Mediating Effects of Cognitive Responses to Advertising on Cognitive Structure

Jerry C. Olson, The Pennsylvania State University
Daniel R. Toy, University of Washington
Philip A. Dover, Dartmouth College
ABSTRACT - This paper reports results of an initial study of the mediating effects of cognitive responses to advertisements on selected elements of cognitive structure. To create variation in cognitive response, consumers were shown one of several new product ads that varied in terms of product price and the presence or absence of information regarding additional (non-price) attributes of the advertised product. Type of cognitive response was sensitive to both aspects of ad content. As expected, counterargument and support argument responses were both strongly related to several cognitive structure elements.
[ to cite ]:
Jerry C. Olson, Daniel R. Toy, and Philip A. Dover (1978) ,"Mediating Effects of Cognitive Responses to Advertising on Cognitive Structure", in NA - Advances in Consumer Research Volume 05, eds. Kent Hunt, Ann Abor, MI : Association for Consumer Research, Pages: 72-78.

Advances in Consumer Research Volume 5, 1978      Pages 72-78


Jerry C. Olson, The Pennsylvania State University

Daniel R. Toy, University of Washington

Philip A. Dover, Dartmouth College

[This research was supported by a grant to the senior author from the Center for Research, College of Business Administration, The Pennsylvania State University.]

[Associate Professor, Assistant Professor and Assistant Professor, respectively. This research was conducted while the second and third authors were doctoral students in Marketing at the Pennsylvania State University.]


This paper reports results of an initial study of the mediating effects of cognitive responses to advertisements on selected elements of cognitive structure. To create variation in cognitive response, consumers were shown one of several new product ads that varied in terms of product price and the presence or absence of information regarding additional (non-price) attributes of the advertised product. Type of cognitive response was sensitive to both aspects of ad content. As expected, counterargument and support argument responses were both strongly related to several cognitive structure elements.


The measurement and explanation of persuasive communications impact has occupied the attention of a wide variety of academic, industry, and public policy researchers. In literally hundreds of studies attempts have been made to model the effects of the persuasive communication process and its subsequent influence on consumer behavior. The importance of this issue for marketing and public policy is evidenced by the more than $28 billion spent in the United States on advertising in 1976.

The lack of a clear understanding of the communication process may be attributed to several factors, the broadest and most critical of which may be the absence of a vigorous, process-oriented, empirical research tradition. The dominant research paradigm for communication research involves the measurement of some dependent variable (attitudes, opinions, sales) following exposure to a persuasive communication. The independent variables typically manipulated in these studies are usually concerned with characteristics of the source or channel of the message, the message itself, or the receiver of the message (Mitchell & Olson, 1977). The purpose of such studies is to identify generalizable relationships between the presumed goal of the persuasive communication (e.g., attitude change) and the communication variable of interest. For the most part, these simplistic attempts have not been successful (cf. Fishbein & Ajzen, 1975).

In contrast to this generally atheoretical approach to studying communication effects are two streams of research based upon attitudinal and information processing perspectives. Both focus on the cognitive states and/or processes that intervene between or mediate exposure to persuasive communications and changes in attitude, behavioral intention, or overt behavior. These two approaches, broadly termed the cognitive structure and cognitive response models, provide the theoretical foundations for this study.

The Cognitive Structure Model

The cognitive structure model has its theoretical roots in learning theory and emphasizes beliefs as the fundamental cognitive element (see Lutz & Swasy, 1977; Olson & Mitchell, 1975). For example, Fishbein proposed expectancy-value models that depict the causal relationships between beliefs and attitudes, intentions, and eventually behavior (cf. Fishbein & Ajzen, 1975). Specifically, attitudes are thought to be based on beliefs (Fishbein, 1967):



A = attitude toward (evaluation of) an object, action or event.

bi = strength of belief about the "ith" attribute or consequence.

ei = evaluation of the "ith" attribute or consequence.

n = number of salient beliefs.

By this model, Fishbein proposed that attitudes are a function of belief strength and the evaluative aspects of beliefs combined in an additive, compensatory manner. An extension of this attitude model (see Fishbein & Ajzen, 1975) relates the attitude concept to behavioral intentions (BI) which, in turn, are causally related to behavior (B) itself, as indicated by arrows in the following simplified schema:


Fishbein's model of attitude formation implies that a persuasive communication will affect attitudes indirectly via the processing of information contained in the communication which first creates bi and ei cognitive elements (see Fishbein & Ajzen, 1975). This approach to monitoring communication effects represents a step beyond the traditional research paradigm by actually measuring the cognitive variables (beliefs) that presumably mediate and determine attitude change and is becoming increasingly evident in the consumer behavior literature (cf. Kuehl & Dyer, 1977; Lutz, 1975; Mazis & Adkinson, 1976; Olson & Dover, 1976, in press). Beliefs formed by processing the information contained in the persuasive communication presumably are integrated into a pre-existing belief framework. According to Fishbein and Ajzen (1975), the resulting change in overall belief structure due to the integration of previous beliefs and beliefs derived from the communication information is the fundamental basis for attitude change and behavior modification.

Cognitive Response Models

An alternative conceptual view of communication impact is provided by cognitive response models which were developed explicitly for monitoring communication effects. The basic premise of these models is that the spontaneous unstructured cognitive responses (thoughts) elicited by a communication message act as mediators of attitude formation or change. In this view, a precise understanding of prediction of communication impact is not possible without attention to the cognitive thought responses emitted by the receiver upon exposure to the communication.

The measurement paradigm for a cognitive response study typically includes the following procedures (see Wright, 1974, for a more complete discussion). First, the subject is asked to either write down or verbalize all of his/her thoughts either during or immediately after exposure to a persuasive communication. These protocols are assumed to indicate the internal subvocal cognitive responses which occurred during and after exposure to the communication. Then, the cognitive responses are categorized according to various criteria, either by the subject or by the experimenter. Possible criteria include whether there is a "critical" or "favorable" orientation to the message, the perceived origin of the thought, and whether or not the response meets the prescribed definition for a class of cognitive responses. Examples of cognitive response categories include counterargument, source derogation, support arguments, and curiosity statements (see Wright, 1973, 1974). The third step is to compute cognitive response indices or scores based upon the categorization. These scores may represent either a simple addition of the number of counterarguments, source derogations, or support arguments, or more elaborately, use of a model in which each cognitive response type is weighted by subjective indications of importance (e.g., Wright, 1973). In both cases the scores represent an operationalization of the cognitive response mediators and are typically used to predict some criterion, usually attitude or intention.

The application of direct thought monitoring in communication research is in its infancy. Only a few studies, mostly in the social psychology literature, have used this conceptual and methodological orientation. In these investigations, the research emphasis has centered on the study of distraction effects (Osterhouse and Brock, 1970; Keating and Brock, 1974; Insko, Turnbull, and Yandell, 1974; Petty, 1977; Petty, Wells, and Brock, 1976), attitude change/persuasive impact (Cook, 1969, Wright, 1973, Petty, 1977), product versus task involvement (Krugman, 1967; Wright, 1973) repetition and media modality effects (Ray, Sawyer, and Strong, 1971; Wright, 1973; Ray, Ward, and Reed, 1976; Ray and Webb, 1976), source credibility (Cook, 1969) and fear arousing messages (Janis and Terwilliger, 1962). Wright (in press) has reviewed the current literature on cognitive response models. The potential of the cognitive response/thought monitoring approach to advertising research, although largely unexplored, appears strong and warrants the attention of consumer behaviorists.

A Combined Cognitive Structure/Cognitive Response Model

Both of the previously mentioned research orientations possess advantages and disadvantages as a modeling perspective for studying the impact of persuasive communications. However, if the two approaches are combined, it may be shown that each complements the other's weak points. Whereas the cognitive structure model focuses on the structural aspects of stored knowledge in the form of beliefs, the cognitive response model is more concerned with cognitive processing activity. Together, the two models can provide a relatively comprehensive model of communication impact. That is, cognitive responses (subvocal thoughts) can be seen as directly mediating the formation of new beliefs or changes in pre-existing beliefs. The belief structure so created would, in turn, mediate attitude formation and change. The advantages of, and potential for, this joint model approach were independently recognized and clearly presented in a particularly thorough paper by Lutz and Swasy (1977).

The present paper represents an initial attempt to examine the conceptual and predictive utility of the cognitive response perspective as a mediating process occurring between exposure to a communication and the formation of or change in cognitive structure variables such as beliefs, attitudes and intentions. Due to space limitations, emphasis is on the more general and heuristic effects of cognitive responses on cognitive structure, as a function of exposure to advertising communications. Our long-range goal is to encourage and contribute to the development of a more precise and complete theory of communication effectiveness, at an individual level.


Research Design

To more clearly identify the mediating effects of cognitive responses, it was necessary to experimentally create alternative communications (in the form of simple, poster-like advertisements) that should lead to variations in magnitude and type of cognitive response. These differences, in turn, would be expected to have a variable impact on cognitive structure variables. Two factors were selected for manipulation, a product attribute and an ad content variable. Ads were created which communicated one of five arbitrary price levels for the ball point pen product--294, 494, 794, $1.49, and $3.95. Additionally, information regarding other salient (non-price) attributes of the pen was either present or absent. Thus the design was a 5 x 2, between-groups factorial involving 5 levels of price and the presence or absence of additional, non-price information about the product (see Table 1).



Basically, consumer subjects were exposed to one of the 10 advertisements (treatment combinations). Then their cognitive responses (thoughts) to the ad were recorded, followed by an extensive measurement of cognitive structure elements.

Although highly specific hypotheses were not considered possible given the current state of theory, we intuitively anticipated greater counterargument for higher prices. However, it was not clear a priori whether additional non-price information would lead to greater or less counterargument compared to price-only conditions. Moreover, we generally expected counterargument and support argument to mediate the formation of less positive and more positive cognitive elements, respectively.

Product and Subjects

Several factors influenced the choice of ball point ink as the focal product in this research. First, pilot interviews indicated a relatively simple cognitive structure for ball point pens in terms of salient product attributes. Thus, measurement complexity was kept within reasonable limits. Second, a relatively new, inexpensive pen possessing several unusual features had been recently introduced in the local market and was still unfamiliar to nearly all subjects. The actual brand name of this pen was used throughout the study, but is omitted in this paper.

Because ball point pens have a high usage rate among college students, a student population was considered appropriate for this research. Subjects were 165 undergraduate students from introductory business courses who were paid $2.00 for their voluntary one-hour participation.


In all, ten separate ads were professionally created to correspond to the ten treatment combinations in the research design (see Table 1). Centered at the top of each in large type was the brand name followed by the headline, "Introducing the New Ball Point Pen." In the center of each ad was a black and white drawing of the pen.

The experimental manipulations of price level and other information were accomplished by varying the remaining information in the ads. The price level was presented in the lower right corner of each ad -- e.g., "only 294," or "only $3.95.'' [The actual price of the pen was 794 in most stores.] In the price only conditions, no information other than price was given. In the "additional information" conditions, four other product features were listed in the lower right corner and very briefly described. These included (a) the physical construction of the ball point tip, (b) the tungsten carbide ball and thin, free flowing ink, (c) the attractive design, and (d) the comfortableness of using the pen.


Subjects were run in groups corresponding to each treatment combination. The procedure by which subjects signed up to participate at a particular time was considered to be essentially equivalent to a formal randomization procedure.

Upon entering the research room, subjects were told in general terms that the study concerned their reactions to a new ball point pen and some promotional material for the product. Then, to meet University regulations, they signed an informed consent form certifying their voluntary decision to participate. Following this, subjects completed a two-page questionnaire regarding general knowledge of and purchase behavior towards ball point pens and the perceived importance of several pen attributes.

Next, subjects were told that a rough mock-up of a forthcoming ad for the new ball point pen had been obtained. This ad was described as one that might be used as a point-of-purchase poster, perhaps on a store counter-top. Each subject was then given a black and white copy of the ad to examine. After two minutes, subjects were instructed to turn the ad face down on the table and to write (on a blank sheet of paper before them) "all the things that came to mind as you read the ad." They were given exactly two minutes to record their cognitive responses. Following, subjects turned to the next page and began responding to the post-exposure questionnaire. To separate somewhat the measurement of cognitive response and cognitive structure, the first page of the questionnaire contained six structured and two open-ended questions regarding subjects' general reactions to the ad itself. Following this were questions measuring bi, ei, Ao, Aact and BI elements of cognitive structure regarding the pen.

After completion of the questionnaire, the study continued with an actual trial experience with the pen and subsequent re-measurement of the cognitive structure variables (this post-trial data is not relevant for the present paper). At the conclusion of the study, subjects were debriefed (the deceptions used were thoroughly explained and justifications were given) and were paid and dismissed.

Dependent Variables

The elements of cognitive structure measured in this paper are those identified most closely with structural models of attitudes (cf. Fishbein and Ajzen, 1975) and are represented in terms of their general causal relationships in (2).

The measurements of the belief strength (bi) and evaluative aspects (ei) were based on the vector model and methodological procedures proposed by Ahtola (1975). A major contribution of Ahtola's modification of the familiar Fishbein expectancy-value model (1) is the emphasis on a more specific and precise identification of the belief concepts. In the Ahtola view, a product attribute is conceptualized as a dimension containing several discriminable levels or amounts of the attribute. In perceiving or encoding information about a product attribute, a consumer essentially assigns the product to one or more of the discriminable categories along that attribute dimension. For example, a wine could be classified as sweet, slightly sweet, medium dry, dry, or very dry along the sweet-dry attribute dimension. From this perspective, each category or level represents a discriminable belief. If, for example, one is positively certain that a particular wine is medium dry, then the belief strength (bi) for that category should be maximal and the bi's for the other categories should be zero. This basic notion suggests that people learn these categories (cf. Wyer, 1974) and thus cognitive structure research should focus on identifying the discriminable categories used by subjects to encode the meaning of a particular stimulus such as a brand.

Through pilot interviews with several undergraduate students, seven modally salient attributes for ball point pens were identified: price, ink flow consistency, comfort, appearance, ease of writing, writing quality, and value for the money. Discriminable categories (levels or amounts) for each attribute were developed based partially on experimenter intuition. For the categories along each attribute (e.g., very good, good, average, poor, and very poor value for the money), subjects indicated their belief strength by assigning 10 points to the several levels or categories such that the number of points indicated the strength of belief that the pen possessed each attribute level. Subjects also rated their evaluation of each attribute level.

An index of the overall evaluation associated with each attribute can be obtained by summing the biei products over the category levels to yield a vector score of evaluation (Sbiei). An index of overall evaluation associated with the entire cognitive structure is represented by summing the separate vector/attribute scores (SSbiei). Attributes toward the pen (Ao) and the act of buying the pen (Aact) were each measured by averaging the scores on the three 5-point scales following Fishbein's recommendations (1967). Intention to purchase for one's own personal use on the next shopping trip (BI) was measured on a 5-point scale (not likely--very likely to buy, 1-5, respectively).

Classifying the Cognitive Response Measures

The measures of subjects' cognitive responses to the various ads were classified by following closely the criteria proposed by Wright (1973). The two cognitive responses of major interest were counterargument and support argument. Briefly, counterargument is indicated by statements that are directed against (disagree or clash with) the product in general, its use, or a specific claim or idea contained in the ad. Support arguments are indicated by thoughts in favor of (agree with) the product, its use, or a specific claim or idea in the ad. Several other types of cognitive response were also identified, including: (a) curiosity statements -- general thoughts evidencing curiosity regarding product characteristics, (b) positive ad-related statements -- thoughts about positive aspects of the ad itself, and (c) negative ad-related statements.


Treatment Effects on Cognitive Responses

Although this paper is primarily concerned with support and counterarguments, Table 2 presents, for the reader's interest, a summary of all the types of cognitive responses elicited by the two types of advertised information (collapsed across all price level conditions) and for the entire design. The overall average of 2.94 cognitive responses per subject is somewhat lower than that typically obtained in other studies (e.g., 3.8 to 10 or more). However, most of these other studies used much more complex communications with rather extensive copy thus providing more material to cognitively respond to. Moreover, most of these studies did not place an explicit time limit on the thought monitoring task and subjects may have generated items in the course of thought recording (cf. Wright, 1977a, 1977b, for excellent reviews and analyses of the methodology used in this research).



Note in Table 2 that curiosity thoughts dominate in the price-only ads where virtually no information other than price was available. In contrast, for the conditions involving information about price and other attributes, counter and support arguments were the most frequently occurring cognitive responses. Because these types of responses are thought to have the major mediating influence on cognitive structure, the remainder of the paper focuses only on support and counterargument effects.

Table 3 summarizes the effects of the experimental manipulations on the number of counterarguments and support arguments elicited by ad exposure. As expected, Price Level had a significant main effect on the number of both counterarguments and support arguments. Figure 1 illustrates this main effect -- i.e., the mean number of counterarguments as a function of Price. A Newman-Kuels analysis revealed that the 29? and $3.95 prices produced significantly fewer and more counterarguments (p's < .05) than the three intermediate prices, which did not differ from one another.

The experimental treatment effects on number of support arguments are less clear-cut. Although a main effect of Price Level was obtained (graphed in Figure 1), Type of Information was found to interact with Price. This two-way interaction is not readily interpretable and, therefore, is not illustrated. Finally, the significant main effect of Type of Information was such that more support arguments were elicited by price-plus-other-attribute information (X = .75) than price-only information (X = .46, p< .05).





Given the above evidence that the experimental treatments did cause differences in type and level of cognitive response, our attention turns to the mediating effects of counter and support arguments on elements of cognitive structure. To keep this paper within its space limitations, it was necessary to examine only selected cognitive elements. In other papers, we will tackle the complex analyses involved in studying cognitive response effects on individual belief elements. In the present paper we will concentrate on the more global cognitive elements such as attitudes and intentions.

Cognitive Response Effects on Cognitive Structure

To determine the relationship between counter and support arguments and these selected cognitive structure elements, two analyses were performed. The first involved collapsing the price level factor and computing correlations between number of counter and support arguments and selected cognitive variables for the two information type conditions. Table 4 summarizes the results of this analysis. As expected, counterargument was negatively related to cognitive structure variables, while support argument was positively related to cognitive structure elements.



Both counterargument and support argument were significantly related to the belief vector regarding value for the money (Sbiei), the overall cognitive structure index (SSbiei), and intentions to purchase. The relationships with attitude (Ao and Aact) were generally significant but somewhat weaker. Moreover, the relationships between cognitive response and structural variables are generally stronger and more consistent for the condition involving information about price and other attributes than for the price-only ads. When the information-type factor is collapsed, both counter and support argument responses are significantly related to all the more global elements of cognitive structure, although none of the relationships is particularly strong (r's from .16 to .33, p's < .05).

An alternative and perhaps more useful perspective on cognitive response influence is gained by examining the magnitude of the mediating effect on selected cognitive variables. This analysis was accomplished by collapsing the price level factor and dividing the subjects in each information type group into those making no counterarguments (CA = 0) vs. those with some (CA > 1) counterarguments (X = 1.47). In effect, this internal analysis created a 2 x 2 design composed of two levels of information type and two levels of counterargument magnitude (some and none). The cell means and ANOVA results for several cognitive variables are presented in Table 5. Scores for Aact are not shown because no effects were found for this variable. Note that Counterargument was consistently and significantly related to the value-for-the-money belief vector, and to belief structure, product attitude, and purchase intention. Information Type had generally weaker main effects. Only one (relatively weak) interaction was obtained. In all cases, the effect of counterargument was to lower the positiveness (increase the negativity) or weaken the cognitive variable.

The same internal analysis was conducted by splitting the sample based on those subjects who reported no support arguments (n=98) and those who reported one or more support arguments (n=67). When this blocking factor was combined with the information-type treatment conditions, a 2 x 2 internal analysis was again conducted on the selected cognitive variables of interest. The results are presented in Table 6. Support Argument was significantly related (main effects) to each of the cognitive elements, such that the variable was more positive or stronger when subjects engaged in support arguing. Information Type had some main effects on the belief and attitude variables. No interaction effects were obtained.


The results of this study provide support for the notion that cognitive responses to persuasive communications (ads in this case) mediate the effect of the message on elements of cognitive structure (Lutz and Swasy, 1977; Wright, 1973; 1974). Both counterarguments and support arguments were found to be related to a wide range of cognitive variables including beliefs, attitudes, and purchase intentions. Although strict cause and effect statements cannot be supported by this design, it seems logically reasonable to presume that cognitive responses precede and thus influence the formation of cognitive structure elements. The direction of influence of the types of cognitive responses was, of course, opposite, with counterarguments leading to less positive or weaker cognitive states and support arguments leading to more positive or stronger cognitive states. The broad consistency of these effects across different analyses and over a variety of cognitive measures, lends additional credence to the basic conceptual ideas about cognitive responses and to their construct and measurement validity. In sum, cognitive response theory appears to have considerable utility for improving our generally weak theories of the persuasive communications process, and therefore warrants specific attention in future advertising/communication research.

The relatively strong and consistent mediating effects of support and counterargument responses as evidenced in this study may be seen as even more robust when contrasted with earlier research. The bulk of the past research has used rather strong, overtly persuasive communications--ones quite likely to maximize cognitive responses, especially support and counterarguments (Wright, 1977a, 1977b, in press). In contrast, the present study examined ads which were primarily informational in "flavor" rather than distinctly persuasive. This was particularly true of the price-only ads. One point is that consumers may counterargue and/or support argue





with message content that does not directly attack established beliefs. Another is that such processes may occur and mediate impact on cognitive structure even for seemingly unimportant, low involvement products such as ink pens. A third and very important applied application of the present results is that specific aspects of ad content (such as the type of information conveyed) may have a fairly strong effect on the cognitive responses produced and, in turn, on subsequent effects on cognitive structure. This latter idea suggests that advertising research could focus on determining specific ad content variables that would minimize counterarguments (or maximize support argument), thus "paving the way" for cognitive structure effects favorable to the firm.

In conclusion, it would seem that cognitive response theory and methodology warrant the attention of consumer researchers. When combined with other theoretical ideas regarding information processing, this approach may offer new insights into the communication process. Moreover, the entire research area holds promise of applicability to real world communication problems.


Olli T. Ahtola, "The Vector Model of Preferences: An Alternative to the Fishbein Model," Journal of Marketing Research, 12(February, 1975), 52-59.

Bobby J. Calder, Chester A. Insko, and Ben Yandell, "The Relation of Cognitive and Memorial Processes to Persuasion in a Simulated Jury Trial," Journal of Applied Social Psychology, 4(1974), 62-93.

Thomas D. Cook, "Competence, Counterarguing, and Attitude Change," Journal of Personality, 37(1969), 342-358.

Martin Fishbein, "A Consideration of Beliefs and Their Role in Attitude Measurement," in M. Fishbein (ed.), Readings in Attitude Theory and Measurement, (New York: Wiley, 1967), 257-266.

Martin Fishbein and Icek Ajzen, Belief Attitude, Intention and Behavior, (Reading, Massachusetts: Addison-Wesley, 1975).

Chester A. Insko, William Turnbull, and Ben Yandell, "Facilitative and Inhibiting Effects of Distraction on Attitude Change," Sociometry, 37(1974), 508-528.

Irving L. Janis and Robert F. Terwilliger, "An Experimental Study of Psychological Resistances to Fear Arousing Communications," Journal of Abnormal and Social Psychology, 65(1962), 403-410.

Herbert E. Krugman, "The Measurement of Advertising Involvement," Public Opinion Quarterly, 30(1967), 583-596.

Philip J. Kuehl and Robert F. Dyer, "Application of the 'Normative Belief' Technique for Measuring the Effectiveness of Deception and Corrective Advertisements," in W. Perreault, Jr. (ed.), Advances in Consumer Research: Vol. IV, (Atlanta: Association for Consumer Research, 1977), 204-212.

Richard J. Lutz, "Changing Brand Attitudes Through Modification of Cognitive Structure," Journal of Consumer Research, 4(March, 1975), 49-59.

Richard J. Lutz and John L. Swasy, "Integrating Cognitive Structure and Cognitive Response Approaches to Monitoring Communications Effects," in W. Perreault, Jr. (ed.), Advances in Consumer Research: Vol. IV, (Atlanta: Association for Consumer Research, 1977), 363-371.

Michael B. Mazis and Janice E. Adkinson, "An Experimental Evaluation of a Proposed Corrective Advertising Remedy," Journal of Marketing Research, 13(May, 1976), 178-183.

Andrew A. Mitchell and Jerry C. Olson, "Cognitive Effects of Advertising Repetition," in W. Perreault, Jr. (ed.), Advances in Consumer Research: Vol. IV, (Atlanta: Association for Consumer Research, 1977), 213-220.

Jerry C. Olson and Philip Dover, "Effects of Expectation Creation and Disconfirmation on Belief Elements of Cognitive Structure," in B. B. Anderson (ed.), Advances in Consumer Research: Vol. III, (Cincinnati: Association for Consumer Research, 1976), 168-175.

Jerry C. Olson and Philip A. Dover, "Cognitive Effects of Deceptive Advertising," Journal of Marketing Research, in press.

Jerry C. Olson and Andrew A. Mitchell, "The Process of Attitude Acquisition: The Value of a Developmental Approach to Consumer Attitude Research," in M. J. Schlinger (ed.) J, Advances in Consumer Research: Vol. II, (Chicago: Association for Consumer Research, 1975), 249-264.

Robert A. Osterhouse and Timothy C. Brock, "Distraction Increases Yielding to Propaganda by Inhibiting Counterarguing," Journal of Personality and Social Psychology, 15(1970), 344-358.

Richard E. Petty, "The Importance of Cognitive Responses in Persuasion," in W. Perreault, Jr. (ed.), Advances in Consumer Research: Vol. IV, (Atlanta: Association for Consumer Research, 1977), 358-362.

Richard E. Petty, Gary L. Wells, and Timothy C. Brock, "Distraction Can Enhance or Reduce Yielding to Propaganda: Thought Disruption Versus Effort Justification," Journal of Personality and Social Psychology, 34(September, 1976), 874-884.

Michael L. Ray, Alan G. Sawyer, and Edward C. Strong, Frequency Effects Revisited," Journal of Advertising Research, 11(February, 1971), 14-20.

Michael L. Ray, Scott Ward, and Jerome B. Reed, "Pretesting of Anti-drug Abuse Education and Information Campaigns: Summary Report of a Marketing Science Institute Special Project," in R. Ostman (ed.), Communication Research and Drug Education, (Beverly Hills, California: Sage Publications, 1976).

Michael L. Ray and Peter Webb, "Experimental Research on the Effects of TV Clutter: Dealing With the Difficult Media Environment," Report No. 76-102, Marketing Science Institute, Cambridge, Massachusetts, (April, 1976).

Peter L. Wright, "The Cognitive Processes Mediating Acceptance of Advertising," Journal of Marketing Research, 10(February, 1973), 53-62.

Peter L. Wright, "On the Direct Monitoring of Cognitive Response to Advertising," in G. D. Hughes and M. L. Ray (eds.), Buyer/Consumer Information Processing, (Chapel Hill: University of North Carolina Press, 1974), 220-248.

Peter L. Wright, "Research on Ad-Stimulated Thought Processes,'' Unpublished working paper, Graduate School of Management, Stanford University, February, 1977(a).

Peter L. Wright, "Speculations and Evidence About Audience Responses to Mass Media Advertisements," a paper presented at the conference, "The Effect of Information on Consumer and Market Behavior," Carnegie-Mellon University, May 20 and 21, 1977(b).

Peter L. Wright, "Cognitive Responses to Mass Media Advocacy and Cognitive Choice Processes," in R. Petty, T. Ostrom, and T. Brock (eds.), Cognitive Response to Communication Processing Approach, (New York: Wiley, 1976).

Robert S. Wyer, Jr., Cognitive Organization and Change: An Information Processing Approach, (New York: Wiley, 1974).