Attribute Determinance: a Function of Past Memory and External Factors

Meryl Paula Gardner, New York University
ABSTRACT - The effect of memory on situation-specific brand evaluation processes is explored. The impact of internal and external factors on attribute and non-attribute strategies is discussed. An attempt is made to integrate past findings and to suggest directions for future research.
[ to cite ]:
Meryl Paula Gardner (1982) ,"Attribute Determinance: a Function of Past Memory and External Factors", in NA - Advances in Consumer Research Volume 09, eds. Andrew Mitchell, Ann Abor, MI : Association for Consumer Research, Pages: 177-182.

Advances in Consumer Research Volume 9, 1982      Pages 177-182


Meryl Paula Gardner, New York University


The effect of memory on situation-specific brand evaluation processes is explored. The impact of internal and external factors on attribute and non-attribute strategies is discussed. An attempt is made to integrate past findings and to suggest directions for future research.


How does memory affect the brand evaluation process used in a particular situation? This question is most interesting, and most difficult to answer, when each of the terms involved is used in its most general sense. The term "memory" refers to all that is stored - not just product knowledge Analogously, the phrase "evaluation process" refers to how brand affect or brand attitude comes about - not just cognitive, attribute-based evaluative procedures.

Any viable framework for approaching the initial, general question must be able to address the following sub-questions:

(1) How does memory affect the degree of attribute processing used in a particular brand evaluation?

(2) If attributes are not used, how does memory affect what is used instead?

(3) If attributes are used, how does memory affect which ones are used?

a) Which general types of attributes?

b) Which specific attributes?

Past research has addressed each of these questions individually. This paper seeks to provide a framework for integrating past findings and suggesting future research directions.

Brucks and Mitchell (1981) provide a general framework for such an approach. A modification of their model appears in Figure 1.



In this model, brand evaluation is viewed as the output of an internal process influenced by internal and external factors. The internal aspects appear in the dotted line box. They include the individual's long term memory, i.e., internal factors and production systems, and his perception of the problem, i.e., problem characteristics and workspace.

Internal factors are organized as schemata, i.e., frameworks into which thoughts must fit (Calder and Schurr, 1979). These structures influence retrieval of stored information and acceptance of incoming information (Neisser, 1976; Rumelhart and Ortony, 1977; Taylor and Crocker, 1981) Schemata may include five kinds of knowledge traditionally studied by information processing research: terminology, specific facts, relationships, criteria for evaluation, and procedural information. (For a discussion of this typology, see Brucks and Mitchell, 1981). Such knowledge may be with respect to a particular domain or it may be general, world knowledge. In addition, schemata may include elements less frequently discussed by information processing researchers. These include cultural values, sensory memories, motor responses, affect, and response tendencies. Such factors are hard to define and assess, but may have important effects upon consumers' brand evaluations.

Production systems have traditionally been used to help explain human behavior at the information processing level (Newell and Simon, 1972). They provide a control structure to organize the elements of the human information processing system (e.g., memories, encodings, primitive operations) to effectively process knowledge (Newell, 1973). The basic element of a production system is a condition-action statement. The condition refers to elements in the consumer t 5 workspace; the action consists of transformations on such elements (e.g., Klahr, 1973). Productions allow dynamic responses. They can self-trigger and trigger in parallel. Recently, production systems have been used as control systems to explain processing of emotions (Bower and Cohen, 1982).

Internal factors and production systems are stored and relatively enduring, though not always activated. A more transient internal aspect is problem perception. This consists of two parts, the individual's characterization of the problem and his active workspace. The term "problem characteristics" refers to how the consumer structures the problem in his mind. It may include problem definition, alternatives considered, and the perceived urgency of the problem. The term "workspace" refers to all of the material activated at an instant.

External factors are characteristics of the decision situation external to the individual. These include outside pressures, situational importance, budget constraints and outside requirements. In addition, they include characteristics of the product, characteristics of the ad, e.g., amount of information, and characteristics of the viewing situation. e.R., external distractions.

How do internal and external factors interact to form overall affect or attitude toward a brand? Attitude formation involves interacting processes and intermediate steps. In order to answer the question and sub-questions posed earlier, the effects of internal and external factors on some of these processes and steps must be examined.


Problem Characteristics: Internal and external factors may affect such problem characteristics as problem definition, alternatives considered and the urgency of the problem.

Problem definition may be affected in the following ways:

1) Sets of requirements or needs associated with particular situations may be stored and elicited directly.

2) Procedural knowledge for a task category may affect problem definition by specifying behaviors with respect to objects in a domain.

3) A culture's goals and values may affect the way its members define their needs.

4) Aspects of the external environment may catch an individual's attention and affect his needs or his awareness of them. For example, hearing the sound of liquid pouring into a glass may make him thirsty or make him think he's thirsty or make him aware of his thirst. Research is needed to fully unravel these possibilities.

The alternatives considered may be affected by the set of products associated with consumption experiences in the consumer's memory. For example, hunger may evoke many different kinds of alternatives, depending upon prior experiences. For some individuals, the evoked set may consist of names of restaurants. For others, it may consist of names of foods to cook at home. In either case, the set of salient alternatives may be influenced by environmental cues, e.R.. billboards.

The urgency of the problem may be affected by the individual's somatic feelings/needs, e.g., extreme hunger, and by external pressures, e.g., a deadline.

Comprehension: A basic understanding of terminology is necessary to understand and interpret incoming information. In addition, differences in schemata may cause familiar and unfamiliar consumers to differ in the difficulty they experience learning new information about a brand or product. Familiar subjects may be able to readily integrate a new attribute into their complex, well-indexed long term memory (Kinder, Fiske, and Wagner, 1978). Converse (1975, p. 79) points out the differences in costs of assimilating new information for those who are well-informed or poorlyinformed about politics:

If an informed observer hears a surprising policy statement in the news by the Secretary of Defense, he may prick his ears and pay close attention. He relates this information to what he knows of recent policy, what he knows of the Secretary's relationship to the President, what he knows of past positions the Secretary may have taken, and the like, since he is intensely interested to detect even small reorientations of national policy. In short, he automatically imports enormous amounts of prior information that lends the new statement with high interest. The poorly informed person, hearing the same statement, finds it as dull as the rest of the political news. He only dimly understands the role of Secretary of Defense and has no vivid image grounded in past information as to the inclinations of the current incumbent. His awareness of current policy is sufficiently gross that he has no expectation of detecting nuances of change. So the whole statement is confronted with next to no past information at all, hence is just more political blather.

Research involving text processing has shown that individuals who are knowledgeable about an area are able to recall more information about a presented passage than individuals who are not knowledgeable (e.g., Spilich, Vesonter, Chiesi and Voss, 1979; Chiesi, Spilich and Voss, 1979). These studies have involved intentional learning of long, complex stimuli. Further research is needed to ascertain the extent to which these results can be generalized to ad-viewing situations.

Inference Formation: People who know the relationships among product attributes may use such knowledge to form inferences. For example, an individual may believe that smooth riding cars often handle poorly. When presented with *formation about a new car's ride or handling, he can make inferences about the car's performance on the missing attribute.

Chi (1981) has found that expert physicists are better able to form inferences needed for problem solving than novice (beginner) physicists. Further research is needed to ascertain the applicability of research on expert - novice differences in physics to consumer differences in product familiarity.

Moesser (1976) has postulated that new information which can be readily related to other information in an encoder's memory system may be added to the existing structure in semantic memory and could be used to generate inferences. New information which cannot be related to other stored information may be stored as a discrete unit in episodic memory. If encoded as a discrete fact, the information could not be used for inference formation.

Ability to Use Information: People differ in their knowledge of procedures for using, interpreting, manipulating, and combining pieces of information. Chi (1981) found that novice (beginner) physicists tended to lack the knowledge of when to use certain physics knowledge.

Redinger and Staelin (1979) investigated consumer evaluations of refrigerators and air conditioners of varying energy efficiencies. They found that consumers provided only with information concerning annual operating costs chose less energy efficient appliances than those provided with information on how to trade off operating costs with initial price. Consumers in the first group, after becoming aware of the fact that appliances cost a significant amount to operate, seemed to attempt to reduce the amount they spent initially so that they could later afford to pay the high energy costs. They did so by purchasing lower priced, less efficient appliances.

These findings suggest that telling consumers about energy costs without teaching them how to use that information to trade off between initial price and operating cost might decrease sales of energy efficient appliances among consumers who do not know how to use the information. Information provided to consumers seems to affect purchase differently for consumers who differ in their abilities to use the information.

Source and Information Evaluation: An individual's knowledge of specific facts and relationships may aid in source and information evaluation. Presented material may be compared to recalled data and to inferences. If the stimulus materials do not jibe with what is known or inferred, counterarguing against the contents in the stimulus may occur. If the ad or spokesperson seems foolish, unreliable or untrustworthy, source derogation may occur. In such cases, the consumer may not use the stimulus information to form his attitude.

Confidence in Evaluative Ability: The discussion, thus far, has dealt with the effects of the consumer's objective knowledge about products or evaluations. It is also important to understand the effects of the consumers' subjective assessment of his knowledge (Park and Lessig, 1981). Eighmey (1978) reported that only about half of those respondents who had recently purchased a major durable felt that they could judge the product they purchased without obtaining additional information. Consumers may, in fact, know more than they think they do. Perhaps, like rats in learned helplessness experiments, they have been taught not to use the skills they possess.

A consumer's assessment of his ability to evaluate brands in a product class may be affected by interactions between his knowledge and external cues. For example, if a salesDan used terminology unfamiliar to the consumer, he may intensify the consumer's feelings of subjective unfamiliarity. This may make the consumer more vulnerable to the salesman's informational influence (e.g., Park, 1982).

Subjects unfamiliar with a product class are likely to be less confident in their product judgements. Markus (1977) found that individuals who have a self-schema in a domain are more confident in predicting schema-related behaviors than those who do not.

In addition, Raju and Reilly (1977) found that subjects selected brands with fewer below average features when making choices for unfamiliar products than for familiar products. They conjectured that subjects were more likely to "play safe" when dealing with unfamiliar products than when dealing with familiar ones. This tendency might extend to avoidance of new attributes by unfamiliar subjects.

Salience: Krech and Crutchfield (1948) are credited with the earliest definition of salience: "Saliency refers to the fact that not all of a man's beliefs stand out with equal prominence in his cognitive field. He may be more acutely aware of certain of his beliefs than others, they may enter his thoughts more readily, they may be more frequently verbalized - they are, in a word, salient." Thus, salience may be associated with "ease of recall" and is frequently assessed with elicitation procedures (e.g., Ryan and Etzel, 1975). According to a spreading activation theory of memory (e.g., Collins and Loftus, 1975), the items that are elicited first are the ones most strongly associated with a cue node in memory.

Things that are familiar may be more readily recalled than those that are unfamiliar. Chi and Koeske (1979) found that the superior recall of familiar items is directly related to the large number of connections among them in memory. Thorndyke and Hayes-Roth (1979) found that interconnected knowledge substructures or schemata may encourage both facilitation and interference effects. Compared to consumers who are unfamiliar with a product, those who are familiar may have more knowledge in memory related to the product and it may be more cohesively interconnected. We expect consumers familiar with a product to exhibit different recall patterns from those who are unfamiliar.

Marks and Olson (1981) postulated that familiarity may affect the nature of elicited attributes. They found marginal support for the elicitation of more abstract attributes by people familiar with the product class, typing chairs. They postulated that people unfamiliar with the product may not be able to go beyond its obvious features.

Salience is strongly affected by external stimuli. Things which stand out in an environment may receive more attention and so, be more easily recalled (e.g., Taylor and Fiske, 1978). Advertisers can use attention getting device such as color, letter size, and typeset to make a particular attribute stand out in an ad. Under such circumstances the prominent attribute is more likely to be recalled than when it is merely mentioned in the body copy. In addition, when one attribute is prominent, other attributes are less likely to be recalled than when all attributes are merely mentioned in the body copy. Making one attribute prominent seems to direct attention to it and away from all others (Gardner, 1981). Calkins (1894) reported similar findings using color to manipulate the prominence of numerals in a list. (For a discussion of the underlying process, see Simon and Feigenbaum, 1962).

Evaluative Criteria: Salient attributes may or may not be determinant, i.e., used to evaluate brands in a product class. (For a discussion of the relationship between salience and determinance. see Gardner. 1981).

Park and Lessig (1981) have investigated the relationship between familiarity level and reliance on nonfunctional cues such as price and brand name. They found that low familiarity consumers placed a higher confidence on brand name and price than on functional dimensions; moderate familiarity consumers displayed less confidence on brand name and price than on functional dimensions; and high familiarity consumers placed as high a confidence on brand name and price as upon the functional dimensions. They interpreted these findings as reflecting the effects of the consumer's shopping and consumption experiences on his evaluative criteria.

In general, consumers differ in their knowledge of criteria to use to evaluate brands in a product class. The subject's own evaluative criteria and the attributes featured in a stimulus ad may interact to determine the set of criteria used in a given evaluation. For example, we might hypothesize that subjects who are familiar with a product and have a stored set of evaluative criteria may handle missing information differently from those who are unfamiliar with a product and do not have a stored set of evaluative criteria. Those in the former group may form inferences about the omitted attributes; those in the latter group may base their attitudes solely on the attribute information provided. In general, consumers who do not know which attributes to use to evaluate a brand, may use the advertised attributes and little else (e.g., Wright and Rip, 1980).

What happens if both the internal and external environment fail to provide information about evaluative criteria? When confronted with such a situation, an individual may use categorical information. Re may base his evaluation upon attributes which are normally used to evaluate objects in a more general class of products to which the unfamiliar product belongs or is related. For example, if you don't know anything about shaving cream, you might use attributes appropriate for evaluating toiletries in general, such as scent, or those frequently found useful in evaluating products in general, such as price. If a product is considered highly representative of or similar to the more general class of products, the tendency to use this strategy may be enhanced due to the representativeness heuristic originally investigated by Kahneman and Tversky (1972).

Evaluative Stole: A consumer may choose products without forming attribute-based evaluations. Instead, affective channels, category-based affect, pattern matching, or motor responses may influence his brand preferences.

Under some circumstances, consumers may use affective rather than cognitive channels for evaluation. Mitchell and Olson (1981) have found that a subject's evaluation of the ad itself may help explain his attitude toward the advertised brand. Affective channels may be more important for consumers who are unfamiliar with a product class. In addition, the advertiser can encourage non-cognitive heuristics. He might fail to provide adequate factual information (moot ads), use pictures rather than words or present the advertisement too quickly to permit processing. He may also encourage consumers to evaluate the ad, a procedure which may discourage brand evaluation (e.g., Advertising Age, 1981). Special camera angles may encourage the viewer to project himself into the picture. According to Krugman (1965), showing a commercial on television may discourage evaluation.

Affect associated with entire categories or types of stimuli may affect brand choice. Fiske (1982) has investigated affect associated with category schemata. She has presented evidence that affect may be triggered by a match with a type rather than by a piecemeal combination of brand attributes. (For a discussion of category-based models of affect in consumer behavior, see Cohen, 1982).

Pattern matching strategies may be used to select products. For example, a shopper may sniff a cantaloupe and put it in her wagon if its smell is as expected.

Motor responses may also affect evaluations. Zajonc (1982) has found that subjects testing headphones for vertical motion, i.e., shaking their heads up and down, evaluated the headphones more favorably than those testing them for horizontal movement, i.e., shaking their heads left and right. Head movements appeared to affect attitude without the subjects' awareness. The processes underlying the connection between the motor system and affect formation require further investigation.

Number of Attributes Used:

Isen (1982) has found that mood affects problem solving behavior. Subjects who are put into good moods seem to search fewer alternatives and to finish the task more quickly than subjects who are put into bad moods. She has postulated that the former did not want to risk losing their good moods by slow, deliberate processing. In a consumer situation, one might hypothesize that people in good moods might examine fewer brands, use fewer attributes or be less likely to use attributes entirely. Perhaps, they might be more likely to use the affective or categorical heuristics discussed earlier.

Clark (1982) has found that the effects of mood on processing seem to be intensified by physiological arousal. Further research is needed to extend these findings to advertising effects and to consumer behavior.

Processing Time:

Mood and external pressures may accelerate the evaluation process. Generalizing from Isen's (1982) findings, one can hypothesize that consumers in good moods may make purchase decisions more rapidly than those in bad moots. Quick decisions may also result from acute needs, rapid presentation of information, or salesperson pressures.

Processing time may also be affected by stored facts about brands in a product class. Newman and Staelin (1972) found support for the idea that consumers who had purchase and use experience with a major durable could use their stored knowledge and did not have to do as much external search as those who lacked that experience. It might be hypothesized that, for some products, experience may allow more rapid decisions by decreasing external search. In addition, repeating the evaluation process may hasten its execution through learning. Under some circumstances, the process may become automatic

Attitude Extremity:

Linville and Jones (1980) have investigated the effects of category familiarity on the extremity of attitude toward category members. They postulated that people have more complex schemata for in-groups than for out-groups and that such complexity would yield more moderate attitudes for members of in-groups than for members of out-groups. The data supported their model.

Tesser (1978) has found that complex schemata may induce more attitude polarization over time than simple schemata. The complex structures may provide more dimensions which, with thought, can be used to back up, strengthen and intensify initial evaluations.

In addition, Lord, Ross and Lepper (1979), working with strong opinions on a complex social issue, found that subjects used prior attitudes to bias their evaluation and interpretation of new information. Work is underway to see if these effects hold in the more mundane world of consumer products.

Attitude polarization might also be used as a dissonance reducing strategy. Sheth (1968) postulated that consumers unfamiliar with buying a given product class might exhibit more post-purchase dissonance and use more dissonance reducing strategies than those familiar with buying the product class. In such cases, the number of attributes used by unfamiliar subjects in their evaluations would be artificially inflated if assessed by traditional correlation/ regression methods.


Thus far, discussion has focused on the effects of internal and external factors on aspects of brand evaluation. Further research is needed to fully understand these effects and their impact on evaluative criteria. A production system paradigm may systematize that endeavor. Our understanding of processing cannot be complete without a conceptualization of its control system. The preceding section discussed some of the findings that a model of attitude formation must be able to explain.

In conclusion, these findings provide partial answers for the sub-questions posed earlier. Memory seems to affect the degree of attribute processing used in a particular brand evaluation through its effects on problem characteristics, comprehension, inference formation, evaluative style, and mood. Memory appears to affect non-attribute evaluative strategies through its effects on problem characteristics, evaluative style and the use of sensual/motor memories. Memory seems to affect the attributes used in evaluations in two ways. It may affect the nature of attributes used through its effects on problem characteristics, categorization, and evaluative style. It may affect the particular attributes used through its effects on problem characteristics, salience, stored evaluative criteria, and confidence in evaluative ability.


Bower, Gordon and Cohen, Paul (1982), "Emotional Influences in Memory and Thinking: Data and Theory," in M. Clark and S. Fiske (eds.), Affect and Cognition: The 17th Annual Carnegie Symposium on Cognition, Hillsdale, NJ: Erlbaum.

Brucks, M. and Mitchell, A. (1981), "Knowledge Structures, Production Systems and Decision Strategies," in R. Monroe, Ed., Advances in Consumer Research. Vol. 8, Ann Arbor, MI: Association for Consumer Research.

Calder, B. and Schurr, P. (1980), "Attitudinal Processes in Organizations," Research in Organizational Behavior, Vol. 3.

Calkins, M. (1894), "Association," Psychological Review, Vol. 1, 476-483.

Chi, M.T.E. and Rees, E. (May 18, 1981), "Expertise in Problem Solving," Technical Report No. 5, Learning Research and Development Center, University of Pittsburgh, Pittsburgh, PA.

Chi, M. and Koeske, R. (1979 Working Paper), "Recall and Network Representation of a Child's Dinosaur Knowledge," University of Pittsburgh, Pittsburgh, PA.

Chiesi, H.L. Spilich, G.J., and Voss, J.F., "Acquisition of Domain-Related Information in Relation to High and Low Domain Knowledge," Journal of Verbal Learning and Verbal Behavior, Vol. 18, 257-273.

Clark, Margaret (1982), "Affect, Arousal and Social Judgments," in M. Clark and S. Fiske (eds.), Affect and Cognition: The 17th Annual Carnegie Symposium on Cognition, Hillsdale, NJ: Erlbaum.

Cohen, Joel B. (1982), "The Role of Affect in Categorization: Toward a Reconsideration of the Concept of Attitude," in A.A. Mitchell (ed.), Advances in Consumer Research.

Collins, A.M. and E.F. Loftus (1975), "A Spreading-Activation Theory of Semantic Processing," Psychological Review, Vol. 82, No. 6, 407-428.

Converse, P.E. (1975), "Voting and Public Opinion," in F. Greenstein and N. Polsby (eds.), Handbook of Political Science, Reading, MA.: Addison Wesley.

Eighmey, John (1978), "Consumer Research and the Policy Planning of Advertising Regulation," in A.A. Mitchell (ed.), The Effect of Information on Consumer and Market Behavior. Chicago, IL: American Marketing Association, 86-91.

Fiske, Susan (1982) "Schema-Triggered Affect: Applications to Social Perception," in M. Clark and S. Fiske (eds.), Affect and Cognition: The 17th Annual Symposium on Cognition, Hillsdale, N.J.: Erlbaum.

Gardner, Meryl (1981), "An Information Processing Approach to Examining Advertising Effects," Unpublished Doctoral Dissertation, Graduate School of Industrial Administration, Carnegie-Mellon University, Pittsburgh, PA.

Giges, Nancy (October 12, 1981), "Campbell Savors Consumer Input," Advertising Age.

Isen, Alice M. (1982), "Positive Affect, Decision Making Style and Risk Taking," in M. Clark and S. Fiske (eds.), Affect and Cognition: The 17th Annual Carnegie Symposium on Cognition. Hillsdale, N.J.: Erlbaum.

Kahneman, D. and A. Tversky (1972), "Subjective Probability: A Judgment of Representatives," Cognitive Psychology, Vol. 3, 430-454.

Kinder, D., S. Fiske and R. Wagner (1978), "Presidents in the Public Mind," presented at a Conference on Mass Response to Political Leadership, sponsored by the Center for Political Studies, University of Michigan, held in New Haven, Conn., May 22 - 23, 1978.

Klahr, David (1973), "A Production System for Counting, Subitizing and Adding," in W.G. Chase (ed.), Visual Information Processing, New York: Academic Press.

Krech, D. and R. Crutchfield (1948), Theory and Problems of Social Psychology, New York, N.Y.: McGraw-Hill Book Co.

Krugman, Herbert (Fall, 1965), "The Impact of Television Advertising: Learning Without Involvement," Public Opinion Quarterly, 29.

Linville, Patricia and Edward Jones (1980), "Polarized Appraisals of Out-Group Members," Journal of Personality and Social Psychology, Vol. 38, No. 5, 689-703.

Lord, C.G., L. Ross and M. Lepper (1979), "Biased Assimilation and Attitude Polarization: The Effects of Prior Theories on Subsequently Considered Evidence," Journal of Personality and Social Psychology, Vol. 37, No. II, 2098-2109.

Marcus, Hazel (1977), "Self-Schemata and Processing of Information about the Self," Journal of Personality and Social Psychology, 35, 63-78.

Marks, L. and J. Olson (1981), "Toward a Cognitive Structure Conceptualization of Product Familiarity," in K. Monroe (ed.), Advances in Consumer Research, Vol. 8, Ann Arbor: Association for Consumer Research, 145-150.

Mitchell, Andrew and Jerry Olson (August, 1981), "Are Product Attribute Beliefs the Only Mediator of Advertising Effects on Brand Attitudes," Journal of Marketing Research, Vol. IIXX.

Moeser, Shannon, "Inferential Reasoning in Episodic Memory," Journal of Verbal Learning and Verbal Behavior, 15. 193-212.

Neisser, U. (1976), Cognition and Reality, San Francisco, CA.: W.H. Freeman.

Newell, Allen (1973), "Production Systems: Models of Control Structures," in W.G. Chase (ed.), Visual Information Processing, New York: Academic Press, 463-526.

Newell, A. and H. Simon (1972), Human Problem Solving, Englewood Cliffs, N.J.: Prentice-Hall.

Newman, Joseph W. and Richard Staelin (August 1972), "Prepurchase Information Seeking for New Cars and Major Household Appliances," Journal of Marketing Research, Vol. IX, 249-257.

Park, C.W. and V.P. Lessig (September, 1981), "Familiarity and Its Impact on Consumer Decision Biases and Heuristics," Journal of Consumer Research, Vol. 8, No. 2, 223-230.

Raju, P. and M. Reilly (1977), "Product Familiarity and Information Processing Strategies: An Exploratory Investigation," Pennsylvania State University Working Paper Series, No. 57.

Redinger, R. and R. Staelin (1979), "Estimating the Demand for Energy Efficient Refrigerators and Air Conditioners," Report prepared for the Department of Energy, 1979.

Rumelhart, D. and A. Ortony (1977), "The Representation of Knowledge in Memory," in R. Anderson, R. Spiro, and W. Montague, (eds.), Schooling and the Acquisition of Knowledge, Hillsdale, N.J.: Erlbaum.

Ryan, M. and M. Etzel (1975), "The Nature of Salient Outcomes and Referents in the Extended Motel," Association for Consumer Research Proceedings.

Sheth, Jagdish (1968), "Cognitive Dissonance, Brand Preference, and Product Familiarity," in J. Arndt (ed.), Insights into Consumer Behavior, Boston, MA.: Allyn and Bacon, 41-53.

Simon, H. and E. Feigenbaum (1962), "A Theory of the Serial Position Effect," British Journal of Psychology, Vol. 53, 307-320.

Spilich, G.J., G.T. Vesonder, H.L. Chiesi, and J.F. Voss (1979), "Text Processing of Domain-Related Information for Individuals with High and Low Domain Knowledge," Journal of Verbal Learning and Verbal Behavior, Vol. 18, 275-290.

Taylor, S. and J. Crocker (1981), "Schematic Bases of Social Information Processing," in Higgins, Herman and Zanna (eds.), So Mark Cognition: The Ontario Symposium on Personality and Social Psychology, dale, N.J. Erlbaum.

Taylor, S. and S. Fiske (1978), "Salience, Attention, and Attribution: Top of the Head Phenomena," in L. Berkowitz (ed.), Advances in Experimental Social Psychology, Vol. II. New York: Academic Press.

Tesser, A. (1978), "Self-Generated Attitude Change," in Advances in Experimental Social Psychology, Vol. II, New York: Academic Press.

Thorndyke, P. and B. Hayes-Roth (1979), "The Use of Schemata in the Acquisition and Transfer of Knowledge," Cognitive Psychology, II, 82-106.

Wright, P. and P. Rip, "Product Class Advertising Effects on First-Time Buyers' Decision Strategies," Journal of Consumer Research, Vol. 7, 176-188.

Zajonc, Robert (1981), "The Interaction of Affect and Cognition: The Motor Stratum," in M. Clark and S. Fiske (eds.), Affect and Cognition: The 17th Annual Carnegie Symposium on Cognition, Hillsdale, N.J.: Erlbaum.