Consumer Skepticism and the Awaiting Room of the Mind@: Are Consumers More Likely to Believe Advertising Claims If They Are Merely Comprehended?

ABSTRACT - Traditional conceptualizations of information processing have held that incoming information is relegated to a sort of Awaiting room of the mind,@ with consumers processing it in a sequential fashion. However, the work of Gilbert and his colleagues poses an alternative Spinozan conceptualization, that consumers initially believe all incoming information, and in some cases go on to process the veracity of this information. This investigation tests Gilbert’s alternative Spinozan model within an advertising context for the first time, and provides support for the findings that it is more difficult to disbelieve than to believe advertising information.

Citation:

Scott Koslow and Richard F. Beltramini (2002) ,"Consumer Skepticism and the Awaiting Room of the Mind@: Are Consumers More Likely to Believe Advertising Claims If They Are Merely Comprehended?", in NA - Advances in Consumer Research Volume 29, eds. Susan M. Broniarczyk and Kent Nakamoto, Valdosta, GA : Association for Consumer Research, Pages: 473-479.

Advances in Consumer Research Volume 29, 2002     Pages 473-479

CONSUMER SKEPTICISM AND THE "WAITING ROOM OF THE MIND": ARE CONSUMERS MORE LIKELY TO BELIEVE ADVERTISING CLAIMS IF THEY ARE MERELY COMPREHENDED?

Scott Koslow, University of Waikato

Richard F. Beltramini, Wayne State University

[The authors warmly than Sheila Sasser who assisted in data collection.]

ABSTRACT -

Traditional conceptualizations of information processing have held that incoming information is relegated to a sort of "waiting room of the mind," with consumers processing it in a sequential fashion. However, the work of Gilbert and his colleagues poses an alternative Spinozan conceptualization, that consumers initially believe all incoming information, and in some cases go on to process the veracity of this information. This investigation tests Gilbert’s alternative Spinozan model within an advertising context for the first time, and provides support for the findings that it is more difficult to disbelieve than to believe advertising information.

INTRODUCTION

Advertising researchers generally believe that consumers acquire information from advertising, and choose for themselves what they will believe, or accept as true, based on the information available. Most information acquisition models hold that believing a persuasive advertising message is a separate and independent step in processing that information (i.e., independent from exposure, evaluation, comprehension, etc.). For example, Bettman (1979) and Mitchell (1983) establish a tradition of models which include the sequential steps from advertising exposure, comprehension, believability, attitude change, and ultimately through behavior. And, Meyer (1982); Brucks, Mitchell, and Staelin (1984); Finn (1984); Moore and Whinston (1986); Urbany, Bearden, and Weilbaker (1988); and Ford, Smith, and Swasy (1990) among others concur that evaluating, comprehending, etc. an advertising claim and actually believing it are separate mental processes.

For the most part, advertising researchers conceptualize consumers’ believability of advertising information as occurring in a sort of a "waitingroom of the mind." That is, upon exposure, advertising information is "ushered" into a sort of mental waiting room in which claims reside until the skeptical consumer decides to label them "true" or "false" given prior information stored from experience. Consumers thus defer judgments about the veracity of these advertising claims for a time spent in waiting again, while the consumer decides whether or not to believe the claim (Simon, Huber, and Payne 1985; Rao and Sieben 1992).

Consumer psychologists too ascribe similar notions of how we come to believe (or not believe) advertising information. For example, in discussing attitude formation, Zimbardo and Lieppe (1991) separate out the distinct, sequential steps of attention, comprehension, and acceptance of a message. That is, message receivers must determine if they will believe the message content. Similarly, Miller and Stiff (1993) note that interpersonal information goes through a separate mental step to determine the veracity of acquired information, much like Margolis’ earlier (1987) cognitive theory in which he inserts the mental step of first determining if the information "looks right" (p. 98), or still earlier, Cofer’s (1979) notion of the internal parser function from an artificial intelligence perspective.

However, there is still a great deal of evidence that suggests that consumers frequently believe information that the "waiting room" approach suggests they should not. For example, repeating exposure to claims without evidence increases their believability (e.g., Hawkins and Hoch 1992). Once baseless claims are believed, it is difficult for consumers to reverse these beliefs (e.g., Tybout, Calder and Sternthal 1981). And frequently people believe claims that are explicitly labelled as incorrect or are misleading (e.g., Gaeth and Heath 1987). If consumers were always capable of judiciously determining what to believe such effects should not occur.

These anomalies to the "waiting room" conceptualization lead Gilbert (1991) to propose a starkly contrasting model. Gilbert reviews a long series of related research and argues that on a mechanical level, the mind operates such that it initially believes all information that is comprehended. The mind then may go on to reprocess this information in a second mental step, and only then considers whether it is true or false. Therefore, it is easy to believe what one reads or hears, but to not believe claims actually takes more effort and processing (Gilbert, Krull and Malone 1990; Gilbert, Tafarodi and Malone 1993). Gilbert and his colleagues designed research to distinguish between the two models, and find support for the alternative model (Gilbert, Tafarodi and Malone 1993). However, this research has not been applied in an advertising context, an area in which consumers are known to be routinely skeptical.

Therefore, our investigation explores this alternative model of information processing and believability specifically within an advertising context. And because this model suggests a somewhat undesirable tendency for consumers to believeCat least initiallyCall information (even when they should not), this paper also addresses the implications for consumer research and advertising regulation.

BACKGROUND

The origins of the "waiting room" approach to believing information can be traced to the writings of Rene Descartes. In reaction to the intellectual current of the time, Descartes proposed the idea that one can and should question the information that one is given. Specificaly, he proposed that individuals hold propositions in aequilibrio, and then later determine whether such propositions are true or false. Over the last three and a half centuries, researchers have adopted this skepticism as the guiding rule of science, and we have developed specific philosophical mechanisms to "ensure" skepticism (e.g., Lincoln and Guba 1985), again assuming that the human mind naturally works in this same mechanistic manner (Gilbert, Tafarodi, and Malone 1993). This "Cartesian" view has become commonplace in modern psychology.

More recently, Gilbert, Krull, and Malone (1990), incorporating the work of Descartes’ contemporary Baruch Spinoza, suggest that the consumer’s mind works differently than Descartes suggested. Spinoza’s alternative view is that information is always first taken to be true, and only then can it be rejected as false. They elaborate on this by presenting an interesting analogy, that of the library of the mindCwhere facts in memory are likened to books on a library’s shelves.

Under Descartes’ view, new books first enter a holding area of the library, where the veracity of the book can be determined. After such a determination is made, the book is coded, say fiction or non-fiction. Therefore, when one goes to access some book from library shelves, it is clearly coded one way or the other. The Spinozan alternative Gilbert and his colleagues suggest is that when books first arrive at the library, they are assumed to be non-fiction (true), and therefore go directly to the shelves unmarked. If books are judged to be fiction, only then are they marked. Descartes’ approach requires all information to be filtered through two processing stages: comprehension, followed by a determination of veracity. Spinoza’s alternative approach is less cumbersome, and in a comparison of the two alternatives, Gilbert et al. provide empirical support for Spinoza’s view rather than Descartes’. It is important to emphasize that Gilbert et al. measured subjects’ perceptions of whether a fact were identified as true or false, and not merely that subjects understood it.

Perhaps from an evolutionary standpoint, this appears justified. Of the great amount of information our minds typically process, a relatively small percentage is faulty, and so the Spinozan alternative system minimizes processing effort. However, when it comes to an information context where "facts" are quite often false, one might suppose the system could be reversed and false statements are left uncoded, while true ones are coded. However, Gilbert et al. find no evidence that such a reversal is possible.

Even when subjects were advised of the falsity of certain information before the actual information was revealed, subjects still processed the information in a manner that indicated they took the information to be true at first, and then reprocessed it later as false. As Gilbert and Malone (1990) note, "merely comprehending a false proposition increased the likelihood that subjects would later consider it true."

However, given the prevalence of skepticism in the advertising believability literature (e.g., Beltramini and Stafford 1993; Brucks, Armstrong, and Goldberg 1988; Burke, DeSarbo, Oliver, and Robertson 1988; Calfee and Ringold 1988, 1994; Goldberg and Hartwick 1990; Kirmani 1990; Preston 1994; and others), this situation may prove an exception to Gilbert and his colleagues’ findings. There are two rationals one may apply.

First, Fiske and Pavelchak (1986) proposed that there are two modes in which people make stimulus evaluations. One is categorization, in which the stimulus is fitted to available schemata and the evaluation follows from the selected schema. For example, if one knows that one is speaking to a used car salesman, and one successfully categorizes the salesperson in one’s schema of persuasion knowledge (see Friestad and Wright 1994) then evaluations of the salesperson will be based on that schema. However, if one cannot easily categorize, then piecemeal processing will occur to arrive at an evaluation. If consumers categorize the claims as advertisements, the all the negativity normally directed at advertising (see Mittal 1994) should lead to negative evaluations. However, if they do not, piecemeal processing should occur, which is fairly deep, and evaluations will probably be negative if the claim is known or can be determined to be false. Thus it should be difficult to get consumers to believe false claims.

Second, though the work of Gaeth and Heath (1987) also suggests that much of this skepticism doesn’t come naturally, they do suggest that it can be learned. So, possibly prior exposure to advertising may be able to teach consumers to question advertising more automatically, and the effect Gilbert et al. found might be avoided in advertising situations.

Therefore, the key research question we pose is when consumers process advertising, do they really first hold claims in a "waiting room" and then later consider whether or not the claims are true or false? In this sense, we attempt to empirically assess this unique alternative approach to consumers belief of advertising information presented above vis a vis the more conventional consumer information processing perspective.

METHOD

A within-subject 2 x 2 experiment was employed to test the hypothesis. The first factor was whether the claims were true to false, and the second was whether subjects were doing deep or shallow processing. Subjects were introduced to a new pharmaceutical brand called SynFOKtin. The product was bogus and claimed to work on a hypothetical disorder called GMS syndrome. Subjects were given time to learn 16 statements regarding the product and were familiarized with versions of SynFOKtin advertisements with claims deleted. Afterwards, subjects viewed advertisements under low and high processing constraints and their judgements of the veracity of claims were recorded and analyzed.

The product category of pharmaceuticals was chosen due to the category’s frequent use of factual claims in advertisements. Five widely circulated magazines were used to identify a set of typical advertisements for the category. Thirteen different advertisements were identified, and the stimulus advertisement was modeled after them. A sample stimulus advertisement is presented in the Figure. Although no assumptions of broad generalizability are made here, the selection was based upon preliminary analysis to identify those publications with the greatest number of pharmaceutical advertisements.

Subjects were 51 students at a major Mid-Western public university. These 51 included 19 male and 16 female undergraduate students and 6 male and 10 female graduate students. Non-native speakers of English were not used and neither were any subjects who guessed the purpose of the study. This sample size is consistent with other research employing in-depth interviews (e.g., Gilbert, Krull and Malone 1990; Gilbert, Tafarodi and Malone 1993). In fact, because of the repeated measures, a total of 816 measures actually represents the database used in our analysis.

Manipulations

Because Gilbert’s (1991) approach suggests that mere comprehension leads to higher believability, it is important to manipulate subjects such that in some situations they merely comprehend, and in others they go beyond comprehension to do deeper processing on the truth or falsity of advertising claims. Gilbert and his colleagues describe two general types of manipulations: interruption and distraction.

In the interruption manipulation, subjects read statements and are given either limited or ample time to process a statement before having to be directed to move on to the next statement. By doing so, subjects who are under tight time constraints have less capacity to consider whether the statements read are true or false, so may be more likely to take them as true. In the second type of manipulation, subjects are sometimes given a modest distraction task while reading statements. In situations with the distraction task, subjects have little surplus capacity to consider the veracity of statements, and therefore may also be more likely to take statements as true.

Limited to a paper and pencil task, we chose to use the interruption manipulation, and therefore in half the situations subjects were instructed to read some advertisements rapidly and then move on to the next page. The rest of the time, we allowed subjects more time and had them process claims more deeply. This was done by asking them to consider the veracity of claims and giving them sufficient time to do it. This manipulation also corresponds well to the kinds of processing consumers bring to understanding print magazine advertisements. Sometimes consumers "flip" through magazine advertisements with little thought other than just scanning the headlines and pictures. Other times consumers read advertisements thoughtfully and consider the claims deeply.

FIGURE 1

However, there is a potential problem with this manipulation. Only for the deeper processing manipulation can we measure whether subjects perceive claims to be true or not. Indeed, just by asking for judgement of veracity, the theory suggests one manipulates the subject’s mental processing. To get avoid this, one may follow up the comprehension manipulation with another prompt for whether the claim is true or false. However, this may cue subjects into guessing the purpose of the study, and makes it an unbalanced design because measurement of veracity is part of the deeper processing manipulation.

To overcome this potential problem, we showed subjects all the advertisements twice. For half the situations, we used the comprehension manipulation for the first viewing, followed by the deeper processing one for the second. Alternatively, we used the deeper processing manipulation for the first viewing, followed by the comprehension for the second viewing. Therefore, the effective manipulation was that half the time subjects would do deeper processing initially and we would take a veracity measure immediately. They would later merely comprehend the same advertisement again, but no measure would be taken. The other half of the time subjects would initially merely comprehend followed by deeper processing accompanied by the veracity measure.

TABLE 1

PERCENT CORRECT FOR EACH OF FOUR CONDITIONS

This solution to the potential problem, therefore, reduces the power of the experiment to isolate the effect of mere comprehension. That is, the solution underestimates the size of the effect rather than overstates it. When subjects are presented with the comprehension manipulation first and deeper processing second, in the second deeper processing step they may draw upon more information than just the prior claim exposure. Therefore, these subjects may self-correct for biases introduced in the initial comprehension manipulation. The net effect should be interpreted as a conservative one.

The second manipulation is whether the advertising claims they saw were true or false. Therefore, we included a learning phase to the research such that they learned a number of claims that were to be considered "true." Later in the study they would see correct re-phrasings of these claims half the time, and incorrect the other half. For each of the two viewings of each advertisement, both viewings were either true or false. Each viewingof the same advertisement was spaced apart such that there were two other advertisements between each viewing.

Procedure

Subjects were invited to participate in a two-part study "about the reading and learning of advertisements." The first part was the learning phase, and the second was the testing phase. In phase one, the product, SynFOKtin, was introduced with a brief description. Subjects were told that sixteen statements were verified by a government agency. Subjects then had 45 seconds to learn these statements. Next, they were asked to familiarize themselves with a print advertisement for SynFOKtin with the claim deleted. Once again, the 16 claims were re-introduced, and 45 seconds were given to learn each statement. After this, the sample advertisement was again shown, and then subjects were given the third and final 45 second learning opportunity for the 16 statements.

The reason why subjects were made to learn the claims was to lessen the possibility that subjects were responding to the repetition of the advertisement alone. Also subjects were shown the advertisements with claims deleted was to insure that when they briefly scanned the advertisement they were focusing on the claim only and not the layout of the advertisement.

For phase two, subjects were instructed to consider a number of print advertisements. Subjects would view 16 advertisements twice each. In half the viewing opportunities, subjects would be instructed to "Rapidly read through the claim in the advertisement on the next page as you normally would skim or glance through an advertisement in a magazine while in a waiting room, on public transportation, at an airport, or similar situation." This was the shallow processing level of the comprehension manipulation. For the other viewings, subjects were asked the following question: "Is the claim on the next page true or false? Please check the appropriate level on the bottom of the next page." This was the deeper processing level. Claims in advertisements were always paraphrases of the 16 statements. The order of claim topics was randomized. Note also that half the time the claims read were actually true (regardless of instructions), according to the learning phase of the research. The other half of the time they were actually false. This order was randomized as well.

For example, in phase two, one subject may first see the "Is [it] true or false?" instruction on its own page. The subject would then flip the page and consider a print advertisement with the claim, "SynFOKtin can sometimes reduce mouth saliva." This claim relates to the 15^th statement in the learning phase, "SynFOKtin can bring about excess saliva," so the claim is false. At the bottom of the page there are two levels one may check, either true or false.

The subject then flips the page and sees another instruction, and again, it happens that it is the "Is [it] true or false?" instruction. After flipping the page, the claim "SynFOKtin is useful for patients whose families have had diseases of the left heart ventricle," is substituted for the prior claim about saliva, and this second claim is true based on the phase one material subjects needed to learn. Again, the subject checks either true or false at the bottom of the page. The subject then proceeds to the next page where the "Rapidly read" instruction is seen and flips the page.

This page has another advertisement with another claim, but now the subject does not check anything at the bottom of the page, and only scans the claim and flips to the next page. Here the subject gets the "Rapidly read" instrucion, flips the page, and sees again the same saliva advertisement with the false claim the subject saw several advertisements ago. Because the "Rapidly read" instruction was given, the subject doesn’t check anything, reads rapidly and flips to the next page.

Next is another "Rapidly read" instruction, flips, and then sees the heart disease advertisement seen for the second time. This process continues until 32 advertisements (16 different advertisement X two viewings each) have been considered. Phase two never took more than 5 minutes for any subject.

RESULTS

Table 1 presents the average of correct assessment for the advertisement claims. Consistent with Gilbert and his colleagues’ theory, if subjects merely comprehended the claim (the "rapidly read" condition) before assessment (the "true or false" condition), subjects had a tendency to believe that what they comprehended was true. Thus, when subjects were presented with true claims when deeper thought was truncated by the instruction to move on the next advertisement, the percentage of time they were correct increased 4.2%. When subjects were first presented with false claims when deeper thought was truncated by the manipulation, the percentage of time they were correct dropped 14.7%. This interaction was found to be significant with p=.0004. Therefore, advertising has not shown itself to be an exception to the theory presented by Gilbert and his colleagues.

To further the analysis, a multifacet analysis was used to decompose the total variances into specific components. For example, multifacet analyses often decompose the total variance into that arising from interpersonal differences, experimental treatment differences, and their interactions. Table 2 details the variance components.

TABLE 2

EFFECT SIZES: VARIANCE COMPONENTS AND OMEGA-SQUARED

The variance components associated with the effects of interest are the depth of processing X veracity of claim interaction and the subjects X depth of processing X veracity of claim interaction. Together, these two variance components sum to 4.7% of the total variance, which is a modest effect.

Another way to look at the effect size is by using omega-squared (Green 1973; Peterson, Albaum and Beltramini 1985) and these are also listed in Table 1. The effects of interest are again the depth of processing X veracity of claim interaction and the subjects X depth of processing X veracity of claim interaction. Although these effects are relatively small, they are not outside the typical range for similar laboratory interaction effects tested on college students.

DISCUSSION, LIMITATIONS AND IMPLICATIONS

The significance of Gilbert and his colleagues’ work is that the perception of facts as first true may be a biologically based system that individuals are incapable of changing. It is not that individuals find it difficult to not believe advertising claimsCmerely that it is more difficult to disbelieve than believe. One has freedom to disbelieve, but in situations where one chooses not to evaluate (e.g., low involvement situations, time pressure, etc.), one tends to believe advertising claims whether patently true or not. This has wide-ranging implications for both consumer research and public policy.

These findings are consistent with and add more detail to current persuasion frameworks like the Elaboration Likelihood Model (Petty, Cacioppo and Schumann 1983) or the Heuristic-Systematic Model (Chaiken 1987; Ratneshwar and Chaiken 1991). These models propose the active processing of claims under central or systematic processing. Here, supporting-claims and counter-claims are self-generated and the model is consistent with the evaluations of veracity Gilbert and his colleagues discuss. Likewise, the importance of motivation and ability to process is consistent with Gilbert and his colleagues’ suggestion that one must choose not to believe advertising claims.

However, under peripheral or heuristic processing, advertising claims, subtle as they are via various cues or heuristics, are more likely to be believed because there is reduced processing capacity applied. Possibly this effect explains why persuasion attempts under peripheral processing frequently appear successful via short term attitude measures, but unsuccessful via long term behavioral change. Because of the shallow processing applied in peripheral situations, claims are not well integrated into memory. When tested immediately, recency effects may indicate that peripheral processing had lead to positive attitude changes. If a decision needs to be made later, the persuasiveness of the original claims is dependent on the ability of one to recall the claim. Thus, often repeated claims, even if peripherally encoded and unsupported, may be mistaken for facts if they are accessible enough.

This work also suggests new directions for schema triggered affect theory. Most testing situations provide unlimited cognitive resources for subjects (e.g., Pavelchak 1989). Thus, subjects may always switch to piecemeal processing if they do not find a suitable schema to which to fit the stimulus. It may be that under limited capacity and no suitable schema fit, no evaluation may be made. Or affect evaluations do not affect believability if mental capacity is limited.

Another implication is that suspension of consumer beliefs may be impossible. Under the Spinozan view of information processing, there is no "waiting room" in which claims may rest not affecting other thoughts. Claims are accepted as they are, or reprocessed as true or false. This approach also changes the nature of skepticism. Under the "waiting room" approach, skepticism was being rather stingy about what one lets out of the "waiting room." However, under the Spinozan approach, skepticism may merely be the desire to challenge a claim. This latter approach mirrors Kunda’s (1990) notion of motivated reasoning. If one is motivated to challenge a claim, one generates counter-arguments. If a counter-argument succeeds in discounting a claim, then it taken to be false. If appropriate counter-arguments are not generated, the claim might be tentatively believed. Therefore, the lack of a mental "waiting room" may lead to honesty paradoxes in which advertisers have incentives to misrepresent the truth (see Koslow 2000).

This research also has implications for the Persuasion Knowledge Model (Friestad and Wright 1994). Target consumers use topic and agent knowledge to understand and evaluate advertising claims, and consumer skepticism (i.e., consumer questioning of advertising claims) should be seen as an application of target topic knowledge. However, these results suggest that persuasion knowledge may not normally be applied under limited cognitive processing. In addition, the goal of holding valid topic attitudes may only apply in cases with deeper processing.

There are also implications for further study of Hawkins and Hoch (1992) finding that repetition enhances believability. It may be that mere comprehension leads to believability. However, such low level learning may not be retained withot repetition. Further research is warranted.

Two limitations are in order. First, because the product category was pharmaceuticals, which is regulated by the government, it may have been easier to believe claims because consumers were not as on guard. Second, the repetitive within subject design may have produced a demand effect. Although no one guess the subject of the research and those who partially did were eliminated, it would be worthwhile to replicate these findings with a between subject design.

These findings also have implications for public policy. Despite numerous criticisms of continuing consumer skepticism toward the overall institution of advertising, it seems that in contrast to such a macro-perspective consumers’ skepticism declines in a micro-context. Our results support the conclusion that consumers process the vast majority of advertising claims information by rapidly comprehending it and accepting it as true. And, unless consumers go to the next step to assess the veracity of these claims, they remain labeled as true until proven false. In a marketplace situation, therefore, such pre-existing beliefs are pulled off the "library shelves" of the mind for reference, still so labeled regardless of the truthfulness of the information. Thus, in protecting consumers from false or misleading advertising claims, it is essential for policy makers to go beyond exposure to assess consumer comprehensability of the claims.

Under typical, low involvement, casual reading conditions (shallow processing), the initial acceptance of advertising information appears to represent the appropriate focus for regulatory efforts. Subsequent regulatory "fixes" which later seek to undo previously accepted claims by advising consumers of the falsity of the prior information appear far less potent. That is, the damage to consumers is already done and it may be particularly difficult to undo that damage. Thus regulators should be vigilant and not assume that consumer skepticism will protect consumers adequately.

REFERENCES

Beltramini, Richard F. and Edwin Stafford (1993), "Comprehension and Perceived Believability of Seals of Approval in Advertising, Journal of Advertising, 22 (3), 3-13.

Bettman, James R. (1979), An Information Processing Theory of Consumer Choice, Reading, MA: Addison-Wesley.

Brucks, Merrie, Andrew A. Mitchell and Richard Staelin (1984), "The Effect of Nutritional Information Disclosure in Advertising: An Information Processing Approach," Journal of Public Policy and Marketing, 3, 1-25.

Brucks, Merrie, Gary M. Armstrong, and Marvin E. Goldberg (1988), "Children’s Use of Cognitive Defenses Against Television Advertising: A Cognitive Response Approach," Journal of Consumer Research, 14 (March), 471-482.

Burke, Raymond R., Wayne S. DeSarbo, Richard L. Oliver, and Thomas S. Robertson (1988), "Deception by Implication: An Experimental Investigation," Journal of Consumer Research, 14 (March), 483-494.

Calfee, John E. and Debra Jones Ringold(1988), "Consumer Skepticism and Advertising: What Do the Polls Show?" in Advances in Consumer Research, Vol. 15, ed. Michael J. Houston, Provo, UT: Association for Consumer Research, 244-248.

Calfee, John E. and Debra Jones Ringold (1994), "The 70% Majority: Enduring Consumer Beliefs About Advertising," Journal of Public Policy and Marketing, 13, 228-238.

Chaiken, Shelly (1987), "The Heuristic Mdel of Persuasion," in Social Influence: The Ontario Symposium, Vol. 5, ed. M. P. Zanna, J. M. Olson, and C.P. Herman, Hillsdale, NJ: Erlbaum, 3-39.

Cofer, Charles N. (1979), "Human Learning and Memory," The First Century of Experimental Psychology, ed. Eliot Hearst, Hillsdale, NJ: Lawrence Erlbaum Associates, Publishers, 323-370.

Finn, David W. (1984), "The Integrated Information Response Model," Journal of Advertising, 13, 24-33.

Fiske, Susan T. and Mark A. Pavelchak (1986), "Category-Based Versus Piecemeal-Based Affective Responses: Developments in Schema-Triggered Affect," The Handbook of Motivation and Cognition: Foundations of Social Behavior, eds. R. M. Sorrentino and E. T. Higgins, New York: Guilford Press, 167-203.

Ford, Gary T., Darlene B. Smith and John L. Swasy (1990), "Consumer Skepticism of Advertising Claims: Testing Hypotheses from Economics of Information," Journal of Consumer Research, 16 (March), 433-441.

Friestad, Marian and Peter Wright (1994), "The Persuasion Knowledge Model: How People Cope with Persuasion Attempts," Journal of Consumer Research, 21 (June), 1-31.

Gaeth, Gary J. and Timothy B. Heath (1987), "The Cognitive Processing of Misleading Advertising in Young and Old Adults: Assessment and Training," Journal of Consumer Research, 14 (June), 43-54.

Gilbert, Daniel T.(1991), "How Mental Systems Believe," American Psychologist, 46 (2) 107-119.

Gilbert, Daniel T., D. S. Krull and Patrick S. Malone (1990), "Unbelieving the Unbelievable: Some Problems in the Rejection of False Information," Journal of Personality and Social Psychology, 59, 601-613.

Gilbert, Daniel T., Romin W. Tafarodi, and Patrick S. Malone (1993), "You Can’t Not Believe Everything You Read," Journal of Personality and Social Psychology, 65 (August), 221-233.

Goldberg, Marvin E. and Jon Hartwick (1990), "The Effects of Advertiser Reputation and Extremity of Advertising Claim on Advertising Effectiveness," Journal of Consumer Research, 17 (September), 172-179.

Green, Paul (1973), "On the Analysis of Interactions in Marketing Research Data," Journal of Marketing Research, 10 (November), 410-420.

Hawkins, Scott A. and Stephen J. Hoch (1992), "Low-Involvement Learning: Memory without Evaluation," Journal of Consumer Research, 19 (September) 212-225.

Kirmani, Amna (1990), "The Effect of Perceived Advertising Costs on Brand Perceptions," Journal of Consumer Research, 17 (September), 160-171.

Koslow, Scott (2000), "Can the Truth Hurt? How Honest and Persuasive Advertising Can Unintentionally Lead to Increased Consumer Skepticism," Journal of Consumer Affairs, 34 (Winter), 245-268.

Kunda, Ziva (1990), "The Case for Motivated Reasoning," Psychological Bulletin, 108 (3), 480-498.

Margolis, Howard (1987), Patterns, Thinking, and Cognition: A Theory of Judgement, Chicago: University of Chicago Press.

Meyer, Robert J. (1982), "A Descriptive Model of Consumer Information Search Behavior," Marketing Science, 1, 93-121.

Miller, Gerald R. and James B. Stiff (1993), Deceptive Communication, Newbury Park, CA: Sage Publications.

Mitchell, Andrew A. (1983), "Cognitive Processes Initiated by Exposure to Advertising," Information Processing Research in Advertising, e. Richard Jackson Harris, Hillsdale, NJ: Lawrence Erlbaum Associates, 13-42.

Mittal, Banwari (1994), "Public Assessment of TV Advertising: Faint Praise and Harsh Criticism," Journal of Advertising Research, 34 (January/February), 35-53.

Moore, James C. and Andrew B. Whinston (1986), "A Model of Decision-Making with Sequential Information-Acquisition," Decision Support Systems, 2, 285-307.

Moriarity, Sandra E. (1983), "Beyond the Hierarchy of Effects: A Conceptual Framework," Current Issues and Research in Advertising, 6 (1), 45-46.

Pavekchak, Mark A. (1989), "Piecemeal and Category-Based Evaluation: An Idiographic Analysis," Journal of Personality and Social Psychology, 56 (3), 354-63.

Peterson, Robert A., Gerald Albaum and Richard F. Beltramini (1985), "A Meta-Analysis of Effect Sizes in Consumer Behavior Experiments," Journal of Consumer Research, 12 (June), 97-103.

Petty, Richard E., John T. Cacioppo and David Schumann (1983), "Central and Peripheral Routes to Advertising Effectiveness: The Moderating Role of Involvement," Journal of Consumer Research, 10 (September), 135-146.

Preston, Ivan L. (1994), The Tangled Web They Weave, Madison, WI: University of Wisconsin Press.

Rao, Akshay and Wanda A. Sieben (1992), "The Effect of Prior Knowledge on Price Acceptability and the Type of Information Examined," Journal of Consumer Research, 19, 256-270.

Ratneshwar, S. and Shelly Chaiken (1991), "Comprehension’s Role in Persuasion: The Case of Its Moderating Effect on the Persuasive Impact of Source Cues," Journal of Consumer Research, 18 (June), 52-62.

Simonson, Itamar, Joel Huber and John Payne (1985), "The Relationship Between Prior Brand Knowledge and Information Acquisition Order," Journal of Consumer Research, 14, 566-578.

Tybout, Alice M., Bobby J. Calder and Brian Sternthal (1981), "Using Information Processing Theory to Design Marketing Strategies," Journal of Marketing Research, 18 (1), 73-79.

Urbany, Joel E., William O. Bearden and Dan C. Weilbaker (1988), "The Effect of Plausible and Exaggerated Reference Price on Consumer Perceptions and Price Search," Journal of Consumer Research, 15 (June), 95-110.

Wright, Peter K. (1983), "The Cognitive Processes Mediating Acceptance of Advertising," Journal of Marketing Research, 10, 53-62.

Zimbardo, P. G. and M. R. Lieppe (1991), The Psychology of Attitude Change and Social Influcence, New York: McGraw-Hill.

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