Toward a Cognitive Structure Conceptualization of Product Familiarity

Larry J. Marks, Pennsylvania State University
Jerry C. Olson, Pennsylvania State University
ABSTRACT - Product familiarity is conceptualized in terms of the cognitive structures of knowledge concerning the product that are stored in memory. Such knowledge could have been derived from direct or indirect experiences with the product in question (e.g., product use vs. advertisements or word-of-mouth). Different amounts and types of product experience should be reflected by variation in cognitive structure. Theoretically, differences in cognitive structures should influence the cognitive processes and outcomes that involve those cognitive structures and, thereby, should also affect overt behavior. Data from an existing study are used to illustrate these ideas. Suggestions for future research are offered.
[ to cite ]:
Larry J. Marks and Jerry C. Olson (1981) ,"Toward a Cognitive Structure Conceptualization of Product Familiarity", in NA - Advances in Consumer Research Volume 08, eds. Kent B. Monroe, Ann Abor, MI : Association for Consumer Research, Pages: 145-150.

Advances in Consumer Research Volume 8, 1981      Pages 145-150


Larry J. Marks, Pennsylvania State University

Jerry C. Olson, Pennsylvania State University

[This research was funded by a corporate grant to Peter D. Bennett, whose inputs are gratefully acknowledged along with the insightful comments of two anonymous ACR reviewers. We also thank Aydin Muderrisoglu and Robert Humberson for conducting the experimental sessions.]


Product familiarity is conceptualized in terms of the cognitive structures of knowledge concerning the product that are stored in memory. Such knowledge could have been derived from direct or indirect experiences with the product in question (e.g., product use vs. advertisements or word-of-mouth). Different amounts and types of product experience should be reflected by variation in cognitive structure. Theoretically, differences in cognitive structures should influence the cognitive processes and outcomes that involve those cognitive structures and, thereby, should also affect overt behavior. Data from an existing study are used to illustrate these ideas. Suggestions for future research are offered.


Concepts such as product familiarity, product experience or expertise, and/or prior information have been popular as mediating variables in many models of consumer behavior. However, despite the frequency with which product familiarity has been proposed as an individual difference variable affecting choice behavior, the concept has not yet been clearly conceptualized in a compelling manner.

The resulting confusion is apparent in the variety of terms researchers have used to describe this general concept. For example, Berelson and Steiner (1964) found that "pre-existing information" was one of several predisposing factors in determining audience receptivity to "congenial and noncongenial messages". Park (1976) measured product familiarity in terms of subjects' agreement with statements designed to operationalize Howard and Sheth's (1969) concepts of extensive, limited, and routinized problem solving. Based on the Bayesian concept of prior distribution, Woodruff (1972) used subjects' evaluations of a brand-attribute combination and their uncertainty about this rating to operationalize "prior information'' about brand attributes. Laatovicka (1979) measured "knowledge about the product class" by asking subjects if they could "talk about a general group of products for a long time." He also measured "remembered personal experience" by subjects' responses to 'I can remember having purchased something in this general group of products." Raju and Reilly (1979) measured product familiarity in terms of subjects' self-reported "frequency of use, overall familiarity, and knowledge of how to select the best brand." These studies exemplify the diverse and occasionally vague approaches taken in much of the past research on "familiarity." Thus far, this research has emphasized developing predictive models of consumer choice, where measures of product familiarity are one element in such models. Unfortunately this approach has not produced a clear conceptualization of the construct of product familiarity.

The purposes of this paper are to propose a broad conceptualization of product familiarity, defined in terms of the cognitive structures of product knowledge that are derived from past product-related experiences, and to present initial data regarding its usefulness and validity. Our broad objectives are to begin to develop a better understanding of how product experiences are represented in cognitive structure, and of how cognitive structures influence the cognitive processes involved in handling product information (e.g., making purchase choice decisions).


Product familiarity has often been operationalized in terms of product-related behavior, perhaps most often in terms of self-reported product usage experience or frequency of use (of. Anderson, Engledow and Becker 1979, Jacoby, Chestnut and Fisher 1978). An information processing approach, however, provides a different perspective. From this point of view, the construct of interest is not the past product experience per se, but the cognitive representations of that experience that are stored in memory (see Bettman 1979, Olson 1978b, Russo and Johnson 1980). These representations can be considered to be organized in memory as a pro-duct-related cognitive structure or schema. Product schemas contain knowledge in the form of coded representations of brands, product attributes, usage situations, general product class information, as well as evaluations and choice rules. This coded information can be considered to be organized or linked together and stored in memory as a structural framework of knowledge (Norman and Bobrow 1975). Thus, product cognitive structures or schemas may contain not only factual knowledge about the product, but also evaluations or affect, and in well developed schemas, purchase criteria and even decision rules or strategies (Olson 1978a).

Given this information processing perspective, product familiarity can be defined in terms of consumers' cognitive structures for a product that were presumably caused by various product-related experiences. Consumers with different amounts or types of product-related experience probably have acquired cognitive structures with differing characteristics. These variations in cognitive structure are assumed to cause the observed behaviors previously attributed to the vague concept of "product familiarity." Specific features or characteristics of cognitive structure are addressed later.

Although this line of reasoning may be conceptually compelling, it will not have much impact on our research unless we can develop reliable and valid measures which tap, at least relatively directly, these cognitive structures for products and brands. Before addressing operational issues, however, consider the developmental aspects of product familiarity.


Several authors have called for a theoretical explanation regarding how product-related experiences create cognitive structures and/or influence existing cognitive structures (cf. Olson and Mitchell 1975, Russo and Johnson 1980). Hayes-Roth (1977) presented such a conceptual framework which seems useful for considering the effects of product experience on cognitive structure (see Olson and Dover 1978). According to her model, processing information about a product creates the elementary coded representations of knowledge--the concept nodes or cogits--of a product schema. As more product-related experiences occur, these coded representations are "strengthened" in memory and associations between the knowledge nodes develop and become stronger (i.e., learning occurs). Eventually, presumably after multiple experiences with the product, the configuration of associated representations has developed to a point that it can be considered a well-developed, reasonably stable structure.

Moreover, substructures of the now "well-developed" schema may be recoded, a chunking or "unitization" process (cf. Miller 1956, Simon 1974) by which several separate components of the cognitive structure come to be treated (cog-natively) as a single representation. Thus, this newly created concept or chunk represents several other concepts. Chunking can be considered as a process of forming abstract concepts by creating more general concepts to represent several more concrete concepts. According to Hayes-Roth (1977), a chunk usually can be cognitively decomposed into its "original" components, given appropriate instructions and processing, although under ordinary circumstances it would be treated as a unitized representation.

Consumers who are quite familiar with a product due to multiple past experiences might be expected to have formed a stable, complex cognitive structure (or set of structures) of product knowledge, perhaps containing multiple chunks (Olson and Dover 1978). At least one published study provides indirect support for the chunking propositions discussed above. Edell and Mitchell (1978) found that subjects with experimentally-manipulated prior experience regarding a previously unfamiliar product category generated fewer total thoughts in response to an advertisement for a brand in that product class than did subjects with little or no prior experience. From the perspective discussed above, the experienced subjects may have generated ad-stimulated thoughts about more abstract concepts--chunks if you will, "bigger" concepts with "extra'' meaning--and thus they produced fewer total responses. Certainly this interpretation is only speculative; however, as a partial test of this notion, subjects' cognitive responses could be analyzed in terms of their level of abstractness or concreteness.


It is generally accepted that people use information from long-term memory to perceive, interpret, and store new information (cf. Bettman 1979, Mitchell 1978, Olson 1978b, 1980). From the perspective advocated here, product-related schemas are activated and provide the necessary cognitive structure for interpreting product information or new product experiences (see Olson 1978a). In this sense, cognitive structures have a fundamental influence on the processes of attention, encoding, evaluating, storing and using information. For instance, differences in memory schemas should be evidenced by variations in how consumers process information for which those cognitive structures are relevant, whether the information is contained in ads, for instance, or is derived from product use experiences. Thus, consumers' cognitive responses to an advertisement (Wright 1973) could provide clues about how cognitive structures influence encoding processes (Olson 1980). In sum, peoples' existing cognitive structures should exert a 'powerful influence on their cognitive processes and on their subsequent overt behavior.


Perhaps the major stumbling block in doing research on cognitive structures is the measurement problem. How does one measure the content and organization of a memory scheme for a product? This issue is too large for a thorough discussion here. Briefly, however, Olson (1978b) has advocated "free elicitation" as a possible measurement technique. In this procedure, subjects are given cue stimuli or memory probes (usually words) and asked to "Tell me everything that comes to mind when I say..." Presumably, the verbal responses elicited in response to product and/or situation-specific probe cues can provide (admittedly imperfect) indicants of the content and perhaps the organization of cognitive structures. [Russo and Johnson (1980) use a thought protocol task similar to free recall tasks and the free elicitation task used here, but that may provide a somewhat different perspective on product cognitive structures.] In our research program thus far, we have had some success with this technique (see Kanwar, Olson and Sims 1981, Olson and Muderrisoglu 1979). A simple free elicitation procedure was used in this research to provide indices of two characteristics of a product cognitive structure.


The study described below serves only as an initial, exploratory investigation of the ideas discussed above. It is exploratory because the research was designed primarily for other purposes; however, part of the original design produced data that are relevant for our objective. Specifically, there are two broad issues of major interest. First, what differences in cognitive structures result from varying levels of product experience or familiarity? That is, do our presumed measures of cognitive structure reflect differences in product familiarity? Second, how do these cognitive structure variables effect subsequent information processing operations and the formation of product attitudes and purchase intentions?


Product and Subjects

The product was an innovative office chair. The chair had been designed to orthopedically accommodate and support the user's body in a manner that would be both healthful and comfortable. Moreover, the chair was intended to enhance the worker's task performance while working at a desk. The specific chair was totally unfamiliar to all subjects, although it was being marketed in other regions of the country. Subjects consisted of 34 business students (mostly MBA's with some senior undergrads) and 20 secretaries from the University. Subjects were volunteers selected on a convenience basis. They were paid $4.00 for their participation.


The part of the original study examined here can be represented as a simple 2 x 2 between-groups design. Two levels of product experience or familiarity (lower and higher) were represented by the student and secretary groups. The second factor was manipulated by exposing half of the subjects in each group (on a random basis) to a fairly elaborate sales promotion message for the desk chair; the remaining subjects received no product message. An after-only design was used because taking pre-exposure measures was considered to be too sensitizing, and besides, subjects had no pre-experimental knowledge of this particular chair that could be measured,

Product Familiarity.  The main focus of interest is the cognitive structure representation of product familiarity. To maximize inferential rigor, it might have been preferable to experimentally manipulate familiarity, and thus cognitive structure. However, it is difficult to create major differences in product familiarity and related cognitive structures in the context of a controlled experiment. These variables develop naturally in the real world over relatively long periods of time and many experiences. Moreover, for the initial studies in a program of research on cognitive structure, it seemed most efficient to first establish whether differences in existing cognitive structures could be measured. Later, after we learn more about cognitive structures, we can make better attempts to manipulate them. Therefore, we chose to use the secretary and student groups as a classification variable, a surrogate "manipulation" of familiarity with office desk chairs. It seemed reasonable to assume that secretaries have greater amounts of experience and familiarity with office desk chairs than do students.

Exposure to Product Message.  The second independent variable was exposure or not to a sales presentation for the new chair. Half of the subjects in each group saw a 6-minute slide show of professional quality, produced by the manufacturer, and accompanied by a synchronized tape-recorded sales presentation. In the promotion message, the product was presented as of high quality and innovative in design. Four major sales points were described verbally and portrayed visually (the chair is accommodating to one's body, health giving, task motivating, and comfortable).


Subjects were run individually. To enhance their motivation and to provide a context for the experimental procedures, subjects were told to imagine that they were part of an employee committee at their company which was to recommend the type and style of furniture to be used in a remodeling of their offices. In fact, the chair is sometimes sold in this way. After completing an informed consent form, subjects in the exposure condition were shown the slide show presentation. Immediately after, they were asked to record "all the thoughts you had while watching the presentation." Then, concepts were elicited in response to three probe cues that were presented via the instruction, "tell me all the characteristics that come to mind when you think about buying an office desk, ... an office desk chair, ... a filing cabinet." Desk chair was always the second probe. The interviewer recorded all responses verbatim. Following the third probe, subjects rated their attitudes ("o and purchase intentions (BI) regarding the advertised chair.

The nonexposed subjects served as an internal control group to provide an index of peoples' normative cognitive structure regarding the three probe cues, uninfluenced by any persuasive message. After completing the informed consent form, these subjects received the free elicitation procedures. No other measures were taken (no cognitive responses, "o or BI).

Dependent Variables

Cognitive Structure.  To measure differences in cognitive structure due to variation in product familiarity, we examined the results of the free elicitation instruction, "Tell me all the characteristics that cone to mind when you think about purchasing an office desk chair." Empirical indicators of two cognitive structure characteristics were developed (see Kanwar, Slash and Sims 1981). The number of concepts elicited served as an indicator of cognitive structure dimensionality and the percentage of the elicited concepts judged more "abstract" (rather than more concrete) was used as a measure of abstractness of cognitive structure. [In another study, we obtained adequate levels of agreement for these two judgments between two independent coders (r = .85 for total number of thoughts; r = .87 for judgment of abstract vs. concrete; see Kanwar, Olson and Sins 1981)] Whether a concept was considered abstract or concrete was based on experimenter judgment. Significant differences in these indicators were expected between the secretary and the student groups due to their presumed differences in product experience. However, exposure to the promotional message might also produce some variation.

Cognitive Processing.  Immediately following the message subjects were given four minutes to write down the thoughts they had while watching the presentation." These cognitive responses were coded in terms of the numbers of counterarguments, support arguments, and total thoughts produced. As discussed earlier, the cognitive responses of the more and less experienced groups were expected to vary because of their different product-related cognitive structures.

Attitude and Intentions.  Attitude toward the chair ("o) was measured by the mean of three bipolar, 5-point scales labeled good-bad, high quality-low quality, and like-dislike (coefficient alpha = .82). Behavioral intention (BI) was measured on a 7-point bipolar scale by asking the subjects how likely they would be to recommend purchase of the chair (not at all likely-very likely).


Check on Familiarity Classification

Intuitively, it seemed reasonable to assume that secretaries, by virtue of their daily experiences with office desk chairs, would be more familiar with the general product class than students. Thus, secretaries' cognitive structures regarding desk chairs should be different from those of students. To establish that their past experiences with desk chairs did indeed differ, we asked both secretaries and students to report several past behaviors. As expected, 1001 of the secretaries had worked at a job "that involved desk work," while only 65% of the students had. Moreover, secretaries reported that 85% of their work time was spent working at a desk, compared to 67% for those students who had ever had a "desk job." [Two independent judges agreed on 85% of these judgments.]

Differences in Cognitive Structure

Next, it is important to establish that the secretary-student distinction, as a surrogate "manipulation" of product familiarity, produced variation in our measures of cognitive structure. Empirically, differences between secretaries and students might be reflected in (a) the total number of salient characteristics of desks elicited, and (b) the level of abstraction of the elicited concepts. Moreover, we need to determine whether or not exposure to the persuasive sales material influenced these elicitation measures. To enhance the study's low power caused in part by the relatively small cell sizes (n's = 17 and 10), and because the exploratory nature of this research makes us relatively more sensitive to Type II than Type I errors, we adopted an alpha level of .15 for statistical significance (cf., Cohen 1977).

A 2-way, Familiarity by Exposure ANOVA showed that exposure to the sales promotion did not affect the number of concepts elicited (p > .50, M's = 6.1 for exposed and 5.8 for not exposed, respectively), or the abstractness of the elicited concepts (p > .20, M's = 44% and 36%). Although not statistically reliable, the modest increase in abstract concepts due to exposure was to be expected, since the sales promotion message did describe and portray several abstract ideas (regarding the comfort, healthfulness, and accommodating properties of the chair) that subjects probably were not aware of prior to message exposure. The ANOVA also revealed that the total number of elicitations for students was somewhat greater than that produced by the secretaries (M's = 6.2 vs. 5.2, F = 6.60, df = 1/50, p = .0l), This difference is consistent with the proposition that greater product familiarity results in the chunking of cognitive elements, which in turn causes fewer verbalized responses to the elicitation probe. Implicit in this notion, however, is the requirement that those fewer, chunked concepts represent higher levels of abstraction. In fact, the presumably less familiar students did produce a lower percentage of abstract concepts than did the secretaries, but not significantly so (M's = 41% and 34%, p > .20). In sum, the present data provide weak support that the surrogate "manipulation" of familiarity produced differences in cognitive structure. [Parenthetically, it might be noted that an office chair is not a highly complex product. Thus the degree of abstraction that logically can be expected may be somewhat limited. If so, the variance in the percentage of abstract concepts to be discovered might be rather small for a product of this type. For more complex products, differences in level of abstraction might be more apparent.]

Effects of Cognitive Structure and Product Message

Next, we examine the effects that subjects' cognitive structures had on their responses to the promotional material. The following results are based only on those 17 students and 10 secretaries who were exposed to the sales promotion message.

Cognitive Responses.  Consistent with Edell and Mitchell's (1978) results and the chunking proposition, we found that the presumably more experienced secretaries produced marginally fewer cognitive responses to the product promotion than the less experienced students (M's = 7.0 vs. 8.0, F = 1.51, df = 1/26, p = .23). Secretaries produced no more support arguments than students (M's = 2.2 versus 2.0, p > .70), but did have fewer counterarguments (M's = .4 versus 1.8, F = 6.16, df = 1/26, p > .02) than the students.

These results provide weak support for the notion that cognitive processing of a product message is influenced by one's product-related cognitive structures, which in turn reflect differences in product familiarity. By this perspective, secretaries may possess better integrated, more abstract, and presumably more effective product structures. Therefore, they might be expected to better understand and appreciate the rather complex information presented in the sales promotion massage. Since the product was convincingly described as of high quality, secretaries would be expected to perceive it as such and support argue for the message, which they did, although not significantly more so than students. Moreover, secretaries would not be expected to find as many points with which to disagree in the apparently accurate product message. In contrast, students with less-well developed, product-related schemas, may not have clear criteria on which to judge the product. Thus, they might be expected to have some difficulty in relating the information in the advertisement to their personal experiences. If so, they would be more likely to disagree with the information and produce more counterarguments, which they in fact did. Alternatively, however, students may simply have been more cynical or critical than the secretaries, or may have felt greater "demands" to appear critical during the experimental task.

Attitude and Intentions Toward the Chair.  Cognitive responses, especially counterarguments, have been shown to mediate the formation of attitudes and intentions (e.g., Olson, Toy and Dover 1978, Wright 1974). Therefore, it was expected that the secretaries would be more likely than the students to perceive the chair as being of high quality and to recommend its purchase. This was indeed the case. Secretaries had more favorable product attitudes than the students (M's = 4.6 versus 4.l, F = 3.03, df = 1/26, p < .10). And, consistent with their more positive attitudes toward the chair, secretaries were more likely than students to recommend purchase of the chair (M's = 6.0 vs. 5.1, F = 2.93, df = 1/26, p < .10). These results lend additional credence to the presumed casual flow between cognitive responses, attitudes, and intentions (cf. Edell and Mitchell 1978. Olson, Toy and Dover 1978, Wright 1974).


Like many initial investigations of new conceptualizations and measures, the present research generates more questions than it answers. The paper's basic value is probably in raising the issue of how "product familiarity" is to be treated as a construct in consumer behavior theory. Although weak, we believe that the pattern of obtained results are sufficiently interesting to warrant further investigation.

Although the present data were derived from a larger study primarily intended to address other issues, this does not invalidate the obtained results, In particular, the elicitations were carried out about as we would do in a study specifically designed to investigate cognitive structure issues. There are, however, problems in interpreting the product familiarity/experience, student-secretary factor. Because levels of familiarity were not manipulated, a variety of other differences exist between the two groups besides their past experiences with desk chairs. These confoundings become serious if one or more of these other factors can account for the observed pattern of results.

A series of internal analyses revealed that neither age, nor sex, nor marital status could consistently account for the obtained results. A stronger rival explanation involves the experimental demand characteristics that may have been felt by the secretary subjects. Perhaps the secretaries felt that they were "expected" to react favorably to the positive information provided by the advertisement because they had job experience with the product class. While this could account for their favorable attitudes and purchase intentions, it is not clear why students would not also feel "obligated" to react favorably. Additionally, if the secretaries felt that their "expertise" was an issue, it seems likely that they would have sought other ways to demonstrate their expertise. Thus, compared to students, the secretaries should have produced more, not fewer, verbal elicitations, and more, not less, counterarguments in response to the product message. Perhaps the most compelling alternative explanation for the elicitation results is that students--especially MBA students--are rather verbal, and therefore likely to produce more concepts in a free elicitation task than secretaries. This explanation, however, cannot account for the other effects that were obtained. For instance, why should "high verbalness" have a negative effect on "O or BI?

In summary, although of course it cannot be proposed with high confidence, the most parsimonious explanation for the entire range of obtained results is that members of the two groups had different cognitive structures. Interestingly, a similar study, recently published by Anderson and Jolson (1980), obtained some of the same patterns of results as found here. They found that advertisements for 35mm cameras containing relatively technical information produced more favorable beliefs and attitudes for more experienced consumers than for less experienced consumers. We would hypothesize that greater product experience produced more complex cognitive structures containing abstract concepts (chunks) and strong, clear interrelationships between concepts. Such memory schemas seem necessary in order to comprehend complex information in a meaningful way. Stated simply, complex information is mere persuasive for people who are able to more fully comprehend it.

Future Research Directions

This study introduces the idea of conceptualizing product familiarity, product experience, and product expertise in terms of the cognitive structures produced by past product experience. Clearly, additional theorizing and empirical research is necessary before this conceptualization can be generally accepted for use in other research. In particular, we need a more explicit model of cognitive structure. What are the relevant characteristics of a product cognitive structure or memory schema? What measures can serve as empirical indicators of these constructs? How do these aspects of cognitive structure affect other cognitive processes such as the encoding/comprehension process or the information integration processes involved in judgment or decision making? In short, how is past experience represented in a cognitive structure? Answers to such questions will require an extensive research program focused on the construct validity of a model of cognitive structure.

We have begun such a program (cf. Olson and Dover 1978, Olson and Muderrisoglu 1979, Olson and Sims 1980) which recently has produced signs of progress (Kanwar, Olson and Sims 1981). Thus far, we have proposed three characteristics of cognitive structure: dimensionality, the number of salient concepts scored in memory, articulation, the number of discriminable categories along each dimension, and abstraction, the level of abstractness of the dimensions. In addition to free elicitations, the repertory grid procedure and a knowledge test appear to provide potential measures of these cognitive structure elements (see Kanwar, Olson and Sims 1981). These developments also indicate a need for further improvements in methodology in order to facilitate further research on cognitive structure.

The first stage of future research could use the three cognitive structure characteristics just mentioned to establish the aspects of cognitive structure that are related to varying degrees of product familiarity. It does not seem important at this time to attempt the actual manipulation of product familiarity. Rather, we should establish the differences in structure that are associated with differences in experience. Presumably, the constructs of dimensionality, articulation and abstraction vary not only across levels of familiarity, but also for types of products. If so, of course, the data patterns obtained in this preliminary research may not hold for very simple or very complex products.

Future research should attempt to replicate the present design in order to establish the generalizability of these results. For instance, differences in cognitive structures should be found between groups of subjects who a priori have different, hopefully more extreme levels of product familiarity for several products of varying complexity. Once this step is accomplished, it is necessary to demonstrate that the measured differences in cognitive structure are related to differences in information processing behavior, to attitudes and behavioral intentions, and to overt behavior. While experimental simulations are useful for much of this research, actual field observations in a natural environment will be necessary to establish external validity.

Finally, research should focus on the developmental aspects of product familiarity created by product experience or consumer socialization or consumer education. Essentially, in a longitudinal study, subgroups of subjects could be brought to various levels of product familiarity through product experience. Both cognitive and behavior measures, which presumably would possess known reliabilities and validities by this rims, should reflect the levels of product familiarity.

In sum, a wide variety of interesting questions regarding the cognitive structure representation of product familiarity can be addressed, once we have suitable criterion measures. Of course, these questions have not been answered by the present research. We have merely introduced the idea of conceptualizing product experience and product familiarity in terms of cognitive structure and have produced some data that point to the potential usefulness of this point of view. We hope that others will become interested in developing these concepts and measurement procedures further.


Anderson, Ralph E. and Jolson, Marvin A. (1980), "Technical Wording in Advertising: Implications for Market Information,'' Journal of Marketing, 44, 57-66.

Anderson, Ronald O., Engledow, Jack L., and Becker, Helmut (1979), "Evaluating the Relationships Among Attitude Toward Business, Product Satisfaction, Experience, and Search Effort," Journal of Marketing Research, 16, 394-400.

Berelson, Bernard and Steiner, Gary A. (1964), Human Behavior: An Inventory of Scientific Findings, New York: Harcourt, grace and World, Inc.

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

Cohen, Jacob (1977), Statistical Power Analysis for the Behavioral Sciences, rev. ed., New York: Academic Press.

Hayes-Roth, Barbara (1971), "Evolution of Cognitive Structures and Processes," Psychological Review, 84, 260-78.

Howard, John A. and Sheth, Jagdish N. (1969), The Theory of Buyer Behavior, New York: John Wiley.

Jacoby, Jacob, Chestnut, Robert W., and Fisher, William A. (1978). "A Behavioral Process Approach to Information Acquisition in Nondurable Purchasing," Journal of Marketing Research, 15, 532-44.

Kanwar, Rajesh, Olson, Jerry C. and Sims, Laura (1981), "Toward Conceptualizing and Measuring Cognitive Structures," in Advances in Consumer Research, Vol. R, ed. Kent Monroe, Ann Arbor: Association for Consumer Research.

Lastovicka, John L. (1979), "Questioning the Concept of Involvement Defined Product Classes," in Advances in Consumer Research, Vol. 6, ed. William L. Wilkie, Ann Arbor: Association for Consumer Research, 174-9.

Miller, George A. (1956), "The Magical Number Seven, Plus or Minus Two: Some Limits on Our Capacity for Processing Information," Psychological Review, 63, 81-97.

Mitchell, Andrew A. (1978), "An Information Processing View of Consumer Behavior," in Research Frontiers in Marketing: Dialogues and Directions, ed. S. C. Jain, Chicago: American Marketing Association, 178-83.

Norman, Donald A. and Bobrow, Daniel G. (1975), "On the Role of Active Memory Processes in Perception and Cognition," in The Structure of Human Memory, ed. Charles N. Corer, San Francisco: W. R. Freemen.

Olson, Jerry C. (1978a), "Inferential Belief Formation in the Cue Utilization Process," in Advances in Consumer Research, Vol. 5, ed. H. Keith Hunt, Ann Arbor: Association for Consumer Research, 706-13.

Olson, Jerry C. (1978b), "Theories of Information Encoding and Storage: Implications for Consumer Research," in The Effect of Information on Consumer and Market Behavior, ed. Andrew; Mitchell, Chicago: American Marketing Association, 49-60.

Olson, Jerry C. (1980), "Encoding Processes: Levels of Processing and Existing Knowledge Structures," in Advances in Consumer Research, Vol. 7, ed. J. C. Olson, Ann Arbor: Association for Consumer Research, 154-60.

Olson, Jerry C. and Dover, Philip A. (1978), "Attitude Maturation: Changes in Related Belief Structures Over Time," in Advances in Consumer Research, Vol. 5, ed. H. Keith Hunt, Ann Arbor: Association for Consumer Research, 333-42.

Olson, Jerry C. and Mitchell, Andrew A. (1975), "The Process of Attitude Acquisition: The Value of a Developmental Approach to Consumer Attitude Research," in Advances in Consumer Research, Vol. 2, ed. M. J. Schlinger, Chicago: Association for Consumer Research, 249-64.

Olson, Jerry C. and Muderrisoglu, Aydin (1979), "The Stability of Responses Obtained by Free Elicitation: Implications for Measuring Attribute Salience and Memory Structure," in Advances in Consumer Research, Vol. 6, ed. William L. Wilkie, Ann Arbor: Association for Consumer Research, 269-75.

Olson, Jerry C., Toy, Daniel R. and Dover, Philip A. (1976), "Mediating Effects of Cognitive Responses to Advertising on Cognitive Structure," in Advances in Consumer Research, Vol. 5, ed. H. Keith Hunt, Ann Arbor: Association for Consumer Research, 72-8.

Park, C. Whan (1976), "The Effect of Individual and Situation-Related Factors on Consumer Selection of Judgmental Models," Journal of Marketing Research, 13, 144-51.

Raju, P. S. and Reilly, Michael D. (1979), "Product Familiarity and Information Processing Strategies: An Exploratory Investigation," Journal of Business Research, 8, 187-212.

Russo, J. Edward and Johnson, Eric J. (1980), "What Do Consumers Know About Familiar Products?" in Advances in Consumer Research, Vol. 7, ed. Jerry C. Olson, Ann Arbor: Association for Consumer Research, 417-23.

Simon, Herbert A. (197&), "How Big is a Chunk?" Science, 183, 482-8.

Woodruff, Robert B. (1972), "Measurement of Consumer's Prior Brand Information," Journal of Marketing Research, 9, 258-63.

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

Wright, Peter L. (1974), "Analyzing Media Effects on Advertising Responses," Public Opinion Quarterly, 38, 192-205.