受众心理与广告效果（英文版）

Understanding Advertising Effectiveness from a Psychological Perspective: The Importance of Attitudes and Attitude Strength Derek D. Rucker, Richard E. Petty and "Advertising is persuasion, and persuasionis not a science. but an art. Advertising is the art of persuasion. " William Bernbach, quoted in Stephen Fox, The Mirror Makers, 1984. Like Bembach, one of advertising's greatest pioneers, many have directly linked the study of advertising to the study of persuasion. For instance, previous volumes on advertising have included chapters that integrate work from the persuasion literature (Tellis, 1998, 2004). Indeed, one of the dominant goals of advertising is to persuade the consumer of the benefits of a product or service and to stimulate a purchase. While we are like­ minded with Bembach when it comes to the notion that advertising is tied to persuasion, we diverge markedly from Bembach in that we believe persuasion, and therefore Joseph R. Priester advertising, is better conceptualized as a science. As a science, we can adopt theoretical perspectives and examine empirical findings to understand factors that contribute to successful advertising. To this end, we draw on the vast scientific literature in psychology to understand some of the underpinnings of successful advertising. CHAPTER OVERVIEW This chapter provides a framework for under­ standing the factors that contribute to adver­ tising effectiveness by attending to two rich literatures in the domain of persuasion. First, we focus on classic and contemporary theo­ retical approaches to understanding attitude change. We discuss conceptualizations and findings that help understand when various ....... 75UNDERSTANDING ADVERTISING EFFECTIVENESS FROMA PSYCHOLOGICAL PERSPECTIVE74 HOW ADVERTISING WORKS elements of advertisements will be successful in promoting favourable attitudes. Second, we focus on work on attitude strength. As will be discussed, attitude strength is important for advertising effectiveness as it determines whether attitudes resulting from exposure to advertisements will be persistent, resistant, and likely to influence behaviour. ATTITUDES ANDATTITUDE CHANGE The attitude construct Advertisers are interested in outcomes such as persuasion, product-relevant thoughts and beliefs, cognitive processes, emotions, inten­ tions to buy, and actual purchase behaviour. Attitudes playa unique and important role in relation to each of these constructs. Attitudes have typically been conceptualized as one's overall, global evaluation of an object. That is, attitudes reflect whether individuals like or dislike specific products, brands, advertisements, or spokespersons (Petty and Cacioppo, 1981; Petty and Wegener, 1999). For example, overall, do you like or dislike Pizza Hut Meat Lover's Pizza? You might answer anywhere from extreme liking to extreme disliking, with varying gradations in-between. Your answer would represent your attitude. To more quantitatively measure attitudes, researchers often use a series of semantic differential scales (e.g., good-bad, favourable-unfavourable), with the specific response associated with a number on the scales.' This overall reaction (i.e., one's attitude) has traditionally been conceptualized as encompassing and reflecting three distinct bases; thoughts, emotions and feelings, and one's behaviours. That is, one's overall attitude is the result of and stems from one's affective, cognitive, and behavioural reactions toward that pizza, for example.? As such, if an advertisement influences any one of the three components, that change will also be reflected in that individual's overall attitude. For example, if a television commercial for a greeting card engenders warm feelings in an individual, the extent to which that warmth is transferred to the specific greeting card brand will emerge on measures of the individual's overall attitude toward the brand. Similarly, changes in cognitions and/or behaviours associated with an attitude object will be reflected in the overall attitude. A veritable bounty of research has demon­ strated that one's attitude influences one's behavioural intentions, and that behavioural intentions best predict behaviour (e.g., Fishbein and Ajzen, 1981; Sheppard et al., 1988). Thus, whether people purchase a particular product will be determined, in part, by their attitudes. Studying and identi­ fying consumers' attitudes also has a num­ ber of important advantages over studying behaviour itself. An attitude is a global evaluation that can be used to make gen­ eral predictions about a variety of specific behaviours (e.g., willingness to receive addi­ tional information about a product, word-of­ mouth, willingness to pay a price premium, purchase behaviour, etc.). Rarely, if ever, is it possible to study all the various behavioural implications of an advertisement. However, by studying global attitudes it is possible to gauge how consumers are likely to act in general across a variety of attitude-relevant situations. In short, the construct of attitude not only captures the affective, behavioural, and cognitive reactions to products, services, and brands, but also predicts behavioural intentions, and ultimately, behaviour. Historical approaches Much contemporary work on persuasion has its roots in the empirical investigations con­ ducted by Carl Hovland and his colleagues at Yale in the 1950s (e.g., Hovland et al., 1953). Importantly, a key emphasis of this work was on the importance of attention, learning and recall processes for message effectiveness, a theme also seen in work on advertising (e.g., Loken and Hoverstad, 1985). Hovland and his colleagues also examined the role of source, message, and recipient factors in persuasion. Source factors included variables associated with the person delivering the message, such as the credibility or attractiveness of the source (Hovland and Weiss, 1951). Message factors consisted of variables such as the number of arguments (Calder et al., 1974), and whether the message presented arguments that were solely in favour of the position or also disclosed the counterpoints (Hovland et aI., 1953). Recipient or audience factors included more stable individual differences in the audience such as intelligence (McGuire, 1968), but also more temporary influences such as the audience's emotional state (Zanna et al., 1970). A focal aspect of early research was the emphasis on uncovering the single effect of a variable on persuasion; that is, whether a variable, such as source credibility, increased or decreased persuasion. This early work was often accompanied by a focus on a "single-process" perspective where researchers attempted to uncover the single means by which a variable produced its effect (see Petty, 1997). For example, if source cred­ ibility increased persuasion, then researchers were interested in identifying the single, and presumably only, process by which it did so (e.g., increasing attention to the message). InitiaIly, work that could be classified into the single effect approach was very fruitful. For instance, regarding the source of the message, credible sources were found to produce greater persuasion than less-credible sources (Hovland and Weiss, 1951). Work on the message itself revealed that more arguments, up to a point, were found to yield greater persuasion (Calder et aI., 1974).Workon audi­ ence characteristics suggested that associating negative emotions with a message decreased persuasion (Zanna et al., 1970). However, the single-effect approach soon became untenable as research began to produce contradictory findings. For instance, some research indi­ cated that credible sources could be associated with less persuasion (Stemthal et aI., 1978), increasing the number of arguments did not always increase persuasion (Norman, 1976), and that negative emotions could actuaily enhance persuasion (Rogers, 1983). Contemporary approach: The elaboration likelihoodmodel The prevalence of conflicting studies posed a serious threat to the field of attitude change and its scholars. Indeed, at one point it seemed as if the field might be destined for oblivion by the perception that there were no reliable or systematic results (e.g., Fishbein andAjzen, 1972). In response to this pervasive threat the Elaboration Likelihood Model (ELM; Petty and Cacioppo, 1981, 1986) of persuasion was developed to resolve these seemingly incompatible findings under an integrated framework. For over 25 years the ELM has guided many studies of persuasion (see Petty and Wegener, 1999) and we use it as an organizing tool here.' The ELM approaches understanding per­ suasion by focusing on two routes to attitude change that serve as endpoints along an extent of thinking continuum - a relatively thoughtful route in which people scruti­ nize the issue-relevant information presented (the central route), and a less thoughtful route in which people rely on simple associations or heuristics to reach decisions (the peripheral route). The route that produces persuasion is determined by the amount of elaboration, or thinking, in which people engage. Furthermore, route selection has implications for the durability and impact (i.e., the strength) of attitudes. Finally, in stark contrast to the "single-effect and single­ process" approaches taken in earlier research, the ELM posits that any variable (e.g., source credibility) can have multiple effects on persuasion and do so through a variety of pro­ cesses in different situations. We next explore the implications of this model in greater detail. The central route According to the ELM, the central route to persuasion involves an effortful scrutiny of issue-relevant information in an attempt to determine the central merits of the product or service under consideration. For example, consumers engaged in the central route while receiving a car advertisement would be prone to carefully scrutinize the advertisement for 76 HOW ADVERTISING WORKS relevant information, In such a situation, in the thoughts they generate the more these consumers would evaluate each piece of thoughts determine consumers' attitudes information available whether it stems from (Brinol et al., 2004; Petty et al., 2002).4 the source, message, or themselves (e.g., There are two prerequisites for consumers brand, warranty, endorser, their feelings) with to engage in the central route to per­ respect to whether the information was a suasion. First, consumers must have the relevant favourable or unfavourable piece ability to scrutinize a message carefully. of evidence. When this occurs, consumers' Consumers must be able to understand the attitudes towards the advertised product message and have the resources to attend are determined primarily by their cognitive to the information present. Factors affecting responses or thoughts to the information consumers' ability to process a message available (see Cacioppo and Petty, 1981; include, consumers' knowledge (Wood and Petty and Cacioppo, 1986; see also Batra and Lynch, 2002), the amount of distraction Ray, 1986 and Wright, 1973, for illustrative present in the environment (Petty et al., examples). If people generate predominantly 1976), and the number of times the message positive thoughts as a result of scrutinizing is repeated (Cacioppo and Petty, 1979b). the information, a positive attitude will result; Some specialized consumer persuasion the­ if scrutiny leads to predominantly negative ories such as the resource matching model thoughts a negative attitude will follow. If (e.g., Anand and Stemthal, 1988; Peracchio people generate a mix of positive and negative and Meyers-Levy, 1997) have focused on thoughts a moderate attitude will result. the importance of matching the cognitive Importantly, the notion of cognitive resources available to the consumer with the responses deviates from earlier approaches resources needed to process the message. that emphasized message learning and recall According to this approach, persuasion is (e.g., Hovland et al., 1953). According to greatest when available resources match those the ELM, under high elaboration, it is not required to process the message. According necessarily the specific information that to the ELM, this would primarily be true consumers can recall about the product that only when the message presents evidence determines their attitudes, but rather it is their that is compelling when scrutinized. If the idiosyncratic responses to this information, message presents weak arguments, then hav­ For example, two consumers might both ing sufficient resources would allow people recall that a new plasma television features to recognize the flaws in the arguments an integrated DVD player. However, one leading to reduced persuasion. When people person might evaluate the integrated DVD have too many resources needed for the feature positively (e.g., "Great, I don't have to message, they may become bored and find buy a DVD player"), but another negatively the message tedious, leading to reduced (e.g., "I already have a great DVD player, persuasion, as is the case with excessive and I don't want to pay for a feature I don't message repetition (see Cacioppo and Petty, need"). Consequently, whereas theories 1979, 1989). Alternatively, if people have focusing on recall would suggest including excess resources, they could generate their this feature would have a similar effect across own unique thoughts that could augment or consumers as long as people attend to and detract from the message depending on their leam the attribute, contemporary theories valence. The latter is most likely under high such as the ELM recognize it is not recall per elaboration conditions when variables in the se, but the idiosyncratic thoughts individuals situation induce a bias to the processing (see have about the advertised features. More discussion of biased processing). recent research has stressed that in addition The central route to persuasion not only to consumers' cognitive responses, the requires ability, but also motivation to confidence people have in their thoughts is process the message. Consumers' motivation important. The more confident consumers are to process a message can be influenced by UNDERSTANDING ADVERTISING EFFECTIVENESS FROMA PSYCHOLOGICAL PERSPECTIVE 77 a number of variables, such as the personal have that seem valid, the more they will resist relevance of an issue (petty and Cacioppo, persuasion. 1979b), people's general enjoyment of thinking (Cacioppo et al., 1983), and being The peripheral route personally responsible for processing the When people lack either the motivation message (petty et al., 1980). When motivation or ability to process advertisements, the and ability to process a message are present, peripheral route is taken. In the peripheral advertisers can expect consumers to rely route, consumers' attitudes are determined upon the cognitive responses in which primarily by simple decision processes such they have confidence to determine their as classical conditioning (Staats and Staats, reaction to the message. Confidence in one's 1958), mere exposure (Zajonc, 1968), and cognitive responses can also stem from use of decision heuristics (Chaiken, 1980). source, message, and recipient factors (see Heuristics represent mental shortcuts that Brinol and Petty, 2004, for a review). For allow for a simple decision based on a rule of example, people will have more confidence thumb. Examples of heuristics are "Experts in their thoughts the more easily they come are usually right," and "Higher prices mean to mind (Tonnala et al., 2002). better quality." When relying on heuristics, This conceptualization stresses that what consumers might be more persuaded by a message recipients carry with them under car advertisement that is accompanied by high elaboration conditions: memory of the an attractive model than an advertisement cognitive responses, rather than memory for without a model. This might occur because the specific executional elements of the ad, the attractive model creates positive affect that is critical (Petty, 1977; see also Mazzocco becomes associated with the car. However, et al., 2005). Thus, it is not the case that to the extent that the attractiveness of the under high elaboration conditions people model is serving as a simple cue, including will necessarily have greater memory for an attractive model in an advertisement would the information in the ad. Rather, when be less likely to persuade consumers who are individuals have the motivation and abil­ engaged in central route processing, unless ity to think, it is the thoughts that they as explained subsequently, attractiveness is have in response to the advertisement (i.e., serving in some role other than as a simple their cognitive responses) that determine ad cue. Other variables that are capable of serving effectiveness. This perspective suggests that as peripheral cues include the expertise of ad recall and attention may be unrelated the message source (Petty et al., 1981), the to whether the advertisement is effective number of arguments presented (Petty and under high rather than low elaboration. Cacioppo, 1984), and one's mood (Petty et al., The process of cognitive responses being 1993). responsible for the formation and change To understand the type of psychological of attitudes is that hallmark of the central process that might be involved in the periph­ route of persuasion. Under high elaboration, eral route, consider work on classical condi­ a number of properties of the thoughts tioning. Classical conditioning represents the themselves are important. The most studied process whereby an unconditioned stimulus aspects of thoughts are their valence (how (e.g., a novel product) becomes associated favourable or unfavourable they are), their with a conditioned stimulus (e.g., music that number (how many thoughts are generated), elicits happiness). After sufficient pairing of and the confidence with which people hold the unconditioned stimulus with the condi­ their thoughts (how valid people believe their tioned stimulus, the novel product becomes thoughts are). Under the central route, the associated with and elicits the feelings of more favourable thoughts people have that happiness, even in the absence of the music seem valid, the more they will be persuaded (a.k.a., conditioned response). Mere exposure and the more unfavourable thoughts people represents the process whereby one's liking ~-_. 78 HOW ADVERTISING WORKS for an attitude object can be increased by ofpersuasion. Arguably themost fundamental repeated, non-focal exposures. For example, reason for studying attitudes, such as those the more one is exposed (albeit out of focal following advertisements, is that people's attention) to a melody, the more one likes evaluations were thought to be instrumental that melody. Consistent with contemporary in guiding action and therefore useful in theorizing, both of these processes are more predicting behaviour (petty and Wegener, influential under conditions oflow elaboration 1999). However, as soon as the first serious likelihood. For example, novel stimuli are investigations of attitudes had begun, chal­ easier to classically condition than familiar lenges were raised regarding whetherattitudes ones (Cacioppo et al., 1992). Similarly, the could predict behaviour (e.g., La Piere, mere exposure effect emerges more when 1934). the stimuli are perceived without awareness In response to this challenge, attitude (Bornstein and Dagostino, 1992; Bornstein r