

3.2.3 Visual metaphors as sources of brand knowledge

There are two types of metaphors: verbal and nonverbal. However, visual (nonverbal) metaphors have been neglected by researchers in the last two decades, although their use is significant in expressing feelings as well as experiences and new aspects of brand knowledge may be deduced (Barner, 2008). Methods that emphasize visual metaphors, such as the collage technique, may be an appropriate instrument.

Richards (1936) was one of the pioneers in providing a definition of visual metaphors. Richards (1936) described them as thinking processes about one thing in terms of another. A building with a facial expression was one such example (Seitz, 1998). El Refaie (2003) explored the ways in which metaphors were expressed in the visual mode and found three major arguments. The first was that visual metaphors are based on metaphorical concepts (in line with cognitive metaphor theory discussed earlier in this thesis). Although the number of works on visual metaphors has increased in recent history (Morris, 1993; Forceville, 1994; Carroll, 1996) there is still no consistent definition of visual metaphors (el Refaie, 2003). El Refaie (2003) suggested defining visual metaphors rather as visual expressions of metaphorical concepts, which was consistent with the current dominant paradigm in metaphor analysis, as opposed to a focus on formal qualities. The second of el Refaie’s (2003) arguments was that visual metaphors were less straightforward and much more susceptible to the socio-political context than their verbal counterparts. The boundaries between the literal and metaphorical are blurred and visual metaphors are more ambiguous than verbal ones. However, rather than investigating whether a metaphor is literal or metaphorical, it makes more sense to determine whether a metaphorical concept has become the natural way of expressing experiences or not (el Refaie, 2003). Moreover, often the concepts of visual metaphors are not present visually and can only be implied by the context, which requires an understanding of the socio-political context in order to determine meaning (el Refaie, 2003). Thirdly, the specific forms of visual metaphors have a great influence on their meaning and impact. One of the major differences between verbal and visual metaphors is that the latter often refer to situations that have no verbal translation, implying that on occasion images are more effective in expressing meaning than words. This also means that visual metaphors are more implicit concepts and therefore more susceptible to different interpretations (el Refaie, 2003). The next section deals with color theory, the second approach of the interpretation procedure.

3.3 Color theory in interpreting collages

Consumer’s personality, mind and behavior are determined by basic influences, of which colors are part (Braem, 1998). As a consequence, colors have an impact on humans, whether this impact is conscious or not. As an example, by wearing colorful clothes and preferring or rejecting certain colors customers are sending permanent signals (Riedel, 1986). Colors trigger automatic-unconscious reactions and associations based on past personal experiences (Heller, 1998) such as blue skies, green meadows, or a red fire (Riedel, 1986). Reactions may include feelings and a direct impact on bodily processes, such as: heart beat, breathing frequency or blood pressure (Braem, 1998). Associations may include preferences, images, and words.

3.3.1 What colors reveal about customers

Before explaining the importance of colors in marketing and why color theory is helpful for interpreting collages, the phenomenon of synesthesia is introduced. Synesthesia describes a process of perceiving a sensation by two different senses simultaneously (Cytowic, 2003). The experience of a sensation with one sense leads to an automatic experience of the same sensation with another sense (Cytowic, 2003). There are several relations between different senses in the field of synesthesia, including those between colors and emotions, colors and music, or line drawings and emotions (D’Andrade & Egan, 1974). The relation between colors and emotions suggests that colors have emotional value and that there are happy colors, sad colors, exciting colors and many more (D’Andrade & Egan, 1974; Singh, 2006). There are two major schools of thought explaining this phenomenon. The first takes culture as the basis of synesthesia, stating that culture contains metaphors, rituals and symbols that link colors to different emotions. According to this perspective, emotions of colors are learned from experiences (D’Andrade & Egan, 1974; Aslam, 2006). The second explanation uses stimuli as the basis for synesthesia and suggests that stimuli trigger distinct, innate and unconditioned responses which serve as connective links combining and eliciting different sensory experiences (D’Andrade & Egan, 1974; Aslam, 2006). Though this is the case, this thesis is particularly interested in meanings and brand knowledge deduced by colors rather than emotions. It is important to note that the emotional responses triggered by colors influence consumer perceptions and behaviors regarding products and brands (Aslam, 2006; Heath, 1997). Without words, colors are able to communicate emotions whilst also conveying characteristics of products and brands. Such characteristics can be as wide-ranging as: sexiness, fragility, durability, youth, and freshness (Heath, 1997). Colors create attention, convey messages, and generate feelings that influence emotional responses and behavioral intentions (Kotler, 1973). Firms and managers have become more keenly aware of the effects of colors and consequently use them knowingly in marketing related to product design, packaging, advertising, branding, corporate identity and consumer perceptions (Aslam, 2006). Colors have the power to reveal product attributes such as price, quality, or flavor. Colors also have the ability to evoke product associations. For example, in the United Kingdom white is associated with cheap and low-quality products, red is perceived as conceited and average priced and beige is seen as boring and expensive (Aslam, 2006). Furthermore, colors can be used to communicate corporate positions and are an effective method to differentiate from competition by launching a product with an atypical color in its category. In the United States for example, blue is associated with responsible financial services, whereas yellow stands for bright and exciting companies (Aslam, 2006). Finally, but importantly, colors reveal one’s own individual personality and self-image, as people buy colors that reflect themselves (Aslam, 2006). This means that color theory helps to reveal experiences, perceptions and meanings of consumers in relation to brands, products and companies. Since this information is very useful in marketing, color theory is a part of the interpretation procedure in this thesis.

3.3.2 Same color, different meaning

While some colors have universal meanings across countries, others represent and offer different meanings and preferences based on culture, ethnicity, region and income (Aslam, 2006; Heath, 1997). White for example symbolizes mourning and death in East Asia, but happiness and purity in North America. Blue represents death in Iran and purity in India, but coldness and masculinity in Western Europe (Aslam, 2006). There are numerous examples showing cultural differences of colors and consequently it is crucial that managers are aware of the different perceptions and interpretations of colors in various geographic areas (Aslam, 2006). This thesis focuses on color theories in Western Europe only. Interestingly however, there are not only differences based on culture, even people from the same culture associate different meanings with the same color and different colors with the same meaning (Heller, 1998). This is possible because the perception of colors is context-dependent, and same colors are associated with different experiences, personal or cultural. Beyond that, even small differences in the hue of a color can impart very different meanings (Heller, 1998). For example, whilst green is the color of youth and represents immaturity (as green tomatoes are unripe), green is also the color of nature representing health and freshness, as green leafs belong to healthy plants (Riedel, 1986). As a consequence, in terms of brands and products green is associated with healthy foods, but it can also represent young and thriving organizations (Aslam, 2006). This shows that colors have an impact on consumer behavior and their perceptions, influence purchase decisions and define brand identity (Heath, 1997). Color theory can reveal valuable information about brands, products and companies.

The remainder of this section illustrates the major meanings of the seven most common colors based on Riedel (1986) and Heller (1998). Riedel (1986) developed a thorough psychological color theory based on findings and results by several researchers and 20th century studies (Stefanescu-Goanga, 1912; Frieling & Auer, 1961; Itten, 1962; Heiss, Halder, & Höger, 1975). Riedel (1986) utilized and illustrated those theories systematically and meaningfully. Nine different colors (red, blue, yellow, green, orange, purple, brown, black, and white) are discussed in terms of associations, psychological impact, symbolism and archetypes (Riedel, 1986). Heller (1998) complemented Riedel (1986) very well and conducted her own study of color theory. 1888 persons were asked to assign colors to forty different terms. Heller (1998) used the findings and results to illustrate the meanings of thirteen colors (blue, red, green, black, pink, yellow, white, purple, gold, brown, gray, silver, and orange) in terms of psychological, symbolic, cultural, political, traditional and creative effects. Color theory of collages in this thesis is based upon the meanings outlined below. White – the “godlike” color

The most common associations with the color white are: light, brightness, and snow. White also has an important status as a godlike color, as it stands for faith, belief, piety, sacrifice and innocence. It is the color of the beginning and resurrection as well as eternity (Heller, 1998). White clothes play a major role at rites of initiation and passage and are associated with bright days, cheerfulness and vitality (Riedel, 1986). White represents openness, innocence, purity, universality, perfection, ideal and the good (Riedel, 1986; Heller, 1998). White can be the color of cleanliness and quietness. It can also represent tenderness and relate to hygiene and hospitals or epitomize good virtues, such as honesty, intelligence, and precision. Moreover it is the color of science referring to dispassion, objectivity, functionality, modernity; and the color of the north referring to winter, cold, and snow (Heller, 1998). White also has a negative connotation as color of emptiness, loneliness, the unknown and a lacking in color (Riedel, 1986; Heller, 1998). Black – the “negative” color

Black has a strong impact on people, as it is based on the experiences of nightfall and the extinguishment of color. Black is associated mainly with death, night, darkness, deepness and also coal or lava (Riedel, 1986). It is the color of negative feelings: looking at the dark side, having a black heart, giving a black look and black humor are only a few examples. Black is the color of dirt and disgrace (blackening, blackmail, blackball or black guard) and the color of bad luck and misfortune (black day, black cat, and black ice). Black reverses meanings into the opposite, such as: red from love into hate, yellow from sociability to selfishness, and blue from harmony into harshness (Heller, 1998). Black is a melancholy color and is often connected to repression, blockage and dissociation. It is the absolute color and represents nothingness, lack of life and evil (Riedel, 1986). Black is the color of mourning, as it is a sign of mourning to ignore one’s personal appearance (Heller, 1998). Clerics also often wear black as it is an ascetic, conservative and plain color (Riedel, 1986). Also importantly, black can be the color of individuality and differentiation (Heller, 1998). Red – the color of fire

The most common associations with red are fire (inflammatory speech and enlightenment), heat and blood. Other nuances are associated with sexual desire, danger, closeness, or strength, power and love (Riedel, 1986). The emotional impacts of red are excitement, warmth, stimulation, and activity; red raises blood pressure, pulse and breathing frequency. It is the urge to do something, to be active and to make an impact (Riedel, 1986). From love to sexuality, to hate, all feelings that make our blood boil are related to red, as blood is a vivid part of our body (Riedel, 1986). The more positive a passion, the lighter the red; the more negative a passion, the darker the red (Heller, 1998). Red is conspicuous in many flags and a political color, as it is related to Communism (Riedel, 1986). Red is also the color of the legally and morally forbidden (red-light district) or the color of warning and help (road traffic and the Red Cross) (Heller, 1998; Riedel, 1986). Red is the color of war as well as justice and red gives power, thus warriors and judges wore red (Heller, 1998). Red is the color of correction, as the red pencil is used for corrections in school, and the color of advertising, as red is active, dynamic and eye-catching (Heller, 1998). Blue – the color of infinite dimensions

The most common associations are sky/heaven, sea/ocean, cold/coolness, and ice (Riedel, 1986). Blue is a cold color and inappropriate for the interior (Heller, 1998). Blue evokes skies in different moods, the atmosphere of the earth and seas, ranging from light turquoise to dark blue ultramarine (Riedel, 1986). Colors have an impact on our perspective and blue is the color of infinite dimensions, as cold colors seem farther away than warm colors (Heller, 1998). Sky and sea refer to unlimited distance and depth, as well as infinity (Riedel, 1986). Blue is the color of faithfulness and desire, as both are connected to distance. Faithfulness can only be proved and desire can only occur, if somebody or something is far away (Heller, 1998). In general, t he color blue triggers wishful, melancholic, calm, and dreamlike emotions. It is the color of introversion, sensibility, secureness and connectedness, in spite of the fact that dark blue often elicits depressing, serious and sad feelings (Riedel, 1986). Blue can also be the color of imagination, representing a utopian idea whose realization is far away. Finally, blue represents masculinity, as it stands for courage, performance, wisdom, intelligence and accuracy (Heller, 1998).



ISBN (eBook)
ISBN (Paperback)
1.9 MB
Institution / Hochschule
Leopold-Franzens-Universität Innsbruck – Institut für Strategisches Management, Marketing und Tourismus
2012 (Mai)
Marketing Brand knowledge Brand Projective technique Qualitative Collage analysis Collages Color theory Knowledge retrieval Retrieval Representation Consumer memory Empathy Visual Image schemata Mental models Semiotics Hermeneutics Collage technique Metaphor analysis People's minds

Titel: How collages reveal your deepest thoughts