Starting From Scratch: Rethinking Brand Image Research and Identifying Cues and Context As Influential Factors

ABSTRACT - The paper proposes to start brand image research from scratch. It is argued that presently applied measurement methods hold an overly reliance on the presence and relevance of certain kinds of associations in consumer memory. First, the concept of brand representation is introduced, grounded on frame theory. Next, activation cues and evaluation context are elaborated upon in the paper, as these factors might bias research findings. The paper intends to make both advertisers and market researchers aware of bias in present brand image research: by relying on method assumption methods may create rather than establish brand images.

Citation:

E.M. Timmerman (2001) ,"Starting From Scratch: Rethinking Brand Image Research and Identifying Cues and Context As Influential Factors", in AP - Asia Pacific Advances in Consumer Research Volume 4, eds. Paula M. Tidwell and Thomas E. Muller, Provo, UT : Association for Consumer Research, Pages: 156-161.

Asia Pacific Advances in Consumer Research Volume 4, 2001      Pages 156-161

STARTING FROM SCRATCH: RETHINKING BRAND IMAGE RESEARCH AND IDENTIFYING CUES AND CONTEXT AS INFLUENTIAL FACTORS

E.M. Timmerman, University of Amsterdam, The Netherlands

ABSTRACT -

The paper proposes to start brand image research from scratch. It is argued that presently applied measurement methods hold an overly reliance on the presence and relevance of certain kinds of associations in consumer memory. First, the concept of brand representation is introduced, grounded on frame theory. Next, activation cues and evaluation context are elaborated upon in the paper, as these factors might bias research findings. The paper intends to make both advertisers and market researchers aware of bias in present brand image research: by relying on method assumption methods may create rather than establish brand images.

INTRODUCTION

Conceptualization of and research on brands has developed in recent years from the conception of brands being a necessary identification mark for products, into an awareness of the growing benefits of owning brands instead of products. This awareness led to attempts to establish the surplus value brands lever over mere products. The establishment of the so-called brand equity, the 'value of brands’, has become a major focus of research on brands.

A great number of authors address the notion of brand equity nowadays (D. Aaker, 1996; Feldwick, 1996; Franzen, 1998; Kapferer, 1994; Keller, 1998). Until so far there seems to be not much consensus on the definition of brand equity. However, based on a great number of conceptualizations of brand equity, Franzen (1998) concludes that brand equity includes four aspects: (1) the presence of a brand in the memory of consumers, (2) its influence on purchase behavior, (3) the effects of purchase behavior on the market position and financial results of a brand and (4) the financial value of a brand as an immaterial asset owned by a company, which can be placed on the balance sheet and is expressed when the selling of the company is at stake. According to these four brand equity aspects, the value of a brand stems from the content of consumer’s memory, and research on this aspect is thus essential for brand equity research.

The brand-related content of consumer memory is most often referred to as brand image. The term image is derived from psychological literature, and has been widely applied in marketing research for years but unfortunately has become more of a generic term. There seems to be no unified definition of brand image. Poiesz (1989) tried to make sense out of the quantity of definitions of brand image by distinguishing three main categories of definitions, differing in level of elaboration: (1) the high elaboration approach defines brand image as a network of meanings stored in memory. This approach employs a means-end chain orientation to the definition and operationalization of images; (2) the intermediate level of elaboration views images as the theoretical and operational equivalent of attitudes in the Fishbein tradition, in which attitudes are viewed as a function of the combination of salient beliefs and belief evaluations; (3) the low elaboration approach views images as general, holistic impressions or perceptions of the relative position of a brand among its perceived competitors. Poiesz’ work illustrates the absence of consensus among researchers on the brand image concept, which caused the term brand image become diluted over years. Although the absence of consensus on the concept of brand image could be of less personal interest to individual researchers, it has a substantial, but unfortunately too often unrecognized or neglected effect in brand image research. The multitude of definitions led to an equal multitude of measurement instruments. As each definition tends to be unique its operationalization might hold a narrow focus on specific brand image aspects. This is especially apparent in the competitive area of applied market research where companies try to distinguish themselves by developing unique methods. As this practice underlies the results of conducted research, it has direct consequences for advertisers, being the clients of brand researchers. As our Holy Grail is to find the ideal way of revealing consumer memory, the question is how can we pass by these assumptions?

THE PRACTICE OF BRAND IMAGE MEASUREMENT

Nowadays, a lot of research agencies state to have a unique ability to measure a brand’s image by applying its own developed measurement method. For illustration, with respect to brand research, Bouwman (1999) collected well over 70 different methods applied by market research agencies in The Netherlands. In practice, an advertiser in need of insight into the consumers perception of his brand will approach one of themany research agencies whose research method seems valid, reliable, appealing, surprising, inventive or whatever to him/her. Since all measurement instruments created by these different research firms are based on personal definitions of brand image, the resulting and reported brand image (which is the end product to the client) is definition-specific. An advertiser becomes more or less dependent on the brand researcher and his conceptual point of view on brand image (especially those advertisers who are unable to conduct more than one study because of low research budgets or time restrictions). For example, researchers conceptualizing brand image as holistic impressions will be more inclined to focus on associations of a psychological and/or symbolic nature like brand personality, feelings and expressive functions and will focus their research methods on these aspects. A reported brand image will thus be described in these kinds of terms. On the other hand, researchers who go less in-depth and conceptualize brands in terms of concrete attributes will probably start their research from associations like quality, pricing and (effects of) usage, and base their description of the brand’s image on these terms. Ergo, had the advertiser chosen another research agency, the reported brand image would be different. One could question whether these results reflect the consumer’s memory content. Would it be both to an equal degree, would it be either one or none of these?

Most brand image definition (and by that their consequent methods) hold (implicit) assumptions that certain associations (1) are representatives of the brand under study, and thus present in the respondent’s memory, and (2) are relevant to the brand. For illustration, suppose a brand researcher defines a brand in terms of its personality. Based on this definition, the researcher has developed measurement methods using personality statements or pictures of faces or such. When an advertiser wants to obtain insight on his brand’s image, and contracts this researcher, the latter will present respondents with his personality oriented methods and present a personality profile of the brand as a result to the advertiser. Although for some brands this procedure might be highly valid and reliable, one could wonder if an overrating of the presence and relevance of the brand personality associations has produced an artificial result. The method might not have measured activated associations, but might have created non-existing associations and hence measured artifacts.

Passing the addressed assumptions would mean to start brand image research from scratch. Ideally research should start without focusing on specific associations at forehand but make allowance for all associations that possibly exist in the respondent’s memory. The first research objective is to know by which kinds of associations the brand is primarily represented in memory (e.g., being actually stored in memory and also of relevance to the brand), and what their nature is (concrete versus abstract or symbolic). Only then are we able to select measurement methods which are well designed to measure the content of the association and obtain valid, unbiased and activated (instead of created) consumer knowledge.

BRAND REPRESENTATION

Starting from scratch also implies to step back from the diluted concept brand image. As brand image definitions differ widely on assumptions on kinds of associations underlying the brand storage, we apply a broader concept that explicitly holds no assumptions on this matter. We introduce the notion of brand representation as starting point for 'start from scratch’ brand image research. Brand representation refers to the cognitive frame on which the memory representation of a brand is build, and is defined as:

'A brand representation is the brad specific collection of product-, brand-, and consumer-related attributes (consisting of all aware and unaware knowledge of, feelings and attitudes towards a brand) which an individual holds in memory, a subset of which becomes active in memory in a moment specific configuration dependent on activation cues and context’.

This definition resulted from addressing the following issues: (1) how are concepts stored in general (and therewith the concepts of brands); (2) which elements underlie the representation of brands and as such are the building blocks of the representational frame; and (3) how can we activate and establish these representation elements in applied research on specific brands. The next paragraphs will elaborate on these issues.

FRAME THEORY AND BRAND REPRESENTATION

If we make allowance for all associations possibly present in consumer memory, we first need to know how knowledge is actually stored in memory. In cognitive science, many memory models have been proposed (see for an overview Barsalou & Hale, 1993), one of the most recent being frame theory (Barsalou, 1992; Barsalou, 1999a; Barsalou & Hale, 1993; Barsalou, Huttenlocher, & Lamberts, 1998; Barsalou et al., 1993). Following frame theory, the representation of knowledge can be conceptualized as a frame structure wherein attribute-value sets are the storage elements. Frames are internal cognitive structures representing constructs in memory. Frames are comprised of attributes, which take on different defining values. For instance, the concept of 'watermelon’ is comprised of attributes like color, taste and shape, the respective defining values of which (like 'green’, 'watery’ and 'oval’) discriminates this fruit from other fruits. Furthermore, according to frame theory, concepts are neither context-independent nor universal, but are situated and local. Situated means that people typically represent concepts in the larger situation that contain them. For example, a person’s concept for chair represents chairs in particular situations, such as chairs in a living room. Local refers to the notion that the concept constructed by a person on a particular situation is local to that situation, only covering exemplars in that situation and not attempting to cover all exemplars universally. Taking the chair example, a person’s concept for chair only covers those chairs in the current situation containing it, such as the chairs in a particular living room (Barsalou et al., 1993). These notions imply that only a subset of the total set of concept-building attributes will be activated, dependent on for one the context in which it appears.

When we apply frame theory to brand-related consumer memory, it first implies that a brand representation in memory is a frame of attributes, each with a specific value (see also Lawson, 1998). We consider attributes as association types, the value on which is the actual association. For illustration, the association with McDonald’s and its golden arches is considered as that the attribute 'brandmark’ (defined by its value 'golden arches’) underlies the representation of the brand McDonald’s. Second, a brand representation is context dependent (situated and local), meaning that the manifestation of a brand (in form of a brand image) may vary over situation.

INVENTORY OF BRAND REPRESENTATIONS ATTRIBUTES

The next issue is to find the attributes that might in general constitute brand representations. If brand image research starts from scratch, and gives allowance for all attributes to be stored, it would be insightful to list the possible collection of attributes that might underlie the brand’s representation. As part of this study, insights from practitioners nd academicians who have proposed models in the literature on brand image have been gathered. Some of these are based on the perspective of the consumer (e.g., Biel, 1992; Brucks, 1986; Franzen, 1998; Franzen & Bouwman, 1999; Franzen & Hoogerbrugge, 1996; Gordon, 1991; Keller, 1993; Keller, 1998; Marder, 1997; Restall & Gordon, 1994), others based on the perspective of brand managers (e.g., Aaker, 1991; De Chernatony, 1993; De Chernatony & Dall’Omo Riley, 1998; Kapferer, 1994). Next to these general models of brand image, a great number of authors have addressed specific types of associations with brands, like for instance country of origin (e.g., Li & Wyer jr., 1994), the personality of a brand (e.g., J. Aaker, 1996; Aaker, 1997; Shank & Langmeyer, 1993; Siracuse, 1997), or image of the company (Brown, 1998; Brown & Dacin, 1997; Simonin & Ruth, 1998). However, most models of brand image components we found present abstractions of association categories. Moreover, because not all models present the same components, and as such completely overlap, each model incorporates at least some unique brand attributes.

In order to obtain a complete (as close as we can get) view of the richness of consumer memory, a combination was made of general brand image models and attribute-related research findings adding unique features. This resulted in the Inventory of Brand Representations Attributes (IBRA), a collection of attributes that most probably underlie the representation of brands. The inventory consists of 10 main types of attributes: product characteristics, product usage, price & quality, brand identifiers, brand personification, market, origin, advertisement, attitudes & purchase behavior and last, personal reference. These 10 main groups cover a total of 57 specific attributes. For illustration, the main attribute group 'brand identifiers’ includes amongst others the appearance of the brand’s logo, like its color and shape.

It is important to stress that the memory representation of any brand is a specific configuration of some or all of the attributes listed in the IBRA. This is a notion as it implies that a specific brand does not necessarily include (values on) each specific attribute per se, nor does it exclude at forehand an attribute as building block of its representation. It states that the memory representation of a brand is composed of a brand specific set of attributes, which are determined by brand specific attribute values (being associations). With respect to the categorization and discrimination of (product-) related brands, three possibilities exist: (1) a brand shares (an) attribute(s) and equal values with competitors; (2) a brand shares (an) attribute(s) with competitors, but has a different value on the attribute(s); (3) a brand has (an) attribute(s) not shared with its competitors, making it an unique attribute for the brand under question (Timmerman, 1999).

FIGURE 1

FRAME REPRESENTATION OF THE MACDONALDS

CONDITIONAL BRAND REPRESENTATION

If brand representations are context dependent, ideally the impact of context dependency should be included explicitly in applied research. Also, from cognitive science we know that processes like availability have their influence on the activation of knowledge by only activating a subset of information. In applied research, ideally we need to find a way to activate the total set. Activation of knowledge requires cues: without a cue, nothing will be activated, but too many or too specific cues might lead to creation or imagery. So, we also need to investigate the role cues play in the activation of knowledge. As Barsalou et. al. (1999a) state and show, the activation of knowledge is dependent on the questions posed. If a person is asked to describe a 'watermelon’ or a 'car’, the responses obtained will differ from responses obtain when asked to describe a 'split watermelon’ or a 'comfortable car’. The provided cue (or 'modifiers’, as Barsalou et. al. call it) yields a different activation ofattribute values, although the concept to be described remain the same. If brand attribute values need to be activated and measured, the cue(s) provided by the method are influential on the obtained results.

The definition of brand representation states that '[] a moment specific configuration dependent on activation cues and context’. The memory representation of a brand is therefore active. This means that there is no static representation, equal under all circumstances. The configuration of the subset of product-, brand-, and consumer related attributes that represent the brand depends on the way their values ('associations’) are activated by cues, and how their relevance is determined by context, and is therefore moment specific. Because the manifestation of a brand representation is dependent on cues and context, the result of a measurement will always be a conditional brand representation. The next paragraphs will elaborate on the influential role of activation cues and evaluation context.

Cues

Research by cognitive scientist has shown the differential effects of providing cues on the activation of knowledge in numerous experiments. Fundamental research in cognitive science on cues is also known as 'priming’ (Shiffrin & Raaijmakers, 1992; Zeelenberg, 1998), and is, because of its fundamental nature, mainly restricted to the effect of providing a priming word on the speed of activation of a target word. For instance, the word 'cat’ as a prime has a facilitating effect on the activation of the word 'dog’, whereas the word 'hat’ does not have this effect. Based on this kind of research, it is known for long that the use of cues has an impact on the activation of knowledge.

In the context of brand research, research on the effect of cues has been performed in different areas of interest like recall of brand related knowledge (Aaker, 1999; Costley & Brucks, 1992; Hastak & Mitra, 1996), evaluation and judgment of brand related knowledge (Maheswaran, Mackie, & Chaiken, 1992; Park, 1995). Timmerman (1999) explicitly investigated the differential effect cues play on the measurement of brand image. In this research, 56 respondents were presented with one of two beer brands and performed thought lists in order to elicit attributes. Two different methods were applied: in the first method, the minimal cue method, only one cue was provided being the brand name. Consequently respondents listed occurring thoughts for ten minutes. The obtained thoughts were later coded into one of 35 attributes. Based on the frequency of obtained attributes, the presence and relevance was established. In the second method, the maximal cue method, 35 cues were provided on cards, each one representing an attribute. After a short thought list on each card, the respondent was asked to select the cards (e.g. attributes) that would describe the brand to them. From this selection process, the presence and relevance of the attributes was established.

TABLE 1

TOP FIVE (OUT OF 35) ATTRIBUTE BASED ON MINIMAL AND MAXIMAL CUE METHOD

Looking at the five highest scoring attributes that resulted, the minimal cue method elicited mainly concrete responses, referring to attributes like advertisement, packing of beer, and social use environment on both brands (table 1). From this method, which actually is a frequently applied method in practice, it seems that the memory representation of both brands is more or less based on the same kind of attributes. However, looking at the results of the maximal cue method, brand A still seems to be comprised of the same concrete attributes, while brand B now seems to be comprised of more abstract attributes like feelings and brand ideology. Apparently the provision of cues activated attributes which are hard to verbalize in a free thought list, but nevertheless once activated indeed relevant to the second brand’s image, but not to the first brand.

If brand attributes need to be activated and measured, the cue(s) provided by the method are influential on the obtained results. It could be that by using a cue, an 'existing’ atribute value is activated, which is good. It can also be the case that an existing attribute value is not activated by lack of an appropriate cue. This is unfortunate. Worse however is the situation in which a cue creates an attribute value, and by creating it, it seems like an activated value is 'measured’. Unfortunately, a lot of presently applied methods are subject to this pitfall by relying on (implicit) assumptions of the presence and relevance of certain attributes.

Context

Context refers to the framing of the task the respondent is exposed to, and relates to the mind setting of the respondent. The influence of context has been widely studied in the field of psychological methodology, wherein context effects refer for instance to question order, time pressure, task complexity, response mode etc. (Bickart, 1992; McNamara & Diwadkar, 1996; Slovic, 1995). Context effects have also been researched to test specific models of memory (Murnane & Phelps, 1995) and as mediators in processes of encoding and retrieval (Hill, Radtke, & King, 1997). In brand research, the influence of context has amongst others been studied in the field of brand preference (Belk, 1974; Lefkoff-Hagius & Mason, 1993; Miller & Ginter, 1979; Nowlis & Simonson, 1997; Schmitt & Shultz II, 1995; Slovic, 1995), consideration set formation (Alba & Chattopradhyay, 1985; Brown & Wildt, 1992; Cowley & Law, 1995; Nedungadi, 1990), and product/brand evaluations (Graeff, 1997; Huffman, 1997; Ratneshwar & Shocker, 1991; Stapel, Koomen, & Veldhuijsen, 1998; WSnke, Bless, & Schwarz, 1998).

In practical research, the respondent is usually exposed to a measurement method without explicitly treating the respondent as a consumer or as a evaluator. In most cases in the practice of market research, the respondent is asked to perform all kinds of tasks to evaluate (a) brand(s), like filling out questionnaires, selecting pictures, making collages etc. Next to performing these tasks, the respondent is asked all kinds of questions on his/her brand attitude, purchase behavior and intentions. Then, researchers are inclined to relate these obtained results, which however might not be that much related at all. Questions on purchase behavior and brand attitude create a 'purchase mood’ for the respondent, whereas evaluating brands on all kinds of tasks does not have this mood effect. The two different mind settings might each yield: (1) specific activation of attributes, and/or (2) differing relevancy’s of activated attributes. Dependent on context, an activated attribute related to a certain brand is relevant or irrelevant to the brand’s image. For example, the 'price’ of a car brand might be extremely relevant at times of purchase ('purchase context’), but not at all when the brand is evaluated in a neutral context.

At the moment of writing, data are gathered in order to establish the differential effect of providing a purchase context on the activation and relevance of brand attributes, compared to a neutral context. This research includes the product categories coffee, detergents, TVs and automobiles, three brands from each category. Results will be obtained from 420 Dutch consumers.

DISCUSSION

This paper discusses the influential effect on the measurement of brand image of overly relying on assumptions of the presence and relevance of assocations in consumer memory. These assumptions, incorporated in measurement methods, can create artificial 'brand images’ as measurement result, whenever the method is applied to brands which are actually not represented in consumer memory as assumed. An argument is made for a 'start from scratch’ approach to brand image. First we need to know what kinds of attributes represents the brand under study, what their relevancy’s are, and only then we can start looking for the appropriate measurement method. Both advertisers as well as market researchers should become more aware of the possible impact of the definition of brand image on the end-product of brand research. There should be an increase of awareness of and support for the notion that the reported end-product of specific brand research is not an objective measure, but is at forehand framed in the researchers definitive terms.

In this paper, two methodological factors that are of influence on the activation and relevance of attributes, and hence the reported brand image, are discussed: activation cues and context. Clearly these factors are influential, but these two are definitely not the only ones. Probably the most influential factor is the respondent himself. Research in psychology has dealt with the problem since decades: creating a situation wherein people are asking questions to people in order to understand them, is a guaranty for bias. Numerous effects are found in social psychology like halo, Rosenthal and demand effects. Respondents live up to expectations created by the researcher, the experimental conditions, and in the practice of market research by the mere fact that they are invited by a research agency to come and talk about brands, product, services etc., and get paid for it as well! The respondent is malleable. This might be one of the underlying reasons for the fact that most research techniques appear to be valid: respondents are inclined to do what they are asked to do. This view might be a bit exaggerated, yet the final conclusion would be that if we want to measure a complex phenomenon as brand image, we should start from scratch for each brand. Before starting a measurement from a set of assumptions covered in a (standard applied) method, first we should try to establish what the most important attributes are to be measured while at the same time be fully aware of the impact of influential factors embedded in measurement methods.

REFERENCES

Aaker, D. A. (1991). Managing Brand Equity: Capitalizing on the Value of a Brand Name. New York: The Free Press.

Aaker, D. A. (1996). Building Strong Brands. New York: The Free Press.

Aaker, J. L. (1996). Conceptualizing and Measuring Brand Personality. Unpublished working paper, UCLA, Los Angeles.

Aaker, J. L. (1997). Dimensions of Brand Personality. Journal of Marketing Research, 34(3), 347-356.

Aaker, J. L. (1999). The Malleable Self: The Role of Self-expression in Persuasion. Journal of Marketing Research, 36(1), 45-63.

Alba, J. W., & Chattopradhyay, A. (1985). Effects of Context and Part-category Cues on Recall of Competing Brands. Journal of Marketing Science, 12, 340-349.

Barsalou, L. W. (1992). Frames, Concepts, and Conceptual Fields. In A. Lehrer & E. F. Kittay (Eds.), Frames, Fields, and Contrasts: New Essays in Semantic and Lexical Organization. Hillsdale, New Jersey: Lawrence Erlbaum Associates.

Barsalou, L. W. (1999a). Perceptual Symbol Systems. Behavioral and Brain Science, 22, 577-660.

Barsalou, L. W., & Hale, C. R. (1993). Components of Conceptual Representation: From Features Lists to Recursive Fames. In I. Van Mechelen, J. Hampton, R. Michalski, & P. Theuns (Eds.), Categories and Concepts: Theoretical Views and Inductive Data Analysis (pp. 97-144). San Diego, CA: Academic Press.

Barsalou, L. W., Huttenlocher, J., & Lamberts, K. (1998). Basing Categorization on Individuals and Events. Cognitive Psychology, 36, 203-272.

Barsalou, L. W., Yeh, W., Luka, B. J., Olseth, K. L., Mix, K. S., & Wu, L. L. (1993). Concepts and Meaning. Paper presented at the Chicago Linguistics Society 29: Papers from Parasessions on Conceptual Representations, University of Chicago.

Belk, R. W. (1974). An Explanatory Assessment of Situational Effects in Buyer Behavior. Journal of Marketing Research, 11, 156-163.

Bickart, B. A. (1992). Question-Order Effects and Brand Evaluations: The Moderating Role of Consumer Knowledge. In N. Schwarz & S. Sudman (Eds.), Context Effects in Social and Psychological Research . New York: Springer-Verlag.

Biel, A. L. (1992). How Brand Equity Image Drives Brand Equity. Journal of Advertising Research, 6, RC-6BRC-12.

Brown, J. J., & Wildt, A. R. (1992). Consideration Set Measurement. Journal of the Academy of Marketing Science, 20(3), 235-243.

Brown, T. J. (1998). Corporate Associations in Marketing: Antecedents and Consequences. Corporate Reputation Review, 1(3), 215-233.

Brown, T. J., & Dacin, P. A. (1997). Company and the Product: Corporate Associations and Consumer Product Responses. Journal of Marketing, 61, 68-74.

Brucks, M. (1986). A Typology of Consumer Knowledge Content. In R. J. Lutz (Ed.), Advances in Consumer Research (Vol. 13, pp. 58-63).

Costley, C., & Brucks, M. (1992). Selective Recall and Information Use in Consumer Preferences. Journal of Consumer Research, 18, 464-474.

Cowley, E. J., & Law, S. (1995). Altering Retrieval Sets: When Will Contextual Cues Make a Difference? Advances in Consumer Research, 22, 323-327.

De Chernatony, L. (1993). Categorizing Brands: Evolutionary Processes Underpinned by Two Key Dimensions. Journal of Marketing Management, 9, 173-188.

De Chernatony, L., & Dall’Omo Riley, F. (1998). Modelling the Components of the Brand. European Journal of Marketing, 11/12, 1074-1090.

Feldwick, P. (1996). What is Brand Equity Anyway and How Do You Measure It? Journal of the Marketing Research Society, 38(2), 85-104.

Franzen, M. P. (1998). Merken & Reclame: Hoe Reclame-Effectiviteit Brand Equity Bennvloedt. Deventer: Kluwer bedrijfsinformatie.

Franzen, M. P., & Bouwman, M. Y. (1999). De Mentale Wereld van Merken. Alphen aan den Rijn: Samson.

Franzen, M. P., & Hoogerbrugge, M. C. (1996). Het merk op weg naar de 21ste eeuw (SWOCC Research Report). Amsterdam: SWOCC, Universiteit van Amsterdam.

Gordon, W. (1991). Accessing the Brand through Research. In D. Cowley (Ed.), Understanding Brands (By People Who Do) (pp. 33-56). London, UK: Kogan Page Ltd.

Graeff, T. R. (1997). Consumption Situations and the Effects of Brand Image on Consumers’ Brand Evaluations. Psychology & Marketing, 14(1), 49-70.

Hastak, M., & Mitra, A. (1996). Facilitating and Inhibiting Effects of Brand Cues on Recall, Consideration set, and Choice. Journal of Business Research, 37, 121-126.

Hill, M. E., Radtke, R., & King, M. (1997). A Transfer Appropriate Processing View of Consumer Memory. Journal of Current Issues and Research in Advertising, 19(1), 1-21.

Huffman, C. (1997). Elaboration on Experience: Effects on Attribute Importance. Psychology & Marketing, 14(5), 451-474.

Kapferer, J., N. (1994). Strategic Brand Management. New York: The Free Press.

Keller, K. L. (1993). Conceptualizing, Measuring and Managing Customer-based Brand Equity. Journal of Marketing, 57, 122.

Keller, K. L. (1998). Strategic Brand Management: Building, Measuring, and Managing Brand Equity. New Jersey: Prentice Hall.

Lawson, R. (1998). Consumer Knowledge Structures: Networks and Frames. Advances in Consumer Research, 25, 334-340.

Lefkoff-Hagius, R., & Mason, C. H. (1993). Characteristics, Beneficial and Image Attributes in Consumer Judgments of Similarity and Preference. Journal of Consumer Research, 20, 100-110.

Li, W. K., & Wyer jr., R. S. (1994). The Role of Country of Origin in Product Evaluations: Informational and Standard-Of-Comparison Effects. Journal of Consumer Psychology, 3(2), 187-212.

Maheswaran, D., Mackie, D. M., & Chaiken, S. (1992). Brand Name as a Heuristic Cue: The Effects of Task Importance and Expectancy Confirmation on Consumer Judgments. Journal of Consumer Psychology, 1(4), 317-336.

Marder, E. (1997). The Laws of Choice: Predicting Customer Behavior. New York: The Free Press.

McNamara, T. P., & Diwadkar, V. A. (1996). The Context of Memory Retrieval. Journal of Memory and Language, 35, 877-892.

Miller, K. E., & Ginter, J. L. (1979). An Investigation of Situational Variation in Brand Choice Behavior and Attitude. Journal of Marketing Research, 16, 111-123.

Murnane, K., & Phelps, M. P. (1995). Effects of Changes in Relative Cue Strength on Context-Dependent Recognition. Journal of Experimental Psychology: Learning, Memory and Cognition, 31(1), 158-172.

Nedungadi, P. (1990). Recall and Consumer Consideration Sets: Influencing Choice Without Altering Brand Evaluations. Journal of Consumer Research, 17, 263-276.

Nowlis, S. M., & Simonson, I. (1997). Attribute-Task Compatibility as a Determinant of Consumer Preference Reversals. Journal of Marketing Research, 34, 205-218.

Park, J. W. (1995). Memory-Based Product Judgments: Effects of Presentation Order and Retrieval Cues. Advances in Consumer Research, 22, 159-164.

Poiesz, T. B. C. (1989). The Image Concept: Its Place in Consumer Psychology. Journal of Economic Psychology, 10, 457-472.

Ratneshwar, S., & Shocker, A. D. (1991). Substitution in Use and the Role of Usage Context in Product Category Structures. Journal of Marketing Research, 28, 281-295.

Restall, C., & Gordon, W. (1994). Merken: Inzicht in de Emotionele Binding. Tijdschrift voor Marketing, 28(3), 72-79.

Schmitt, B. H., & Shultz II, C. J. (1995). Situational Effects on Brand Preferences for Image Products. Psychology & Marketing, 12(5), 433-446.

Shank, M. D., & Langmeyer, L. (1993). Does Personality Influence Brand Image? The Journal of Psychology, 128(2), 157-164.

Shiffrin, R. M., & Raaijmakers, J. (1992). The SAM Retrieval Model: A Retrospective and Prospective. In A. F. Healy, S. M. Kosslyn, & R. M. Shiffrin (Eds.), From Learning Processes to Cognitive Processes. Essays on Honor of William K. Estes (Vol. 2, ). Hillsdale, New Jersey: Lawrence Erlbaum Associates.

Simonin, B. L., & Ruth, J. A. (1998). Is a Company Known By the Company It Keeps? Assessing the Spillover Effects of Brand Alliances on Consumer Brand Attitudes. Journal of Marketing Research, 35, 30-42.

Siracuse, L. (1997). Looks Aren’t Everything: An Examination of Brand Personality. IMC Research Journal, 33-39.

Slovic, P. (1995). The Construction of Preference. American Psychologist, 50(5), 364-371.

Stapel, D. A., Koomen, W., & Veldhuijsen, A. S. (1998). Assimilation or Contrast?: Comparison Relevance, Distinctness and the Impact of Accessible Information on Consumer Judgments. Journal of Consumer Psychology, 7(1), 1-24.

Timmerman, T. (1999). Establishing the Structure of Brand Images from a Feature-Based Perspective on Memory. In C. Nobles (Ed.), Developments in Marketing Science (Vol. 22, pp. 1-5). Coral Gales, FL, USA: Academy of Marketing Science.

WSnke, M., Bless, H., & Schwarz, N. (1998). Ad Hoc Categorizations and the Emergence of Context Effects. Marketing Bulletin, 10, 1-4.

Zeelenberg, R. (1998). Testing Theories of Priming. Amsterdam: Universiteit van Amsterdam.

----------------------------------------