literature review

ConsumerAttitudesRevisitedAReviewofAttitudeTheoryinMarketingResearch..pdf

Home >Business & Finance homework help >Marketing homework help >literature review

Consumer Attitudes Revisited: A Review of Attitude Theory in Marketing Researchijmr_

299 431..451431..451

Evmorfia Argyriou and T.C. Melewar1

Department of Management, King’s College, University of London, London Franklin-Wilkins Building, 150 Stamford Street, London SE1 9NH, UK, and 1Brunel Business School, Brunel University West London, Uxbridge,

Middlesex UB8 3PH,UK Corresponding author email: [email protected]

Few concepts in the marketing literature have proliferated like the concept of attitude. However, a closer look at studies investigating attitudes as consumers’ responses to marketing efforts reveals a considerable diversity in perspectives about the concept of attitude and its formation. Attitudes are considered either relatively stable object– associations, or temporarily constructed evaluations, which are formed through memory (cognitive)-based information processing or contextual and affect-based infor- mation processing. The current paper discusses and organizes these different theoreti- cal viewpoints on what attitudes are and how they are formed. By approaching the topic through an integrative lens, the paper provides a solid conceptual foundation and roadmap for marketing researchers.

Introduction

Marketing researchers and managers alike rely heavily on attitudinal surveys to estimate people’s preferential responses to a range of marketing objects, such as products, brands and advertisements (Grewal et al. 2004; Lee and Labroo 2004; Pieters et al. 2010), retail sites (Yoo et al. 1998), websites (Chen and Wells 1999; Schlosser 2003b) and Web commercials (Bruner and Kumar 2000; Stevenson et al. 2000). As a result, understanding the concept of attitude and the process of attitude formation is important to researchers and managers interested in altering consumers’ evaluation of marketing objects in order to influence their preferences and tendencies to engage in particular behaviour. However, the marketing literature is filled with diverse views on attitude (Cohen and Reed 2006; Schwarz 2006). The diversity stems from two issues surrounding the attitude concept: the first is whether attitude is a stable object-related association stored

and then evoked in memory (Fazio 1990), or a tem- porary evaluation of an object constructed in situ at the time of a judgment (Feldman and Lynch 1988; Schwarz and Bohner 2001). The second and inter- related issue is whether attitude formation is a strictly cognitive process, stemming from analyti- cal, deliberative evaluative categorization (Fishbein and Ajzen 1975; Fishbein and Middlestadt 1995, 1997) or an affective one, based on emotions and feelings which guide categorization at the time of evaluation (Schwarz 1997, 1998; Zajonc 1984). Several theorists have discussed their views on these issues, thus building a clear distinction between functional vs constructive views of attitude and its process of formation.

Functional theory supports the notion that attitudes are stored in memory in the form of object-related associations, and hence are relatively stable and evoked in memory when needed (Eagly and Chaiken 1993; Katz 1960; Shavitt 1990). For example, when a decision is to be made about an

International Journal of Management Reviews, Vol. 13, 431–451 (2011) DOI: 10.1111/j.1468-2370.2011.00299.x

© 2011 The Authors International Journal of Management Reviews © 2011 British Academy of Management and Blackwell Publishing Ltd. Published by Blackwell Publishing Ltd, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA

object, the relevant attitude will be activated to inform the decision. The process can certainly be influenced by situational factors, but it is safe to assume that certain attitudes are stored in the form of memory associations, because they serve basic human needs (functions), such as the need to sum- marize knowledge and express values. Contrary to the functional view on attitudes, constructive views suggest that people, including consumers, are evalu- ating objects by on the spot decisions, without engaging in memory-based processing, simply by using their perception and decision context as an information source (Bettman et al. 1998; Feldman and Lynch 1988; Reed et al. 2002). This would mean, in turn, that most attitudes are temporary and depend on the situation. The attractiveness of this view lies in the notion that humans tend to avoid cognitive effort and therefore do not engage in memory-based processing whenever an evaluative judgment takes place. This distinction between func- tional and constructive views has profound implica- tions for the study of the process of consumer attitude formation, as it suggests that attitudes towards marketing objects (e.g. products, brands, advertisements) do not necessarily stem from an evaluation of the product attributes, but may be entirely contextual, driven by factors such as consumers’ goals, presentation characteristics and transient affective states (Bettman et al. 1998; Clore 1994; Schwarz and Clore 1996). Thus an affective account of attitude formation may be more plausible than a cognitive, deliberative model.

Throughout the past 20 years, publications in the Journal of Consumer Psychology, Journal of Advertising, Journal of Business Research and Journal of Consumer Research have reflected the debate between functional and constructive views on attitude theory, with no apparent resolution. A resolution – if there can be one – would probably shift marketing and communication strategies from a long-term (changing memory) to a short-term (influencing the purchase situation) focus or vice versa. In light of the expansion of attitude theory into new marketing research contexts, we deem it necessary to pause, summarize and reflect on the key issues surrounding different conceptualiza- tions and views on the concept of attitude. In this vein, we start by explaining the relationship between attitude and evaluative categorization, which we believe is an operational issue and does not really enhance understanding of attitudes and the way they are formed. Secondly, we discuss

alternative theoretical perspectives on the process of attitude formation: (1) the functional vs con- structive view on attitudes; and (2) the cognitive vs affective view on attitudes. Lastly, we review current research in cognitive science and discuss how this may provide answers and how marketing managers can benefit from an integrative per- spective on attitude formation. Space limitations confine this review to topics directly relevant to consumer attitudes in marketing research as one of the most controversial topics of the last decade (Ajzen 2001; Cohen and Reed 2006; Schwarz 2006). Therefore, this paper contributes to the realm of marketing research by collecting the key theories of attitude formation, linking them to their philosophical underpinnings, and discussing their application in different research contexts. Effec- tively, we hope to provide a roadmap for marketing researchers in their efforts to understand what atti- tudes are, and how they can be formed.

Attitude vs evaluative categorization and judgment

A recent article by Cohen and Reed (2006) proposes a comprehensive model of attitude, namely the Multiple Pathway Anchoring and Adjustment of attitude generation and recruitment, which is an excellent account of attitude formation accommo- dating multiple theories. A pre-cursor of this model is Feldman and Lynch’s (1988) Accessibility– Diagnosticity model (refer to Lynch 2006). Still, these models have raised immediate criticism (Schwarz 2006) with respect to a fundamental problem: the term ‘consumer attitudes’ per se is constructed by researchers and therefore may be redundant. Schwarz (2006) suggests that, for the sake of parsimony, attitude should be treated no dif- ferently from evaluative categorization, a part of judgment, and therefore there is no need to explain the process of its formation separately. We believe that the essence of this debate is again the result of a basic two-part question: What does the concept of attitude represent, and does it have a distinct process of formation?

From a first glance at the literature, one can observe that, despite the diversity in views about the concept and formation of attitude, there is reasonable agreement that as a construct attitude represents ‘a summary evaluation of a psychological object captured in such attribute dimensions as

432 E. Argyriou and T.C. Melewar

good–bad, harmful–beneficial, pleasant–unpleasant, and likeable–dislikeable’ (Ajzen 2001, p. 29).

This definition equates the construct ‘attitude’ to an evaluative judgment, thus supporting Schwarz’s (2006) argument against the need for a separate hypo- thetical construct. In addition, it helps to distinguish attitude from simple categorization or other non- evaluative judgment. Empirical research using physio- logical measures supports this critical difference between categorization processes, indicating that there is a clear distinction between evaluative and non-evaluative categorization processes. For ex- ample, when people categorize food items as positive vs non-positive and as vegetable vs non-vegetable, evaluative categorizations evoke a different brain potential from non-evaluative categorizations, one which is larger and occupies a wide area on the right rather than the left region of the brain (Ajzen 2001; Crites and Cacioppo 1996). Hence, attitude is essen- tially an evaluative categorization distinct from a simple non-evaluative categorization, which uses distinct brain resources and activates distinct brain processes. The above-mentioned definition and con- ceptualization is dominant and universally accepted among prominent attitude theorists (e.g. Ajzen and Fishbein 2000; Eagly and Chaiken 1993; Petty and Wegener 1997).

Since attitude as a construct is equivalent to evalu- ative categorization, do we really need the term ‘atti- tude’? To answer this, it is necessary to explain alternative conceptualizations and processes of atti- tude formation in order to elucidate facets of attitude that distinguish it from other contextual evaluative categorizations. The review will hopefully shed light on the uniqueness of attitude as a concept beyond its operationalization as an evaluative categorization of an object on a continuum of positive–negative attributes.

The functional–constructive debate

Functional theory

Originally, functional theory defined the concept of attitude as a ‘predisposed tendency to respond towards an object in a favourable way’ (Eagly and Chaiken 1993). According to this view, consum- ers possess a state of readiness to respond to classes of objects, which is enduring and stored in the memory (Eagly and Chaiken 1993; Katz 1960; Shavitt 1990). Classical functional theory

assumes that people hold attitudes because they are functional, serving psychological motivations and needs (Shavitt 1989). These include the knowledge function, value-expressive function, social-adjustive function and utilitarian function (Ajzen 2001; Grewal et al. 2004; Schlosser 1998). This traditional functional view of attitudes is heavily memory based and refers to an intrinsic tendency rather than a simple evaluative categorization. Let us explain this in more detail.

According to functional theory, people form atti- tudes in order to organize, structure and summarize large amounts of information about an object (Grewal et al. 2004). Consequently, all attitudes serve pri- marily the function of knowledge (Fazio 1989). Knowledge can be conceptual or perceptual in nature, and supporters of the functional view believe that people organize knowledge in global attitude con- cepts, and in a manner along the lines of gestalt psychology, wherein the whole is greater than the sum of the parts (Schlosser 2003a). Additionally, attitudes express values, preferences and self-perceptions to others (Shavitt 1990). These attitudes serve a value-expressive function, and facilitate interpersonal communication.

Furthermore, attitudes help individuals conform to the expectations of others, so that social interaction runs smoothly and efficiently (Smith et al. 1956). These attitudes serve the social-adjustive function. The value-expressive and social-adjustive functions constitute the social-identity function, which is also known as ‘symbolic’ (Shavitt 1989, p. 125). Here- after, the term social-identity is used to refer to value-expressive and social-adjustive functions, unless a study specifically differentiates between the two. Finally, but importantly, attitudes summarize rewards and ‘punishments’ obtained from an object, and thus serve a utilitarian function (Katz 1960). An attitude serves this function to the extent that it sum- marizes the positive and negative outcomes that one associates with an object. Utilitarian functions guide approach and avoidance behaviour (Shavitt 1989). Prior to this taxonomy, Smith et al. (1956) discuss an ‘object-appraisal’ function, which includes attitudes that summarize an object’s evaluation against people’s concern about their well-being. This func- tion subsumes the knowledge and the utilitarian functions (Shavitt 1989). We revisit this function of attitudes subsequently in our discussion of the affective antecedents of evaluative judgment. For the time being, we refer to knowledge and utilitarian functions separately.

Consumer Attitudes Revisited 433

The notion that people hold attitudes because they are functional implies that, after their first exposure to stimuli, consumers form an attitude which they store in memory and, when prompted by marketers, advertisers or, more importantly, researchers, retrieve their attitudinal responses in memory. Func- tional theory adheres to the information-processing paradigm, according to which cognitive processing takes place in between a stimulus exposure and the formation of an attitude. With respect to type of processing, the majority of attitude models within the functional approach distinguish between delib- erative (rule-based) and spontaneous (associative) information processing (Fazio 1986, 1990; Fazio and Zanna 1981; Sloman 1996). Several dual-process models of information processing for attitude forma- tion exist, including Chaiken’s (1980) systematic– heuristic model and Cacioppo and Petty’s (1984) central–peripheral routes to persuasion, both stem- ming from the persuasive communications literature. The following section elaborates on the two modes of processing in order to elucidate the different routes via which an attitude can be formed and then evoked.

Deliberative or rule-based models assume that attitudes are formed from cognitive beliefs that are stored in explicit memory. Spontaneous or associative models, in contrast, assume that atti- tudes are evoked automatically through association, and relate mostly to affective factors such as emo- tions or feelings. Fishbein and Ajzen’s (1975) expectancy–value (EV) model, and the theories of reasoned action and planned behaviour that stem from it, represent the deliberative class of attitude models. We discuss these models further in sub- sequent sections, as they are more relevant to the distinction between cognitive vs affective debate regarding attitudes.

Modern functional theory does not preclude the operations of deliberative processing, but is more apt to modelling attitude retrieval as a spontaneous process, which is consistent with the functionalists’ assumption that attitudes lie in memory as associa- tive knowledge, and that these global associations are greater than the sum of their parts (Schlosser 2003a). Specifically, spontaneous processing models define attitude as ‘an association [stored] in memory between a given object and one’s evaluation of this object’ (Fazio 1990, p. 81). Fazio (1990) recognizes that attitude formation can be deliberative, depend- ing on factors such as motivation and opportunity, but evidence suggests that the mechanisms that elicit

attitudes from memory, such as priming,1 are largely automatic and occur below the level of conscious awareness (Fazio 2001; Olson and Fazio 2001). According to this model of attitude, whether a stored attitude will influence a subsequent evaluation, or other response, depends on its accessibility in memory or probability of activation. Among the factors that can influence the accessibility of atti- tudes is the salience or importance of its function, frequency of prior activation and the strength of the association between the concept and its attitude (Fazio 1990). For example, Fazio and Zanna (1981) suggest that attitudes formed through direct experi- ence are more readily accessible in memory and more strongly associated with behaviour than those formed indirectly, for instance via advertising, or word of mouth.

This associative model of attitude formation finds ardent support from researchers in classical condi- tioning approaches to advertising and marketing communications (e.g. Gorn 1982; Grossman and Till 1998; Kellaris and Cox 1989; Shimp et al. 1991). According to the associative perspective, in order to evoke a positive attitude a stimulus needs to empha- size the appropriate attitude function. Functional theory and the associative model also suggest that attitude objects, such as products and brands, empha- size different functions (Shavitt 1989). Therefore, in order to elicit a favourable attitude, marketers need to present them in functionally congruent ways. For example, Shavitt (1990) shows that the use of social- identity appeals in advertising results in less favour- able responses than utilitarian appeals for products that serve utilitarian functions, such as air condition- ers and coffee. Correspondingly, utilitarian appeals result in less favourable responses compared with social-identity appeals for products that serve social- identity functions, such as greeting cards. In addi- tion, within each product category, communicating brand messages through functionally congruent

1Priming refers to testing responses to objects after exposing subjects to some prior event (the prime). There are three types of priming tests: (1) semantic priming, e.g. testing whether recognition of the item ‘dog’ as a word or a non- word is faster when it is preceded by the associated word ‘cat’; (2) categorical priming, e.g. testing whether response to an object changes as a function of a construct, such as motivation or emotions, activated in a prior task; and (3) feature priming, e.g. making an object feature more salient so that it is weighted more heavily in later evaluations (Mandel and Johnson 2002).

434 E. Argyriou and T.C. Melewar

advertising results in more favourable responses than does communicating through incongruent advertis- ing (Shavitt 1990).

Functional theorists recognize the possibility that attitude objects offer more than one type of benefit, or serve multiple functions (Shavitt 1989). For instance, a sweatshirt with a university logo has func- tional benefits and serves a utilitarian function on a cold day at home, but has symbolic benefits and serves a social-identity function when worn at a uni- versity sports game (Schlosser 2003a). However, the notion that people ‘attach’ value to products and ser- vices implies that these benefits are relatively stable in the minds of consumers, even if their relative importance varies across different situations. In a recent study, Grewal et al. (2004) examine the inter- active effects of product attributes, attitude functions and context of the purchase decision (forced or unforced) on the timing of repeat purchases. Results are based on a study across seven product categories, i.e. cameras, cars, vacuum cleaners, watches, DVD players, refrigerators and televisions. Product cat- egory is not found to be a significant moderator of the results, which indicates that all products poten- tially serve all functions. However, certain product characteristics moderate the effect of attitude func- tions on the timing of purchases.2 In addition, Grewal et al. (2004) find that the decision context has a direct effect on the timing of a repeat purchase. Forced purchase decisions always result in longer inter-purchase intervals than unforced purchase deci- sions. In this sense, product attributes, salient atti- tude functions and situational parameters jointly determine consumer behaviour.

We need to stress that, although functional theory assumes that attitudes are stored in the memory, it allows for situational parameters to operate as cues for memory-based attitude retrieval (within the asso- ciative model). This notion is more evident in the study of retail atmospherics. For instance, Schlosser (1998, 2003a) explains the influence of aesthetic cues in a consumption environment on the salience of attitude functions in memory. She finds that shop- ping atmospheres interact with product attributes to influence consumer responses to stores and their

products. In particular, store atmospheres trigger the social-identity function; hence, aesthetically pleasant atmospheres have a positive effect on consumers’ evaluations of the store’s image and on their inten- tions to patronize the store, and a positive influence on perceptions of the quality of products with sym- bolic benefits (e.g. goblets for wedding toasts, perfume and jewellery), but not of products with functional benefits (e.g. vacuum cleaner, can opener, toothbrush).

In a different study of consumer attitudes towards products in Internet advertising contexts, Schlosser (2003a) finds that computers operate as a situational cue which evokes consumers’ need for knowledge. The researcher argues that consumers’ attitudes towards products on the Internet are influenced not only by the medium’s characteristics, but also by the computer as a context in which the Internet operates. Similarly, De Pelsmacker et al. (2002) suggest that the media context in which an advertisement is dis- played operates as a situational cue to determine attitude towards the advertisement. Along these lines, Shavitt (1989) has long highlighted the interactions between product attributes, attitude functions and situational factors. The recognition that situations can interfere with the elicitation of attitudes is fun- damental, as it contrasts the traditional functional belief that attitudes are stable predispositions (e.g. Eagly and Chaiken 1993).

Furthermore, Shavitt (1989) has highlighted that individual differences can also interact with product attributes and situational factors and influence attitu- dinal responses. Along these lines, Lutz (1981) finds that ego-defensive tendencies do not influence product attitudes, and Snyder and DeBono (1987) find that people who have a tendency for self- monitoring differ in their degree of social-identity needs. In addition, researchers have suggested that the salience of knowledge and utilitarian functions depends upon traits such as need for cognition, the most widespread measure of an individual’s personal motivation to engage in deliberative information pro- cessing (Cacioppo et al. 1984). In sum, functional theory treats attitudes as memory-based, object- related associations which are evoked in a spontan- eous manner, and acknowledges the possibility that the elicitation of these attitudes may vary across situ- ations and individuals. Hence, in order to explain and change attitudes, marketers need to understand their motivational underpinnings (Schlosser 2003a).

The recognition of the role of situational cues in the elicitation of attitudes facilitates the transition

2For example, products that score highly on the knowledge function have shorter inter-purchase intervals overall, although within this category luxuries have longer inter- purchase intervals than necessities. Similarly, products that score highly on the social-adjustive function have shorter inter-purchase intervals, but public products have longer intervals than private products.

Consumer Attitudes Revisited 435

from the functional to the constructive view. Similar to modern functional theorists, constructive attitude theorists acknowledge the contingencies between attitude objects and situational parameters (Bettman et al. 1998). However, they argue that human beings lack the working memory capacity, and hence the computational ability, to hold and retrieve an infinite number of well-defined attitudes in memory. Thus, people are likely to construct their attitudes based on the salience and diagnosticity of any information available in a given context (Bettman et al. 1998; Feldman and Lynch 1988; Schwarz and Bohner 2001). We elaborate on this approach in the follow- ing section.

Constructive theory

The constructivist view on attitude formation is part of a greater discussion in economics and psychology regarding whether well-defined preferences exist (McFadden 1999) or whether they are constructed at the time of valuation (Payne et al. 1992; Slovic 1995) Constructivists (or ‘constructionists’ as referred to in Reed et al. 2002) stem from the behavioural tradition in psychology, and are closely related to the intellec- tual movement of social cognition (Bandura 2001; Schwarz and Bohner 2001), which is rooted in the paradigms of cognitive (e.g. Bruner 1990) and social (e.g. Vygotski 1978) constructionism. A basic con- structivist assumption is that consumers do not retrieve their attitudes in memory, but compute on the spot according to their contextual goals (Bettman et al. 1998; Schwarz and Bohner 2001; Reed et al. 2002). Such goals involve reducing cognitive pro- cessing effort, increasing the accuracy of a decision, minimizing negative emotions or maximizing ease of justification (to oneself and others) for preferring an alternative (Bettman et al. 1998). Although these goals resemble attitude functions, there is a con- ceptual difference between functional and construc- tive perspectives. Functional theory assumes that functions correspond to enduring motivations, and situations merely invoke their salience in memory (Schlosser 2003a).

Constructive theory suggests that goals generate temporary motivations, which may determine the salience and diagnosticity of internal and external information in the process of attitude construction (Feldman and Lynch 1988; Reed et al. 2002). There- fore, constructive theory differs from functional theory, in that the former argues against automatic knowledge retrieval in memory, and endorses

situated learning and cognition (Brown et al. 1989; Schwarz 1998). In effect, the constructivist per- spective suggests that what researchers truly measure in attitude surveys are temporary evaluations (Schwarz and Strack 1991). Hence, attitudes are merely evaluative judgments, formed via a construc- tive process, in the context of an attitude question, or a decision problem (Bettman 1979; Bettman et al. 1998; Bettman and Park 1980; Reed et al. 2002; Payne et al. 1992; Slovic 1995; Tversky et al. 1988).

Despite it being associated with behavioural decision-making, constructive theory should not be confused with behaviourism, as it acknowledges the intermediate role of cognitive processing in attitude formation. Furthermore, constructivists coincide with Fazio and Zanna’s (1981) proposition that direct experience with an object may result in stronger atti- tudes, which will be better predictors of behaviour than attitudes formed without such experience. However, the main contribution of the constructive approach is an alternative explanation as to why this may be the case: direct experiences make strong use of the human perceptual system. As a result, percep- tually salient object attributes are remembered quickly and effortlessly at the time of evaluation (Reed et al. 2002). The focus on perception rather than cognition is an important difference between functional and constructive attitude theories. Although the constructive viewpoint acknowledges the role of cognition, it also highlights the human tendency to avoid engaging in cognitive effort, owing to limited working memory capacity. The behav- ioural literature proposes two general decision- making frameworks, namely the cost–benefit and the perceptual frameworks (Payne 1982; Tversky and Kahneman 1991). The vast majority of studies in constructive choice processes explain the variance in consumer preferences within a cost–benefit frame- work (Payne et al. 1992; see also Bettman et al. 1998 for a meta-analytic discussion). Even when the con- sumer goal is to minimize negative emotional expe- rience, or maximize ease of justification, they make trade-offs between effort and benefit (achieving the goal). Cognitive effort is a cost, and consumers are ‘cognitive-misers’ (Fiske and Taylor 1984, p. 12; in Garbarino and Edell 1997, p. 147). Therefore, the notion that consumers are willing to trade-off ben- efits in return for conservation of cognitive effort (cost–benefit framework, Payne 1982) is central to constructive preference processes.

Constructive information processing comprises a series of elementary information processes (EIPs),

436 E. Argyriou and T.C. Melewar

namely searching for, comparing, multiplying or adding and eliminating items of information. The sum of duration and number of EIPs equals the total cognitive effort of processing (Bettman et al. 1998). Owing to consumers’ tendency to avoid cognitive effort, they often rely on heuristic processing strat- egies in order to arrive at a decision. In attitude research, this mostly refers to affect referrals (Wright 1975) and/or anchoring and adjustment heuristics (Tversky and Kahneman 1974). The notion of heu- ristic strategies of attitude formation requiring more effort is consistent with Chaiken’s (1980), Cacioppo and Petty’s (1984) and Fazio’s (1990) models, which also discriminate between systematic and heuristic modes of processing based on the amount of cogni- tive effort (Chaiken 1980). Similar to modern func- tional theory, which is inclined towards associative processing vs deliberative processing, constructive theories support that consumers do not engage in systematic processing of information to form their attitudes; instead they are prone to heuristic process- ing, which is not exactly systematic, but not auto- matic either.

With respect to the context of construction, Bettman et al. (1998) and Payne et al. (1992) propose that preferences are contingent on the inter- actions between goals and aspects of the context. Such contingencies include (a) the complexity of a context, i.e. when task demands are high, options that are superior on prominent attributes tend to be favoured, (b) the characteristics of other options in the choice set, (c) the framing of the question that aims to elicit a preference, and (d) the representation or display (frame) of the choice set. Because these suggestions stem from research on decision-making rather than attitude, the contingencies identified above are more pertinent to choices rather than evaluative judgments.

In particular, the impact of context complexity on consumer choices suggests that, when cognitive demands are high, consumers focus on perceptually salient options in an attempt to minimize their cog- nitive effort. Applying this reasoning to the construc- tion of attitudes as evaluative judgments implies that conspicuous information (attributes) about an object will be more salient and hence more heavily weighted in evaluation. Context complexity may be merely the result of demanding goals, but also depends on the complexity of object or information presentation in the environment (this notion is equiva- lent to the ‘display of the choice set’, where informa- tion refers to different options). Coupey (1994)

suggests that consumers tend to avoid restructuring information into appropriate mental representations of an object unless the cost of losing accuracy by processing information directly in the form in which it is presented is high. By processing information that is conspicuous, and in its original form, consumers minimize their cognitive processing effort (Coupey 1994; Slovic 1972). Hence, researchers propose that marketers can influence consumer processing and therefore the construction of attitudes, by ‘matching’ information presentations to the cognitive demands of different shopping goals (e.g. Bettman and Kakkar 1977; Bettman and Zins 1977, 1979; Jarvenpaa 1989; Jarvenpaa and Dickson 1988; Russo 1977).

Constructivists attribute the positive effects of such congruencies between object presentations and task demands to the operations of the perceptual system, rather than those of working memory. This is consistent with the constructivist belief that people have limited working memory capacity; hence, they are not likely to engage in effortful processing (Bettman et al. 1998). Specifically, constructivists suggest that the positive effects of the above- mentioned contingencies are due to attention, selec- tivity and perceptual interpretation of information (Bettman et al. 1998; Lohse 1997). According to the psychology of attention (Kahneman 1973), people devote their voluntary attention to information in the context that is perceived to be relevant to their goals. Attention enhances information salience, while relevance enhances information diagnosticity in later stages of the attitude construction process (Feldman and Lynch 1988). In addition, aspects of the context that are complex, novel or extremely per- ceptually salient capture involuntary attention, thus facilitating external information search and accessi- bility (Bettman et al. 1998). Therefore, attention and selectivity can be influenced by both goal-directed and involuntary/perceptual factors.

To date, empirical studies that are clearly claiming to adopt a constructive viewpoint on attitude forma- tion are not easy to find. The reason for this may be (as mentioned previously) that the theory’s supporters do not use the term attitude, but refer to the underlying definition of attitude construct directly, i.e. evaluative judgment (Schwarz 1998; Schwarz and Bohner 2001; Yeung and Wyer 2004). A considerable body of research on the constructive nature of consumer responses is concerned with the informational role of affect in evaluative judgment and, hence, will be extensively discussed in the next section (e.g. Clore 1994; Garbarino and Edell 1997; Isen 2001; Schwarz

Consumer Attitudes Revisited 437

1998; Schwarz and Bohner 2001; Schwarz and Clore 1996; Yeung and Wyer 2004). For now, we illustrate two studies that explicitly refer to constructive pro- cesses of attitude formation, one demonstrating serial information acquisition, and a second one illustrating the anchoring and adjustment heuristic. Computa- tional processes are the central notion in information integration models (Anderson 1981, 1991) and the anchoring and adjustment judgment heuristic, which relates to Tversky and Kahneman’s (1974) research on judgment. Both processes describe contextual information processing in the course of evaluative judgment, and both suggest that consumers tend to anchor heavily on initial information and do not adjust their judgments significantly by evaluating informa- tion acquired at later stages.

Jacoby et al. (2002) study attitude construction from ‘online’ (on the spot) rather than memory-based information (Jacoby et al. 2002). Online processing begins with an initial piece of internal or external information, which is then revised and updated as each subsequent piece of information is acquired. Jacoby et al. (2002) suggest that online processing models can include prior attitudes – if any – as intern- al information, but the final attitudes are a function of incremental information input. This model of information integration recognizes every step as a dynamic EIP. These study findings conform to con- structive process theories: people tend to minimize their cognitive effort by reducing the number of EIPs. In addition, initial information is heavily weighted in attitude, unless an incremental input is of opposite valence, in which case the previously formed attitude is updated.

In addition, Reed et al. (2002) illustrate how con- sumers construct their attitudes towards advertise- ments based on information from two different sources: (a) direct experience with an advertisement, and (b) external information from others. In order to distinguish construction sources from attitude retrieval cues, constructivists refer to them as ‘schemas’ (Reed et al. 2002). Results from their study suggest that consumers with an evaluative goal anchor on initial information (either neutral from direct experience, or valenced external information from others), and then adjust for subsequent infor- mation. Similar to Jacoby et al. (2002), Reed et al. (2002) find that the initial piece of information is always more salient during the construction of evaluative judgment (primacy effect), regardless of its source. On the contrary, consumers with proofreading goals (non-evaluative) anchor on more

recent information and then adjust for information presented earlier (recency effect).

Overall, both studies demonstrate the potential fluctuation in attitudes that are temporarily con- structed. Although online processing models and the anchoring and adjustment heuristics may also result in attitudes being stored in the memory, constructiv- ists strongly believe that this dynamic process results in ‘any number of attitudes depending on the number of different schemas available for thinking about the object’ (Tesser 1978, pp. 297–298; in Reed et al. 2002, p. 375).

Summary of the functional vs constructive debate

On the one hand, functional theory conceptualizes attitudes as global, object-related associations retrieved in consumers’ memory according to a psy- chological motivation that becomes salient in a given situation. On the other hand, the constructivist per- spective suggests that consumers construct their attitudes when requested to, and in accordance with the cognitive demands induced by the context. Con- temporary functionalists concur with constructivists that attitudes are essentially situation- or context- dependent; therefore, both perspectives endorse the importance of interpreting consumer attitudinal responses in situ. Furthermore, both theories recog- nize that (a) people prefer to minimize their cognitive effort by relying on information that becomes more easily accessible and (b) direct experience with an object increases such accessibility. Finally, both adhere to the information-processing perspective, but in significantly different ways. While functional theory suggests that attitudes can be stored in memory as knowledge (information) and are retrieved in a deliberative or spontaneous manner, constructive theory suggests that attitudes (and knowledge) are generally constructed heuristically, based on any salient and diagnostic information accessible in a given context. This information may include affective concepts such as feelings or emotions at the time of judgment. Hence, while functional theory assigns an important role to memory-based processing, con- structive theory endorses heuristic and perceptual processing. Table 1 summarizes the similarities and differences between the two theories.

Critique

The debate between the functional and constructive viewpoints on attitude formation has important

438 E. Argyriou and T.C. Melewar

philosophical and practical dimensions. For example, although constructivists generally reject the idea that consumers store knowledge about an object in the form of global attitude in their memory, they acknowledge that consumers are likely to hold well- articulated preferences when they are familiar or experienced with an object (Bettman et al. 1998; Reed et al. 2002; Schwarz and Bohner 2001). An implication of this could be that consumer-novices are more susceptible to external influences from the context than consumer-experts, who are more likely to process internal information and do not turn to external search (Alba and Hutchinson 1987). However, in some cases (e.g. Mandel and Johnson 2002) visual priming appears to affect both novices and experts via different routes. Thus, visual primes affect novices’ evaluations and choice by pushing them to search more in external sources. In contrast, visual primes affect experts by making them look deeper internally. The constructivist view would attribute the effects of visual priming for novices to the operations of attention, selectivity and perceptual interpretation (Bettman et al. 1998). Prime features capture voluntary and involuntary attention, thus resulting in higher salience of the feature at the time of evaluation. But functionalists would argue that, for consumer-experts, visual primes merely trigger an association in memory, hence evoking an attitude that relies more heavily on the prime attribute (Mandel and Johnson 2002). Given the identical result of the two processes (i.e. change in attitude), it

is difficult if not impossible to accept that the two processes of attitude formation coexist harmoniously and are applicable to different people and situations.

Furthermore, although constructivists reject the notion that people rely on global object–attribute associations, they are ardent supporters of the limi- tations of working memory, and the operation of holistic, perceptual processing and judgment heuris- tics, thus not being entirely different from functional theorists. One of the oldest decision heuristics, namely Wright’s (1975) ‘affect referral’ heuristic, suggests that people simplify their decisions by relying on emotions, general attitudes and global impressions of objects instead of attribute informa- tion (Erevelles 1998). This resembles the associative model of attitude formation in functional theory. Fur- thermore, Priester et al. (2004) suggest that strong attitudes result in affect referrals, which in turn can moderate perceptual sensitivity to external stimuli. As a result, it is also difficult to detect whether atti- tude precedes perceptual interpretation and heuristic processing or vice versa.

Hence, we believe that the functional–constructive debate is largely philosophical and cannot be resolved by deciding on one correct process. Perhaps at a preconscious level, the mind’s ‘choice’ of creat- ing knowledge by retrieving global attitudes vs (re)constructing them from available salient informa- tion is itself contextual. A small number of research- ers support our speculation (e.g. Mantel and Kardes 1999; Nowlis and Simonson 1997; Sanbonmatsu and

Table 1. Functional vs constructive theory

Assumptions about Functional theory Constructive theory

Concept Classical: A predisposed tendency to respond towards an object in a favourable way.

Modern: An association [stored] in memory between a given object and one’s evaluation of this object.

Evaluative judgments constructed via a contextual process, at the time of a question, or a decision problem.

Attitude foundations Functions (motivations – needs). Knowledge. Utilitarian. Social-identity (Value-expressive, social-adjustive).

Goals. Minimizing effort. Maximizing accuracy. Minimizing negative experience. Maximizing ease of justification.

Information processing Deliberative or Spontaneous/Associative. EIPS: Computational strategies or heuristics (anchoring and adjustment).

Attitude retrieval Retrieval from memory. Accessibility based on salience or importance of

function, frequency of prior activation and the strength of the association between the object and its attitude. Can be influenced by situational cues.

Construction on the spot. Accessibility based on perceptual salience, diagnostic

internal and external information about an object in a given context.

Nature and operation Attitudes are stored in memory as object associations, but their manifestation may vary across situations and personalities.

Attitudes are formed in the process of attaining a goal, hence they are temporary and contextual constructions.

Consumer Attitudes Revisited 439

Fazio 1990) by showing that, when attribute infor- mation is inaccessible in memory or external sources, or decision-makers are low in their need for cognition, they construct their preferences based on global impressions and intuition.

Regardless of the problems associated with the empirical detection of functional vs constructive pro- cesses of attitude formation, there is a common, growing belief that cognition is situated, a notion that lies in the heart of functional and constructive theories (Brown et al. 1989; Fiske and Taylor 1991; Herrington and Oliver 1995, 1999; Schwarz 1998; Schwarz and Bohner 2001; Schwarz and Strack 1991). For social scientists, the constructive view- point is perhaps more appealing than the functional one, as it envisions humans as creative learners who build on existing, brand new and unused knowledge, rather than solely reflecting on prior knowledge structures (Brown et al. 1989). Still, the notion of situated cognition is growing among researchers in social sciences and will help researchers to resolve the debate between the functional and constructive views on attitude formation by adopting a unified view on attitudes being contextual owing to the process of their formation being situationally bound. This does not preclude some attitudes from being relatively stable and evoked in memory.

The notion of situated cognition also relates to the notion of dual cognition, i.e. rational and affective cognition. For several years, researchers treated attitudes as affective constructs that mediate the rela- tionship between cognitive information processing, i.e. reason-based beliefs, and subsequent behaviour (Fishbein and Ajzen 1975; Fishbein and Middlestadt 1995, 1997). Nonetheless, Zajonc (1968, 1980, 1984) argued that affect is distinct from attitude and can arise in the absence of conscious cognitive pro- cessing and influence the way people evaluate. This argument has shifted the focus from attitude as an affective construct per se to the cognitive and affec- tive components and/or determinants of attitude (Bagozzi et al. 1999; Erevelles 1998). This shift is consistent with the constructivist view on attitude as a process of knowledge and judgment construction, based on a variety of informational sources, includ- ing rational appraisals or thoughts, as well as feel- ings (e.g. Schwarz and Bohner 2001; Schwarz and Clore 1988, 1996). The next section illustrates the basic concepts underlying the cognitive vs affective debate about the origins of attitude and sheds more light onto a different area of the debate on attitude formation.

The cognitive–affective debate

Multi-attribute models: the primacy of cognition

The cognitive view on attitude formation is inter- twined with functional theory and deliberative models of attitude formation in that it sustains that attitudes are formed based on salient beliefs about an object’s attributes and the categorization of these beliefs on an affective continuum, i.e. like–dislike (Fishbein and Ajzen 1975; Fishbein and Middlestadt 1995, 1997). In this view, attitudes are conceptualized as affective responses stemming from systematic and deliberative cognitive processing of informa- tion about an object. As previously noted, a well- documented deliberative model of attitude is Fishbein and Ajzen’s (1975) EV model (Ajzen and Fishbein 1980; Cohen et al. 1972; Engel et al. 1990; Fishbein 1963). According to the EV model, a person’s attitude towards an object is a result of their salient (cognitive) beliefs about the object and the evaluative aspect of those beliefs. The EV model is a multi-attribute model of attitude formation and measurement.

According to the EV model, in order to arrive at a valid measure of attitude, one must: (i) carefully identify the salient attributes associated with the atti- tude object; (ii) assess the strength of beliefs that the attributes are associated with the object; (iii) assess the evaluation of the attributes; and (iv) create a sum of the products formed by multiplying the strength of each belief by its evaluative aspect (Fishbein and Middlestadt 1995). Furthermore, the EV model assumes that evaluation arises inevitably, as we form cognitive beliefs about an object (Ajzen and Fishbein 2000). Although people may form various beliefs about an object, the EV model postulates that only those beliefs that are readily accessible in memory will influence our evaluation (Ajzen 2001). Delibera- tive processing models such as the EV emphasize the process of accessing information, which has been acquired on a previous occasion, in memory, even if the occasion had occurred a few minutes earlier (Jacoby et al. 2002).

Multi-attribute cognitive models such as the EV were popular in consumer behaviour in the 1960s and 1970s, but in the early 1980s, a number of research- ers rejected the notion of purely cognitive bases of attitude formation, and argued for a non-belief-based process of attitude formation and change. Around that time, empirical evidence from studies in persua- sive communication were demonstrating that rea- soned arguments based on object attributes are

440 E. Argyriou and T.C. Melewar

effective for well-educated or analytical people (Cacioppo et al. 1983) and highly involved audi- ences (Petty and Cacioppo 1981). Hence, researchers began to propose alternative theories about the direct determinants of attitude formation and change. The argument for affect operating independently of cog- nition and being a distinct predictor of consumer attitudes is rooted in Zajonc’s (1968, 1980, 1984) research.

Primacy of affect theory

Zajonc (1968) postulated that affective responses to a stimulus object could influence subsequent evalua- tion without people being consciously aware of the stimulus’s properties (hence without people forming cognitive beliefs). Effectively, attitudes are based on elements such as liking, feelings and emotions gen- erated by mere exposure to the stimulus, rather than attribute information (Vakratsas and Ambler 1999; Zajonc 1980). Zajonc’s (1998) own interpretation of his research is that repeated exposures to stimuli may improve processing fluency, which in turn can influ- ence evaluation and judgment of stimuli in two ways. Fluency itself can be affectively positive, and this effect may be misattributed to characteristics of the object, or else fluency creates familiarity, which results in positive affect hence positive valence (Lee and Labroo 2004; Lohse 1997; Winkielman et al. 2003; Zajonc 1984). This explains why attitudes towards familiar objects will tend to be more favour- able than attitudes towards novel ones (for more interpretations of mere exposure theories, refer to Zajonc 1998). Although Zajonc (1968, 1984) has been criticized for relying on measures of stimulus recognition in order to eliminate the operations of cognition (Mandler et al. 1987; Pham et al. 2001), his ‘primacy of affect’ theory has inspired a consid- erable body of research (e.g. Batra and Ray 1985, 1986; Burke and Edell 1989; Derbaix 1995; Edell and Burke 1987; Holbrook and Batra 1987; Janisze- wski and Warlop 1993; Mitchell and Olson 1981; Shimp 1981; Srull 1983; Stuart et al. 1987).

Dual process models

In addition to primacy of cognition vs primacy of affect theories, attitude research on persuasive com- munications gave rise to dual-process models, such as Chaiken’s (1980) heuristic–systematic processing, and Cacioppo and Petty (1984) central–peripheral routes to persuasion. These models essentially con-

ceptualize attitudes as the result of different levels of effort (involvement) devoted to cognitive processing. High effort results in systematic, reason-based evalu- ations, while low effort results in heuristic and peripheral evaluations. Batra and Ray (1985), still within the cognitive–affective hierarchy, propose that consumers may develop hedonic attitudes based on pure liking, and utilitarian attitudes based on reason- based beliefs about hard product attributes (Hir- schman and Holbrook 1982; Holbrook and Batra 1987; Vakratsas and Ambler 1999). Hence, attitudes may be a product of two belief structures. The two structures reflect different qualities and depths of cognitive processing, rather than different amounts of cognitive elaboration.3

Modern primacy of affect theories

Since these early theories on attitude formation based on cognition, affect or both, researchers have refined the definitions of affect and, as a result, their views on the nature of attitude. Specifically, Bagozzi et al. (1999) define affect as a mental state of readiness that can arise from automatic or conscious appraisals of events or thoughts. Affect is an umbrella term for emotions and moods, and possibly attitudes (Bagozzi et al. 1999; Schwarz and Clore 2003). The difference between emotions and mood(s) is that mood is not attributed to a stimulus, i.e. we are ‘in’ a good mood, while emotions are attributed to a stimulus; for example we are angry ‘with’ or ‘at’ or ‘about’ some- thing (Clore et al. 1994; Schwarz and Clore 2003). The confusion between affect and attitude stems from the fact that, much like emotions, attitudes have objects. A fundamental difference between attitude and emotions is that people form attitudes towards mundane objects, while arousal is an important part of emotions (Bagozzi et al. 1999).

3Since Batra and Ray’s (1985) proposition, researchers have developed attitude measures that include two distinct but correlated dimensions, one utilitarian and one hedonic (Batra and Ahtola 1991; Spangenberg et al. 1997; Voss et al. 2003). The first dimension results from functional benefits of objects, while the second dimension results from sensa- tions derived from direct experience with the object. This distinction between utilitarian and hedonic components of attitudes is operational and not meant to be confused with the ‘utilitarian’ function of attitudes, although evidently, utilitarian and hedonic attitudes summarize a greater variety of positive and negative outcomes that one associates with an object (Shavitt 1989), i.e. functional and experiential benefits (Keller 1993). Hence, the two dimensions capture different aspects of the utilitarian function.

Consumer Attitudes Revisited 441

Pham et al. (2001) suggest a comprehensive typ- ology of affective responses that helps distinguish affect from attitudes. Type I affect is a bio-regulatory response, triggered by the sensory-motor mechanism (e.g. response to spoiled food). Type II affect is also triggered automatically by mapping stimulus attributes onto schematic structures that have been associated, possibly through conditioning, with par- ticular emotional responses (e.g. response to fear scenes in films). Type III affect is a controlled assess- ment of a stimulus object which involves cognitive appraisal of its significance for well-being (Lazarus 1991). Hence, affect is not identical to, but can enter evaluative judgment through, spontaneous associa- tion or deliberative processing.

Supporters of the purely cognitive basis of attitude (e.g. Fishbein and Middlestadt 1995, 1997) propose that the deliberative influences of peripheral cues, heuristics, emotional reactions, affective responses, classically conditioned reactions and attitudes are mediated by a change in cognitive structure. For instance, feelings of frustration may reinforce beliefs that a product is not reliable. This type of mechanism is consistent with the mood literature. Incidental moods change the accessibility of mood-congruent beliefs in memory (Schwarz 1997). Nonetheless, Schwarz (1997, 1998) suggests a different mecha- nism: affect can operate as a unique source of infor- mation, which enters evaluation in a deliberative manner, but not necessarily through changes in cog- nitive beliefs. Such an affect-based inference is the ‘how do I feel about it’ heuristic (Schwarz and Clore 1988), where people infer the direction and strength of their attitude towards an object from the valence and intensity of their feelings (Gorn et al. 2001; Pham 1998). Although a heuristic strategy, the ‘how do I feel about it’ heuristic is inferential, not purely associonistic (Pham 2004). Hence, heuristic infor- mation processing (e.g. Cacioppo and Petty 1984; Chaiken 1980) is not equivalent to spontaneous, associative processing (Fazio 1990). This heuristic is consistent with constructive theories of attitude for- mation, as it allows a role for affect as diagnostic information that can be used in processing and evalu- ation. Pham et al. (2001) suggest that affect-based inferences are considerably faster than non-affective, ‘cold’ and reason-based inferences, that they are more consistent among individuals, and are more predictive of the number of thoughts towards a stimulus. However, they involve considerable cogni- tive mediation and are slower than automatic affec- tive responses (Cohen and Areni 1991). An important

issue in this heuristic, affect-based inference is that, while in some cases affect is evoked by the stimulus being judged (Schwarz 1999; Shiv and Fedorikhin 1999), individuals cannot always distinguish bet- ween the sources of affect they happen to experience. Effectively, the ‘how do I feel about it’ heuristic may also involve misattribution of the mood that people are in (Schwarz 1999; Yeung and Wyer 2004) to the object that they are evaluating.

Hence, following Zajonc’s view on the primacy of affect in attitude formation, researchers have pro- gressed in their findings about affective processes. This has significantly changed marketing research- ers’ views on information processing, thus causing a need for change in attitude research. Findings from research in the primacy of affect in judgment are pointing towards a role for affect as information that enters attitude formation in a deliberative (e.g. Fishbein and Middlestadt 1995, 1997), spontaneous or more importantly, in a heuristic inferential manner (Lazarus 1991; Schwarz 1997, 1998). We illustrate the implications of this for attitude theory and relate it to the functional vs constructive debate in the summary below.

Overlaps with functional/constructive views

Theories of attitude formation based on the primacy of cognition are closely related to functional theory and deliberative models of attitude formation. Primacy of affect theories, such as Zajonc’s (1968, 1984), discard cognitive processing and propose that attitudes may be the result of fluency in stimulus recognition, stemming from familiarity, which in turn causes positive feelings and thus positive atti- tudes. This notion is closer to constructivists’ view on perceptual processing being a dominant strategy in attitude formation. Recent findings from research in the primacy of affect recognize different types of affect. therefore allowing feelings, emotions or moods to enter attitude formation in a spontaneous (affect I and II) or heuristic manner (affect III). Still, supporters of the cognitive basis of attitude insist that affect enters attitude formation in a deliberative manner by reinforcing people’s negative or positive beliefs (e.g. Fishbein and Middlestadt 1995, 1997).

A functional interpretation of affect type I and II would imply that affective information is stored in memory as part of object-related associations, i.e. attitudes, and enters evaluation and judgment spon- taneously (e.g. Fazio 1986, 1990, 2001). Earlier in our discussion of the functional theory of attitudes

442 E. Argyriou and T.C. Melewar

we referred to an ‘object-appraisal’ function (Smith et al. 1956), which subsumes attitudes that serve knowledge and the utilitarian functions and essen- tially summarizes evaluations of an object’s signifi- cance to well-being (Shavitt 1989). There is clearly a connection between object-appraisal functions and contemporary primacy of affect theories. In line with this view, Chaiken et al. (1999) suggest that the func- tional perspective on attitudes as object-related asso- ciations does not refer to mere object-impressions, but to a complex structure of cognitive, affective, and behavioural associations, which are summarized in object–association linkages. Similarly, a functional interpretation of affect type III would imply that affective responses arise from deliberative appraisals about an object’s significance for well-being. This would be consistent with deliberative multi-attribute attitude models (e.g. EV, Fishbein and Ajzen 1975), in which judgment is the product of salient beliefs and the evaluative aspect of those beliefs.

However, the affect-as-information view, accord- ing to which affect enters evaluative judgment through a heuristic inferential process, does not imply changes in memory-based belief structure (Pham et al. 2001; Schwarz 1998). As mentioned above, the notion of the ‘how do I feel about it’ heuristic process makes this perspective pertinent to constructive attitude processes. Constructive attitude models recognize that people may form evaluative judgments from any information available in a context, including their feelings about an object (Schwarz and Bohner 2001). Garbarino and Edell (1997) also suggest that the process of information processing generates affect. As cognitive effort increases, consumers experience negative affect which lowers the likelihood of an object being pre- ferred. Along the same lines, Clore (1994) suggests that people interpret stimuli that facilitate goal achievement as pleasant. Hence, stimulus-based or process-generated affect influences the construction of evaluative judgment on the spot.

It appears from this discussion that the affect-as- information perspective fits both the functional and constructive theories by promoting a typology of affect that (a) provides a comprehensive argument for the informational value of affect in the formation of attitude, and (b) distinguishes between spontaneous and deliberative processing but also adds a third type of processing, namely heuristic information pro- cessing, thus accommodating more routes to attitude formation, including memory-based associations, evaluative categorizations from rule-based process-

ing and evaluative categorizations on the spot, in the context of a decision. The acknowledgement of the informational value of affect allows for three types of attitudes: object associations, rule-based evaluations and affective temporary evaluations.

Critique

An influential counter-argument against modern primacy of affect theories comes from modern Pav- lovian theory. What this theory contends is that con- ditioning involves contingency awareness, defined as the individual’s learning that the conditioned stimu- lus precedes the unconditioned stimulus in classical conditioning experiments (Allen and Janiszewski 1989; Kim et al. 1996; Priluck and Till 2004; Shimp 1981). Thus, conditioning is a form of cognitive associative learning in which the individual pro- cesses information to form connections between events that occur in the environment (Shimp et al. 1991). Such learning does not occur without cogni- tive mediation. Several studies demonstrate that increased awareness leads to an understanding of the temporal relationship between the two stimuli and, as a result, the individual can clearly articulate the con- nection (Priluck and Till 2004; Shimp 1981). This proposition has strong implications for the study of affect as information input into evaluative judg- ments. If classical conditioning, an inherently asso- ciative process, can occur at a conscious level, perhaps affect type II and even type I (referred to above) also involve cognitive mediation.

This reasoning is not meant to dismiss the inde- pendent and informational value of affect in attitude formation, but to illustrate the difficulty of pinpoint- ing, conceptually and empirically, the primacy of affect vs the primacy of cognition in evaluative judg- ment. Judgment resulting from object-appraisals is more likely to refer to spontaneous retrieval of asso- ciations, or affective beliefs, rather than deliberative analysis of individual attributes (Lazarus 1991; Yeung and Wyer 2004). These findings are consistent with judgment referral heuristic choice strategies (Baker 2001; Wright 1975), as well as the function- alists’ view that attitudes can be retrieved in memory, where they are stored as global object–evaluation associations (Chaiken et al. 1999; Fazio 1990). Nonetheless, regardless of the lack of correspon- dence between attitudinal responses and deliberative processing of attributes of an object, appraisals are inherently cognitive. Hence the majority of attitudi- nal and affective responses involve a minimum level

Consumer Attitudes Revisited 443

of cognitive mediation (Clore 1994; Yeung and Wyer 2004), even if they do not necessarily involve attribute-based inference (Baker 2001; Mantel and Kardes 1999). This raises an issue with regard to whether deliberative and conscious processing should be equated with cognitive processing. The answer comes from cognitive science and is dis- cussed below.

General discussion and conclusions

Notwithstanding the above extensive literature on attitude and its formation, the question concerning the need for an attitude construct, given its equiva- lence to evaluative judgment, remains puzzling. From our standpoint and from our review of the debates and conceptualizations regarding attitude, we conclude that attitudes are indeed evaluative judgments measured via categorization on a con- tinuum involving several attribute dimensions; however, their formation may stem from qualitatively different processes. The term attitude, which implies a more enduring state of mind, is helpful in distin- guishing contextual evaluations from stronger, previ- ously formed object-related associations stored in the memory. In the previous sections, we discussed how object-related associations may involve affective and rational components, and how affect differs from attitude in that it may also occur automatically in the absence of cognitive processing, while attitudes involve a minimum level of cognitive involvement in the form of appraisals. Appraisals, in turn, can be highly automatic or heuristic (e.g. how do I feel about it?) or rational. Therefore, the different per- spectives on attitude formation are not necessarily contrasting views, but supplementary, giving way to more integrated theories of attitude.

Researchers continue, and we hope they will keep on continuing to theorize upon the process of attitude formation, sometimes setting its boundaries (e.g. Park and MacInnis 2006) or questioning its relevance (e.g. Schwarz 2006). Recent advancements in cogni- tive science and neuroscience will add to the debate, by shedding some light on the workings of attitude formation as brain processes. In the light of such research, we expect that the debates between func- tional and constructive theories, as well as primacy of cognition vs primacy of affect models, will soon find their resolution. We discuss a few examples to illustrate how cognitive science may prove useful in resolving the debates.

Neurophysiological research has long distin- guished between two systems of cognition, namely syncretic and analytic, associated with emotion and reason, respectively, in the left and right cerebral hemispheres (Buck et al. 2004). In this view, affect becomes another type of cognition, another type of knowledge. LeDoux (1998) explains that cortico- cognitive processes in the hippocampus and neocor- tex are different from affective processing involving the amygdala. Affective processing may occur even in the absence of cortico-cognitive processes, contrary to what most hierarchical frameworks (Vakratsas and Ambler 1999) have hitherto assumed. In turn, pure affective processing may result in implicit knowledge. This, however, does not imply that cognition is absent. In fact, cognitive science identifies two central memory networks, one ex- plicit or declarative (hippocampus) and one implicit (amygdala), which operate in parallel. Explicit memory may be episodic (factual knowledge about a specific time and place) or semantic (theoretical knowledge independent of time and place). Implicit memory is pure emotional memory. In this view, human beings may experience affect immediately and directly, without necessarily being consciously aware of it, or forming explicit beliefs about it. This view supports an integration of cognitive and affec- tive views regarding attitude formation, and directs marketing research which distinguishes between explicit, semantic or implicit memory-based attitude formation (e.g. Greenwald and Banaji 1995).

This view of dual cognition that recognizes a role for emotion as knowledge in evaluation is further supported by the cognitive-experiential self-theory (CEST) of personality (Epstein 1994). CEST posits that there are two cognitive sub-systems, namely the experiential and the rational, which organize human knowledge and interact to shape judgment. The dif- ference between them is that the first system has a longer evolutionary history and is shared with other animals, while the rational system is unique to humans (Epstein 1994). In addition, the experiential system stores knowledge in implicit beliefs or sche- mata, whereas the rational system stores them in explicit beliefs (see also Epstein, Pacini, Denes-Raj and Heier 1996 for details of the systems’ charac- teristics). Both systems are memory-based as differ- ent sub-systems of cognition. Hence, another possible route in attitude research is investigating attitudes as object-related associations, i.e. schemata vs attitudes that are explicit beliefs. This will help marketing researchers to integrate prior debates on

444 E. Argyriou and T.C. Melewar

purely cognitive vs purely affective models of atti- tude formation.

One may argue that models such as the elaboration likelihood model (Cacioppo and Petty 1984; Petty and Cacioppo 1981) or MacInnis and Jaworski’s (1989) integrative model have long incorporated both cognitions and emotions and thus constitute adequate frameworks of attitude formation. In response to this argument, we note an important con- ceptual difference: traditional dual-process models of attitude formation conceptualize cognition and emotion as distinct processes, and amount of involvement as the moderator of processing route (cognitive-central vs peripheral-affective). However, dual-cognition models see involvement as the com- bination of the two cognitive systems. Quoting Chaudhuri (1993), ‘. . . involvement is composed of feeling and thinking’ (p. 154). Although a few researchers discuss that there may be different quali- ties of cognitive elaboration (e.g. Petty and Wegener (1998) note that low elaboration may indicate either less effortful examination of the same information or effortful examination of less information4), they still place feelings and affective states at the low end of the elaboration continuum. In contrast, Batra and Ray’s (1985) conceptualization of involvement as the depth and quality of cognitive processing reflects more accurately the duality of cognition; in this view, cognitive processing may be both affective/ emotional and rational.

In accordance with recent findings in cognitive science, and conclusions from all perspectives on the nature and origins of attitude formation, we propose the following generalizations:

(1) In order to form attitudes, consumers adopt different processing strategies, i.e. systematic– deliberative, associative–spontaneous or heuris- tic, which reflect parallel but different qualities (types) of cognitive processing.

(2) The adoption of a processing strategy depends on the goal (e.g. minimization of cognitive effort, maximization of benefits) and the context of evaluation (e.g. accessibility of information), and is moderated by the human perceptual system. This is consistent with both functional and constructive viewpoints on attitude forma- tion. The perceptual system facilitates process- ing fluency and augments the effects of attitude

priming (functional theory). In addition, the perceptual system facilitates attention and sele- ctivity during information acquisition, hence reducing effort from the cognitive system (con- structive theory).

Implication for marketing researchers: The context of the decision problem, situational parameters and individual differences will determine the extent to which consumers will rely on deliberative, spontane- ous or heuristic information processing. Encourage- ment of heuristic, perceptual processing may help alleviate consumers’ cognitive load and result in desirable outcomes in terms of consumers’ preferences.

(3) Information input can vary from attribute-based beliefs to affect-based beliefs, past general atti- tudes, and global impressions. Attribute-based inferences enter judgment deliberatively, in the form of beliefs about the individual attributes of an object. Affect-based inferences also enter judgment deliberatively, in the form of experien- tial emotional information, which can be integral or incidental to the object. Finally, affect (type II), general attitudes and global impressions can also enter evaluation spontaneously.

Implications for marketing researchers: Attitudes are evaluative judgments of objects that may be retrieved, or constructed, or a combination of both, according to the consumers’ goals and the informa- tion available in the context in which an evaluation occurs. Either way, information can enter evaluation deliberatively, in the form of pure cognitive beliefs or affect-based heuristics, or spontaneously, in the form of affective responses or general attitudes and global impressions. Table 2 is a classification of research studies according to assumptions about the way information enters evaluation and the cognitive pro- cesses behind it. For example, the top left cell indi- cates that the view that information enters evaluation deliberatively in the form of declarative knowledge assumes a strictly cognitive process where attitude is formed through attribute-based processing. In con- trast, the bottom left cell indicates that information may enter evaluation in the form of associations, but still there is a lever of awareness of the associated stimuli such that cognitive processing can be inferred. The second column from the left highlights that most studies in the affective realm of research assume a heuristic processing of affect as informa-

4We thank an anonymous reviewer for this valuable comment.

Consumer Attitudes Revisited 445

tion, while the next column shows the integrative nature of studies within the dual-process framework, where information is assumed to enter evaluation in different fashions. Finally, the last column on the right displays the special case of studies in the behav- iourist tradition, where evaluation is simply a reac- tion to stimuli that can be learned and reinforced by the environment. Thus, the table might assist market- ing researchers in understanding the theoretical underpinnings of evaluative judgment (and therefore attitude) formation in a variety of research contexts. Yet, and as Sloman (1996) correctly points out in his discussion of associative vs rule-based human rea- soning, the debate about the cognitive systems and information processes that result in people’s evalua- tions and attitudes is largely philosophical as:

When a person is given a problem, however, both systems may try to solve it: each may compute a response, and those responses may not agree. Below, I argue that cases can be found in every domain of reasoning that has been studied in detail in which they do not [agree]. Because the systems cannot be distinguished by the problem domains to which they apply, deciding which system is responsible for a given response is not always easy. It may not even be possible because both systems may contribute to a particular response. (Sloman 1996, p. 6)

In conclusion, assuming that a single process underlies attitude formation is impossible and, in fact, would deprive attitude of its richness as a concept. Marketing researchers need to recognize the

existence of alternative processes underlying the process of attitude formation under different circum- stances. Whether functional or constructive, attitudes continue to reflect people’s evaluative stance apropos object, which is guided by both thought and affective processes. Dual-cognition and modern functional and constructive theories will dominate the realm of attitude research as technological advances will con- tinue to shed light on unexplored areas of the human cognition. Studies from now on should specify care- fully the conditions under which attitude measure- ment occurs and the variables involved in order for the correct process of attitude formation to be inferred. Although the current article hopes to serve as a building block for such an effort, we encourage future researchers to investigate the concept of atti- tude further and relate it back to its theoretical under- pinnings. Research of this sort will undoubtedly assist managers in their efforts to pinpoint the origins and nature of their consumers’ attitudes in different marketing contexts.

References Ajzen, I. (2001). Nature and operation of attitudes. Annual

Review of Psychology, 52, pp. 27–58. Ajzen, I. and Fishbein, M. (1980). Understanding Attitudes

and Predicting Social Behaviour. New York: Prentice Hall.

Ajzen, I. and Fishbein, M. (2000). Attitudes and the attitude–behaviour relations: reasoned and automatic

Table 2. Classification of studies on evaluative judgment (attitude) formation

Models processing

Cognitive Affective Dual-process Behaviourist

Deliberative (strictly declarative knowledge)

Attribute-based processing (e.g. Ajzen and Fishbein 1980; Fishbein and Ajzen 1975; Fishbein and Middlestadt 1995, 1997)

Persuasive hierarchies systematic vs heuristic (e.g. Chaiken 1980), Central vs peripheral routes (e.g. Cacioppo and Petty 1984), Utilitarian vs hedonic (e.g. Batra and Ray 1985)

Deliberative (rule-based) and spontaneous (associative) information processing (Fazio 1986, 1990; Fazio and Zanna 1981)

Heuristic (involves inferential judgment)

Affect-as-information (e.g. Lazarus 1991; Pham 1998; Pham et al. 2001; Schwarz and Clore 1996)

Affect transfer (e.g. Gorn 1982; Zajonc 1968, 1980, 1984)

Spontaneous (associative knowledge)

Modern Pavlovian Theory (contingency awareness), e.g. Allen and Janiszewski 1989; Kim et al. 1996; Priluck and Till 2004; Shimp 1981

Operant conditioning (e.g. Foxall 1997; Foxall and Greenley 1998; Guerin 1994; Foxall and Yani-de-Soriano 2005)

446 E. Argyriou and T.C. Melewar

processes. In Stroebe, W. and Hewstone, M. (eds), Euro- pean Review of Social Psychology, Vol. 11. Chichester: Wiley, pp. 1–33.

Alba, J.W. and Hutchinson, J.W. (1987). Dimensions of con- sumer expertise. Journal of Consumer Research, 13, pp. 411–454.

Allen, C.T. and Janiszewski, C.A. (1989). Assessing the role of contingency awareness in attitudinal conditioning with implications for advertising research. Journal of Market- ing Research (JMR), 26, pp. 30–43.

Anderson, N.H. (1981). Foundations of Information Integra- tion Theory. San Diego, CA: Academic Press.

Anderson, N.H. (1991). Contributions to Information Integration Theory, Vols 1–3. Hillsdale, NJ: Lawrence Erlbaum Associates.

Bagozzi, R.P., Gopinath, M. and Nyer, P.U. (1999). The role of emotions in marketing. Journal of the Academy of Marketing Science, 27, pp. 184–206.

Baker, W.E. (2001). The diagnosticity of advertising gener- ated brand attitudes in brand choice contexts. Journal of Consumer Psychology, 11, pp. 129–139.

Bandura, A. (2001). Social cognitive theory: an agentic per- spective. Annual Review of Psychology, 52, pp. 1–26.

Batra, R. and Ahtola, O.T. (1991). Measuring the hedonic and utilitarian sources of consumer attitudes. Marketing Letters, 2, pp. 159–170.

Batra, R. and Ray, M.L. (1985). How advertising works at contact. In Alwitt, L.F. and Mitchell, A.A. (eds), Psy- chological Processes and Advertising Effects: Theory, Research and Application. Hillsdale, NJ: Erlbaum, pp. 13–43.

Batra, R. and Ray, M.L. (1986). Affective responses medi- ating acceptance of advertising. Journal of Consumer Research, 13, pp. 234–249.

Bettman, J.R. (1979). Issues in research on consumer choice. Advances in Consumer Research, 6, pp. 214–217.

Bettman, J.R. and Kakkar, P. (1977). Effects of information presentation format on consumer information acquisition strategies. Journal of Consumer Research, 3, pp. 233– 240.

Bettman, J.R. and Park, C.W. (1980). Effects of prior know- ledge and experience and phase of the choice process on consumer decision processes: a protocol analysis. Journal of Consumer Research, 7, pp. 234–248.

Bettman, J.R. and Zins, M.A. (1977). Constructive processes in consumer choice. Journal of Consumer Research, 4, pp. 75–85.

Bettman, J.R. and Zins, M.A. (1979). Information format and choice task effects in decision making. Journal of Consumer Research, 6, pp. 141–153.

Bettman, J.R., Luce, M.F. and Payne, J.W. (1998). Construc- tive consumer choice processes. Journal of Consumer Research, 25, pp. 187–217.

Brown, J.S., Collins, A. and Duguid, P. (1989). Situated cognition and the culture of learning. Educational Researcher, 18, pp. 32–42.

Bruner, J. (1990). Acts of Meaning. Cambridge, MA: Harvard University Press.

Bruner, G.C., II and Kumar, A. (2000). Web commercials and advertising hierarchy of effects. Journal of Advertis- ing Research, 40, pp. 35–43.

Buck, R., Anderson, E., Chaudhuri, A. and Ray, I. (2004). Emotion and reason in persuasion: applying the ARI model and the CASC scale. Journal of Business Research, 57, pp. 647–656.

Burke, M.C. and Edell, J.A. (1989). The impact of feelings on ad-based affect and cognition. Journal of Marketing Research (JMR), 26, pp. 69–83.

Cacioppo, J.T. and Petty, R.E. (1984). The elaboration like- lihood model of persuasion. Advances in Consumer Research, 11, pp. 673–675.

Cacioppo, J.T., Petty, R.E. and Kao, C.F. (1984). The effi- cient assessment of need for cognition. Journal of Person- ality Assessment, 48, pp. 306–307.

Cacioppo, J.T., Petty, R.E. and Morris, K.J. (1983). Effects of need for cognition on message evaluation, recall and persuasion. Journal of Personality and Social Psychology, 45, pp. 805–818.

Chaiken, S. (1980). Heuristic versus systematic information processing and the use of source versus message cues in persuasion. Journal of Personality and Social Psychology, 39, pp. 752–766.

Chaiken, S., Duckworth, K.L. and Darke, P. (1999). When parsimony fails. Psychological Inquiry, 10, pp. 118–123.

Chaudhuri, A. (1993). Advertising implications of the plea- sure principle in the classification of products. In Van Raay, W.F. and Bamoosy, G.J. (eds), European Advances in Consumer Research, Vol. 1. Provo, UT: Association for Consumer Research, pp. 154–159.

Chen, Q. and Wells, W.D. (1999). Attitude toward the site. Journal of Advertising Research, 39, pp. 27–37.

Clore, G.L. (1994). Why emotions are felt. In Ekman, P. and Davidson, R.J. (eds), Nature of Emotion: Fundamental Questions. New York: Oxford University Press, pp. 103– 111.

Clore, G.L., Schwarz, N. and Conway, M. (1994). Affective causes and consequences of social information process- ing. In Wyer, R.S. and Srull, T.K. (eds), Handbook of Social Cognition, Vol. 1. New Jersey: Hillsdale, pp. 323– 417.

Cohen, J.B. and Areni, C.S. (1991). Affect and consumer behaviour. In Robertson, T.S. and Kassarjian, H.H. (eds), Handbook of Consumer Behavior. Englewood Cliffs, NJ: Prentice Hall, pp. 188–240.

Cohen, J.B. and Reed, A., II (2006). A multiple pathway anchoring and adjustment (MPAA) model of attitude gen- eration and recruitment. Journal of Consumer Research, 33, pp. 1–15.

Cohen, J.B., Fishbein, M. and Ahtola, O.T. (1972). The nature and uses of expectancy–value models in consumer attitude research. Journal of Marketing Research (JMR), 9, pp. 456–460.

Consumer Attitudes Revisited 447

Coupey, E. (1994). Restructuring: constructive processing of information displays in consumer choice. Journal of Consumer Research, 21, pp. 83–89.

Crites, S.L.J. and Cacioppo, J.T. (1996). Electrocor- tical differentiation of evaluative and nonevaluative categorizations. Psychological Science, 7, pp. 318– 321.

De Pelsmacker, P., Geuens, M. and Anckaert, P. (2002). Media context and advertising effectiveness: the role of context appreciation and context/ad similarity. Journal of Advertising, 31, pp. 49–61.

Derbaix, C.M. (1995). The impact of affective reactions on attitudes toward the advertisement and the brand: a step toward ecological validity. Journal of Marketing Research (JMR), 32, pp. 470–479.

Eagly, A.H. and Chaiken, S. (1993). The Psychology of Attitudes. Orlando, FL: Harcourt Brace Jovanovich College Publishers.

Edell, J.A. and Burke, M.C. (1987). The power of feelings in understanding advertising effects. Journal of Consumer Research, 14, pp. 421–433.

Engel, J.F., Blackwell, R.D. and Miniard, P.W. (1990). Con- sumer Behaviour. Orlando, FL: Dryden Press.

Epstein, S. (1994). Integration of the cognitive and psycho- dynamic unconscious. American Psychologist, 49, pp. 709–724.

Epstein, S., Pacini, R., Denes-Raj, V. and Heier, H. (1996). Individual differences in intuitive–experiential and analytical–rational thinking styles. Journal of Personality and Social Psychology, 71, pp. 390–405.

Erevelles, S. (1998). The role of affect in marketing. Journal of Business Research, 42, pp. 199–215.

Fazio, R.H. (1986). How do attitudes guide behaviour? In Sorrention, R.M. and Higgins, T.E. (eds), The Handbook of Motivation and Cognition: Foundations of Social Behavior. New York: Guildford, pp. 204–243.

Fazio, R.H. (1989). On the power of functionality of attitudes: the role of attitude accessibility. In Praktanis, A.R. et al. (ed.), Attitude Structure and Function. Hillsdale, NJ: Lawrence Erlbaum Associates, pp. 153– 179.

Fazio, R.H. (1990). Multiple processes by which attitudes guide behaviour: the MODE model as an integrative framework. Advances in Experimental Social Psychology, 23, pp. 75–109.

Fazio, R.H. (2001). On the automatic activation of associ- ated evaluations: an overview. Cognition & Emotion, 15, pp. 115–141.

Fazio, R.H. and Zanna, M.P. (1981). Direct experience and the attitude–behaviour consistency. In Berkowitz, L. (ed.), Advances in Experimental Social Psychology, Vol. 14. New York: Academic Press, pp. 161–202.

Feldman, J.M. and Lynch, J.G. (1988). Self-generated valid- ity and other effects of measurement on belief, attitude, intention, and behaviour. Journal of Applied Psychology, 73, pp. 421–435.

Fishbein, M. (1963). An investigation of the relationships between beliefs about an object and the attitudes toward that object. Human Relations, 16, pp. 233–240.

Fishbein, M. and Ajzen, I. (1975). Belief, Attitude, Intention and Behaviour: An Introduction to Theory and Research. Reading, MA: Addison-Wesley.

Fishbein, M. and Middlestadt, S. (1995). Noncognitive effects on attitude formation and change: fact or artifact? Journal of Consumer Psychology, 4, pp. 181–202.

Fishbein, M. and Middlestadt, S.E. (1997). A striking lack of evidence for non belief-based attitude formation and change: a response to five commentaries. Journal of Con- sumer Psychology, 6, pp. 107–115.

Fiske, S.T. and Taylor, S.E. (1984). Social Cognition. Reading, MA: Addison Wesley.

Fiske, S.T. and Taylor, S.E. (1991). Social Cognition, 2nd edn. New York: McGraw Hill.

Foxall, G. (1997). Marketing Psychology: The Paradigm in the Wings. London: Palgrave Macmillan.

Foxall, G. and Greenley, G. (1998). The affective structure of consumer situations. Environment and Behaviour, 30, pp. 781–798.

Foxall, G.R. and Yani-de-Soriano, M.M. (2005). Situational influences on consumers’ attitudes and behaviour. Journal of Business Research, 58, pp. 518–525.

Garbarino, E.C. and Edell, J.A. (1997). Cognitive effort, affect and choice. Journal of Consumer Research, 24, pp. 147–158.

Gorn, G.J. (1982). The effects of music in advertising on choice behaviour: a classical conditioning approach. Journal of Marketing, 46, pp. 94–101.

Gorn, G., Pham, M.T. and Sin, L.Y. (2001). When arousal influences ad evaluation and valence does not (and vice versa). Journal of Consumer Psychology, 11, pp. 43– 55.

Greenwald, A.G. and Banaji, M.R. (1995). Implicit social cognition: attitudes, self-esteem and stereotypes. Psycho- logical Review, 102, pp. 4–27.

Grewal, R., Mehta, R. and Kardes, F.R. (2004). The timing of repeat purchases of consumer durable goods: the role of functional bases of consumer attitudes. Journal of Mar- keting Research (JMR), 41, pp. 101–115.

Grossman, R.P. and Till, B.D. (1998). The persistence of classically conditioned brand attitudes. Journal of Adver- tising, 27, pp. 23–31.

Guerin, B. (1994). Analyzing Social Behaviour: Behaviour Analysis and the Social Sciences. Reno, NV: Context Press.

Herrington, J. and Oliver, R. (1995). Critical characteristics of situated learning: implications for the instructional design of multimedia [electronic version]. Proceedings of the Annual Conference of the Australian Society for Computers in Learning in Tertiary Education, Melbourne. Available at: http://www.ascilite.org.au/conferences/ melbourne95/smtu/papers/herrington.pdf (accessed 14 February 2011).

448 E. Argyriou and T.C. Melewar

Herrington, J. and Oliver, R. (1999). Using situated learning and multimedia to investigate higher-order thinking. Journal of Interactive Learning Research, 10, pp. 3–24.

Hirschman, E.C. and Holbrook, M.B. (1982). Hedonic con- sumption: emerging concepts, methods and propositions. Journal of Marketing, 46, pp. 92–101.

Holbrook, M.B. and Batra, R. (1987). Assessing the role of emotions as mediators of consumer responses to advertis- ing. Journal of Consumer Research, 14, pp. 404–420.

Isen, A.M. (2001). An influence of positive affect on deci- sion making in complex situations: theoretical issues with practical implications. Journal of Consumer Psychology, 11, pp. 75–85.

Jacoby, J., Morrin, M., Jaccard, J., Gurhan, Z., Kuss, A. and Maheswaran, D. (2002). Mapping attitude formation as a function of information input: online processing models of attitude formation. Journal of Consumer Psychology, 12, pp. 21–34.

Janiszewski, C. and Warlop, L. (1993). The influence of classical conditioning procedures on subsequent attention to the conditioned brand. Journal of Consumer Research, 20, pp. 171–189.

Jarvenpaa, S.L. (1989). The effect of task demands and graphical format on information processing strategies. Management Science, 35, pp. 285–303.

Jarvenpaa, S.L. and Dickson, G.W. (1988). Graphics and managerial decision making: research-based guidelines. Communications of the ACM, 31, pp. 764–774.

Kahneman, D. (1973). Attention and Effort. Englewood Cliffs: Prentice Hall.

Katz, D. (1960). The functional approach to the study of attitudes. Public Opinion Quarterly, 24, pp. 163–204.

Kellaris, J.J. and Cox, A.D. (1989). The effects of back- ground music in advertising: a reassessment. Journal of Consumer Research, 16, pp. 113–118.

Keller, K.L. (1993). Conceptualizing, measuring, managing customer-based brand equity. Journal of Marketing, 57, pp. 1–22.

Kim, J., Allen, C.T. and Kardes, F. (1996). An investigation of the mediational mechanisms underlying attitudinal conditioning. Journal of Marketing Research (JMR), 33, pp. 318–328.

Lazarus, R.S. (1991). Emotion and Adaptation. New York: Oxford University Press.

LeDoux, J.E. (1998). The Emotional Brain: The Mysterious Underpinnings of Emotional Life. New York: Simon & Schuster.

Lee, A.Y. and Labroo, A.A. (2004). The effect of conceptual and perceptual fluency on brand evaluation. Journal of Marketing Research (JMR), 41, pp. 151–165.

Lohse, G.L. (1997). The role of working memory on graphi- cal information processing. Behaviour & Information Technology, 16, pp. 297–308.

Lutz, R.J. (1981). A re-conceptualization of the functional approach to attitudes. Research in Marketing, 5, pp. 297– 308.

Lynch, J.G. Jr (2006). Accessibility–diagnosticity and the multiple pathway anchoring and adjustment model. Journal of Consumer Research, 33, pp. 25–27.

McFadden, D. (1999). Rationality for economists. Journal of Risk and Uncertainty, 19, pp. 73–106.

MacInnis, D.J. and Jaworski, B.J. (1989). Information pro- cessing from advertisements: toward an integrative frame- work. Journal of Marketing, 53, pp. 1–23.

Mandel, N. and Johnson, E.J. (2002). When Web pages influence choice: effects of visual primes on experts and novices. Journal of Consumer Research, 29, pp. 235–245.

Mandler, G., Nakamura, Y. and Van Zandt, B.J. (1987). Nonspecific effects of exposure on stimuli that cannot be recognized. Journal of Experimental Psychology: Learn- ing, Memory, and Cognition, 13, pp. 646–648.

Mantel, S.P. and Kardes, F.R. (1999). The role of direction of comparison, attribute-based processing, and attitude- based processing in consumer preference. Journal of Con- sumer Research, 25, pp. 335–352.

Mitchell, A.A. and Olson, J.C. (1981). Are product attribute beliefs the only mediator of advertising effects on brand attitude? Journal of Marketing Research (JMR), 18, pp. 318–332.

Nowlis, S.M. and Simonson, I. (1997). Attribute–task com- patibility as a determinant of consumer preference rever- sals. Journal of Marketing Research (JMR), 34, pp. 205– 218.

Olson, M.A. and Fazio, R.H. (2001). Implicit attitude for- mation through classical conditioning. Psychological Science, 12, pp. 413–417.

Park, C.W. and MacInnis, D.J. (2006). What’s in and what’s out: questions on the boundaries of the attitude construct. Journal of Consumer Research, 33, pp. 16–18.

Payne, J.W. (1982). Contingent decision behaviour. Psycho- logical Bulletin, 92, pp. 382–402.

Payne, J.W., Bettman, J.R. and Johnson, E.J. (1992). Behav- ioural decision research: a constructive processing per- spective. Annual Review of Psychology, 43, pp. 87–131.

Petty, R.E. and Cacioppo, J.T. (1981). Issue involvement as a moderator of the effects on attitude of advertising content and context. Advances in Consumer Research, 8, pp. 20–24.

Petty, R.E. and Wegener, D.T. (1997). Attitudes and attitude change. Annual Review of Psychology, 48, pp. 609–647.

Petty, R.E. and Wegener, D.T. (1998). Attitude change: multiple roles for persuasion variables. In Gilbert, D., Fiske, S. and Lindzey, G. (eds), The Handbook of Social Psychology, 4th edn. New York: McGraw-Hill, pp. 323– 390.

Pham, M.T. (1998). Representativeness, relevance and the use of feelings in decision making. Journal of Consumer Research, 25, pp. 144–159.

Pham, M. (2004). The logic of feeling. Journal of Consumer Psychology, 14, pp. 360–369.

Pham, M.T., Cohen, J.B., Pracejus, J.W. and Hughes, G.D. (2001). Affect monitoring and the primacy of feelings in

Consumer Attitudes Revisited 449

judgment. Journal of Consumer Research, 28, pp. 167– 188.

Pieters, R., Wedel, M. and Batra, R. (2010). The stopping power of advertising: measures and effects of visual com- plexity. Journal of Marketing, 74, pp. 48–60.

Priester, J.R., Dholakia, U.M. and Fleming, M.A. (2004). When and why the background contrast effect emerges: thought engenders meaning by influencing the perception of applicability. Journal of Consumer Research, 31, pp. 491–501.

Priluck, R. and Till, B.D. (2004). The role of contingency awareness, involvement, and need for cognition in attitude formation. Academy of Marketing Science Journal, 32, pp. 329–344.

Reed, A., II, Wooten, D.B. and Bolton, L.E. (2002). The temporary construction of consumer attitudes. Journal of Consumer Psychology, 12, pp. 375–388.

Russo, J.E. (1977). The value of unit price information. Journal of Marketing Research (JMR), 14, pp. 193–201.

Sanbonmatsu, D.M. and Fazio, R.H. (1990). The role of attitudes in memory-based decision making. Journal of Personality and Social Psychology, 59, pp. 614–622.

Schlosser, A.E. (1998). Applying the functional theory of attitudes to understanding the influence of store atmo- sphere on store inferences. Journal of Consumer Psychol- ogy, 7, pp. 345–369.

Schlosser, A.E. (2003a). Computers as situational cues: implications for consumers’ product cognitions and atti- tudes. Journal of Consumer Psychology, 13, pp. 103–112.

Schlosser, A.E. (2003b). Experiencing products in the virtual world: the role of goal and imagery in influencing attitudes versus purchase intentions. Journal of Consumer Research, 30, pp. 184–198.

Schwarz, N. (1997). Moods and attitude judgments: a comment on Fishbein and Middlestadt. Journal of Con- sumer Psychology, 6, pp. 93–98.

Schwarz, N. (1998). Warmer and more social: recent devel- opments in cognitive social psychology. Annual Review of Sociology, 24, pp. 239–264.

Schwarz, N. (1999). Self-reports. American Psychologist, 54, pp. 93–105.

Schwarz, N. (2006). Attitude research: between Ockham’s razor and the fundamental attribution error. Journal of Consumer Research, 33, pp. 19–21.

Schwarz, N. and Bohner, G. (2001). The construction of attitudes. In Tesser, A. and Schwarz, N. (eds), Handbook of Social Psychology: Intraindividual Processes. Hills- dale, NJ: Lawrence Erlbaum Associates, pp. 436–457.

Schwarz, N. and Clore, G.L. (1988). How do I feel about it? Informative functions of affective states. In Fiedler, K. and Forgas, J. (eds), Affect, Cognition and Social Behav- iour. Toronto: Hogrefe International, pp. 44–62.

Schwarz, N. and Clore, G.L. (1996). Feelings and phenom- enal experiences. In Higgins, T.E. and Kruglanski, A. (eds), Social Psychology: Handbook of Basic Principles. New York: Guildford, pp. 433–465.

Schwarz, N. and Clore, G.L. (2003). Mood as information: 20 years later. Psychological Inquiry, 14, pp. 296–303.

Schwarz, N. and Strack, F. (1991). Context effects in attitude surveys: applying cognitive theory to social research. European Review of Social Psychology, 2, pp. 31–50.

Shavitt, S. (1989). Products, personalities and situations in attitude functions: implications for consumer behaviour. Advances in Consumer Research, 16, pp. 300–305.

Shavitt, S. (1990). The role of attitude objects in attitude functions. Journal of Experimental Social Psychology, 26, pp. 124–148.

Shimp, T.A. (1981). Attitude toward the ad as a mediator of consumer brand choice. Journal of Advertising, 10, pp. 9–48.

Shimp, T.A., Stuart, E.W. and Engle, R.W. (1991). A program of classical conditioning experiments testing variations in the conditioned stimulus and context. Journal of Consumer Research, 18, pp. 1–12.

Shiv, B. and Fedorikhin, A. (1999). Heart and mind in con- flict: the interplay of affect and cognition in consumer decision making. Journal of Consumer Research, 26, pp. 278–292.

Sloman, S.A. (1996). The empirical case for two systems of reasoning. Psychological Bulletin, 119, pp. 3–22.

Slovic, P. (1972). From Shakespeare to Simon: speculations and some evidence about man’s ability to process infor- mation. Oregon Research Institute Bulletin, 12, pp. 10–23.

Slovic, P. (1995). The construction of preference. American Psychologist, 50, pp. 364–371.

Smith, M., Brewster, J.S. and White, R.W. (1956). Opinions and Personality. New York: Wiley.

Snyder, M. and DeBono, K.G. (1987). A functional approach to attitudes and persuasion. In Zanna, M., Olson, J.M. and Herman, C. (eds), Social Influence: The Ontario Symposium, Vol. 5. Hillsdale, NJ: Erlbaum, pp. 107–125.

Spangenberg, E.R., Voss, K.E. and Crowley, A.E. (1997). Measuring the hedonic and utilitarian dimensions of atti- tude: a generally applicable scale. In Brucks, M. and MacInnis, D.J. (eds), Advances in Consumer Research, Vol. 24. Provo, UT: Association for Consumer Research, pp. 235–341.

Srull, T.K. (1983). Affect and memory: the impact of affec- tive reactions in advertising on the representation of product information in memory. Advances in Consumer Research, 10, pp. 520–525.

Stevenson, J.S., Bruner, G.C., II and Kumar, A. (2000). Webpage background and viewer attitudes. Journal of Advertising Research, 40, pp. 29–34.

Stuart, E.W., Shimp, T.A. and Engle, R.W. (1987). Classical conditioning of consumer attitudes: four experiments in an advertising context. Journal of Consumer Research, 14, pp. 334–349.

Tesser, A. (1978). Self-generated attitude change. In Berkowitz, L. (ed.), Advances in Experimental Social Psychology. New York: Academic Press, pp. 408–448.

450 E. Argyriou and T.C. Melewar

Tversky, A. and Kahneman, D. (1974). Judgment under uncertainty: heuristics and biases. Science, 185, pp. 1124–1131.

Tversky, A. and Kahneman, D. (1991). Loss aversion in riskless choice: a reference-dependent model. Quarterly Journal of Economics, 106, pp. 1039–1061.

Tversky, A., Sattath, S. and Slovic, P. (1988). Contingent weighting in judgment and choice. Psychological Review, 95, pp. 371–384.

Vakratsas, D. and Ambler, T. (1999). How advertising works: what do we really know? Journal of Marketing, 63, pp. 26–43.

Voss, K.E., Spangenberg, E.R. and Grohmann, B. (2003). Measuring the hedonic and utilitarian dimensions of con- sumer attitude. Journal of Marketing Research (JMR), 40, pp. 310–320.

Vygotski, L. (1978). Mind in Society: The Development of Higher Psychological Processes. Cambridge, MA: Harvard University Press.

Winkielman, P., Schwarz, N., Reber, R. and Fazendeiro, T.A. (2003). Cognitive and affective consequences of visual fluency: when seeing is easy on the eye. In Scott, L.M. and

Batra, R. (eds), Persuasive Imagery: A Consumer Response Perspective. Mahwah, NJ: Lawrence Erlbaum Associates, pp. 75–89.

Wright, P. (1975). Consumer choice strategies: simplifying vs. optimizing. Journal of Marketing Research (JMR), 12, pp. 60–67.

Yeung, C.W.M. and Wyer, R.S. Jr (2004). Affect, appraisal, and consumer judgment. Journal of Consumer Research, 31, pp. 412–424.

Yoo, C., Park, J. and MacInnis, D. (1998). Effects of store characteristics and in-store emotional experiences on store attitudes. Journal of Business Research, 42, pp. 253–263.

Zajonc, R.B. (1968). Attitudinal effects of mere exposure. Journal of Personality and Social Psychology, 9, pp. 1–27.

Zajonc, R.B. (1980). Feeling and thinking: preferences need no inferences. American Psychologist, 35, pp. 151–175.

Zajonc, R.B. (1984). On the primacy of affect. American Psychologist, 39, pp. 117–123.

Zajonc, R.B. (1998). Emotions. In Gilbert, D.T., Fiske, S.T. and Lindzey, G. (eds), The Handbook of Social Psychol- ogy, 4th edn. Boston, MA: McGraw-Hill, pp. 591–632.

Consumer Attitudes Revisited 451

may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express

written permission. However, users may print, download, or email articles for individual use.