Consumers' Cognitive Journey Through the Product Forest

Banwari Mittal, State University of New York, Buffalo
ABSTRACT - This paper develops a broad theoretical framework on Consumers' cognitive as well as affective processes underlying their product selection tasks. The consumer is viewed as constructing cognitive structures of alternative offerings which aid initial implementation of a choice task; a combination of more specific information use heuristics and autonomic priming of emotional schemata is hypothesized to complete- the choice task.
[ to cite ]:
Banwari Mittal (1983) ,"Consumers' Cognitive Journey Through the Product Forest", in NA - Advances in Consumer Research Volume 10, eds. Richard P. Bagozzi and Alice M. Tybout, Ann Abor, MI : Association for Consumer Research, Pages: 464-469.

Advances in Consumer Research Volume 10, 1983      Pages 464-469

CONSUMERS' COGNITIVE JOURNEY THROUGH THE PRODUCT FOREST

Banwari Mittal, State University of New York, Buffalo

[I am grateful to Dr. Gerald Zaltman, Dr. C. Whan Park, Dr. Joel B. Cohen, and Dr. Irene Frieze for encouragement (without implying their endorsement of the views expressed herein) and to the two reviewers and ACR editors 'or comments on an earlier draft of this paper.]

ABSTRACT -

This paper develops a broad theoretical framework on Consumers' cognitive as well as affective processes underlying their product selection tasks. The consumer is viewed as constructing cognitive structures of alternative offerings which aid initial implementation of a choice task; a combination of more specific information use heuristics and autonomic priming of emotional schemata is hypothesized to complete- the choice task.

INTRODUCTION

Most consumer decision research has concerned the strategic (i.e., deliberately pursued) process of comparisons among a specified set of alternatives (e.g., lexicographic, conjunctive, etc., judgmental models) and/or of judging the overall worth of a brand from its disparate attribute values (e.g., Anderson's information integration model, Fishbein's multi-attribute model). This process is an important subset of the larger decision process which has not been explicitly modeled in the extant literature. This larger process extends beyond the subset in the following ways: (a) consumers develop product knowledge over time through voluntary as well as involuntary acquisition of information. This knowledge must somehow be structured before a specific set of alternatives to compare may be identified. (This knowledge structure may, of course, be rudimentary, and may grow further from additional information acquired during the comparison process.) The role of "pre-existing cognitive structures" and of the related memory and attention factors has recently been emphasized by Cohen, Miniard, and Dickson (1980) and Lynch and Srull (1982). (b) The judgment of overall worth are not always the outcome of any cognitive integration of disparate, componential brand attributes (e.g., Zajonc 1980, and Cohen 1982).

There is a need, therefore, to develop a more general model of the larger decision process which (a) incorporates consumers' cognitive structuring of their product environment, and (b) accommodates the role of affective as well as cognitive processes in their brand choices. This paper attempts to develop one such conceptual framework. First, general notions of cognitive classification are introduced. Next, instrumental use of cognitive class structures in consumer choice tasks is described. Variations in these structures and related choice processes are hypothesized to depend upon the "subjective importance" of products. Finally, the nature and sources of affective processes accompanying these cognitive phenomena are discussed.

HUMAN COGNITION AND PRODUCT CLASSIFICATION STRUCTURES

Two axioms, drawn from cognitive psychology literature, will motivate the framework's development:

Axiom 1: Humans inevitably "order" their environment: Man's environment is complex and overwhelming. He puts t some order--order being a generic need of humans--in i this environment by ignoring many elements of its objective diversity and grouping objects into a smaller number of categories. As Rosch et al (1976) state, "one of the most basic functions of all organisms is the cutting F up of the environment into classifications by which nonidentical stimuli can be treated as equivalent" (p. 382). Likewise, Cantor and Mischel (1979) point out, "In order to reduce the complexity of the external stimulus world, the lay person may group both objects and people according to similarities in their essential features, label these natural categories and communicate about the similarities and differences between these kinds or types of objects through this system of shared names or category labels" (p. 4).

The product environment of consumers, when viewed objectively, is also quite complex. A consumer acquires hundreds of products for his life needs. The marketplace offers a large number of alternatives in each product class, all differing more or less from one another. Each marketer seeks to create in the consumer mind a distinct image of his offering. The consumer adapts to this diversity and complexity by sorting various alternatives into a small number of groups or classes. Hence, the proposition: As a matter of natural cognitive process, consumers sort product alternatives into classes and sub-classes (or categories), and thus construct cognitive classification structures of products.

Axiom 2: The "ordering" is directed at achieving efficiency in humans' transactions with their environment: Humans' structuring of their environment is teleological. They cut up and sort the environment so that the resulting categorization may make their transactions with the environment more efficient. Bruner (1958) has pointed out:

Categorization seems to allow one to simplify and reduce an otherwise potentially overwhelming number of stimuli. It selectively focuses attention on certain aspects of particular stimuli, grouping these stimuli under a unifying label and then allowing the perceiver to predict the specific features of any one of the category members on the basis of general expectations about the category.

Categorization thus enables one to predict object or stimuli performance with a minimal intake of information. Given limited memory, attention, and reasoning capabilities (Simon 1957; Slovic 1972), this parsimony in information processing is useful to humans in their role as consumers seeking "best" products. Based only on a limited number of dimensions, available product alternatives are readily categorized. There are overall "consequences" attached to the various resultant categories. Even though limited information is processed for specific product offerings, consequences are attributed to them following their placement in a particular category. The consequences of some categories are judged "undesirable" or "sub-optimal," and these categories are assigned a connotative label of "not-to-be-further-considered." Placement of a brand into one such category eliminates that brand from the choice set and frees a consumer's information processing capacity for alternatives assigned to other categories whose overall consequences are considered more desirable. To illustrate, a cognitive classification structure for television sets is drawn for a hypothetical consumer. (See Figure 1).

FIGURE 1

AN ILLUSTRATIVE 'COGNITIVE CLASS STRUCTURE' FOR ALTERNATIVE OFFEREINGS OF TELEVISION FOR A HYPOTHETICAL CONSUMER

A legitimate question that might arise at this point is, "how do we know whether, in fact, products are sorted it.in the consumer mind in a decision-tree-like manner?" However, this issue is not germane to our discussion since the above representation is proposed as an external one, and no assumption of isomorphism between it and the internal representation is implied. Rather, the external representation adopted here strives only for a functional equivalence in regard to which the following assumptions are advanced:

1. Cognitive classification structures use various dimensions of the product. These dimensions may be intrinsic or extrinsic to the product; they may be performance or image (Myres and Shocker 1978) related. Objects are categorized by the presence or absence of these dimensions or by differential presence (i.e. by a polychotomous division) of these. Dimensionality and intra-dimensional divisions are the key elements of cognitive structures (Scott 1969; Biere 1971; Kanwar, Olson and Sims 1981).

2. The building blocks of the structure are some abstracted features of stimuli/objects, not the objects themselves. Accordingly, categories are specified by product dimensions, not by concrete products. Awareness of all product alternatives is therefore not a precondition for sorting the product environment. A category is formed when a consumer is exposed to just one alternative differing from others in an important dimension. When more alternatives with specifications matching those of an already delineated category are encountered, the category membership grows.

3. There is some reason to believe that when more than one dimensions are used they are used sequentially. That is, the categorization is progressively constructed. The basis of this postulate is that because consumers strive for parsimony in information processing, they can be expected to initially employ a dimension which can eliminate some of the alternatives. (Conjunctive and Lexicographic procedures (Einhorn 1970) typify such a strategy.) Consequently, we can expect that some branches will stop short while others will spread farther into sub-branch structures. For example, a consumer desiring a color TV will be expected to disregard any distinctions among B&W TVs.

4. A consumer sorts alternative offerings in a way which ensures that his utility among alternatives differs minimally within classes and maximally across classes. Otherwise there is no reason for sorting product alternatives.

5. Categorization has its benefits as well as costs. The benefits flow from predictability of consequences, based only upon minimal information. Costs result from misclassification leading to erroneous predictions and hence suboptimal selections (Ratchford 1980). A property of product classification structures will be termed "elaborateness." The structure is elaborate when more dimensions are used or when more values on a dimension are distinguished. It is "simple" when only a few dimensions are used. [The term is used by Craig and Tulving (1975) in a modification of the L-O-P concept earlier proposed by Craik and Lockhart (1972). The L-O-P concept characterizes information encoding as a progression from sensory to semantic levels of analysis and concerns the memorability of the resulting trace. While the usage herein is not inconsistent with the L-O-P notions in that a more elaborate structure implies greater semanticity in encoding, it is believed that the simplest of product structures would entail more than the mere sensory analysis. So, here, all product structures are assumed to entail semantic analysis. And this paper does not dwell upon the memorability of various product structures. Reviewing the inadequacies of the L-O-P concept, Olson (1980) has argued that neither the "progression" nor the "memorability" hypothesis is sufficiently established. Rather than invoking all the nuances of the L-O-P concept, therefore, it is best to adhere to the limited specification, proposed above, of the term "elaborateness" as representing the dimensionality and intradimensional divisions entailed in the product classification structures.] Where the cost of misclassification is high, consumers would want to fine tune their classification of products. This would lead them to sort alternatives into more categories. The elaborateness of classification structures would therefore seem to depend upon relative costs of misclassification as well as benefits of Categorization.

Choice Process and Categorization

It is easy to see that a classification structure with the above properties will direct the choice process. Here, a two stage view of the choice process is adopted (Payne 1976; Wright and Barbour 1977; Park 1978). In the first stage (called the elimination stage) the alternatives not satisfying certain criteria are eliminated yielding a smaller set for closer examination in the second stage (called the selection stage).

It is proposed that "cognitive classification" of products is a means of implementing the first stage of the choice process. As an illustration of how this would transpire, consider the following metaphoric description: Let us imagine the multiple product alternatives as constituting a forest which a consumer muse negotiate in order to reach his desired destination. The decision-tree-like classification structures can be viewed as "pathways" which guide this journey. These pathways are hypothesized to become learned to varying degrees depending upon the amount of product information to which a consumer is incidentally as well as voluntarily exposed. This learning would differ across product classes, as argued in the next section. Accordingly, the amount of cognitive effort in delineating and traversing necessary pathways is hypothesized to vary across product classes. At the end of one of these paths is deemed to lie what will be termed here one's "own-category." ["own-category," as defined here, is specified by criterial dimensions, not by member products. Not all its member-products are considered. In contra-distinction, "evoked-sets" (Howard and Sheth 1969, p. 26) are the number of brands considered.] This is a category which includes one's final choice-to-be. It is not further categorizable because no criterion (i.e. "pathway") of accept/reject type is identifiable. Several alternatives in this category can now be processed only with respect to "degree of acceptability." The first stage is completed when a consumer traveling along the cognitive pathways arrives at his "own category." Interesting variations in this journey can be observed across product classes as discussed next.

Subjective Importance of Products

One fairly general dimension that can be used to order products is their "subjective importance," that is, the importance as perceived by consumers. (This concept is similar to that of Howard and Sheth 1969, p. 419.) The subjective importance of products may stem from multiple sources: "centrality" of the needs for whose satisfaction the product is being sought; social, economic or financial risks perceived in product selection; variation in product offerings, etc. One or more of these aspects may combine to make a product class more or less important to consumers so that they become more or less motivated to make an optimal selection. [The treatment in this paper is at a level of generality that makes it unnecessary to determine the particular source for subjective importance in any given instance. A structural analysis of "subjective importance" of products will be useful but is deemed beyond the scope of this paper.] A lack of motivation to optimize his choice should cause a consumer to expect only a minimal effort (Burnkrant 1976) in making inter-brand distinctions. Consequently, one would expect that when products are of little importance, the classification structures would be "simple" (i.e., less elaborate). To use a previously introduced metaphor, the forest of alternative offerings in a trivial (i.e., low in subjective importance) product class would be cognized/ negotiated by using a short "pathway." Moreover, the pathway must be easy to learn so that traversing it does not entail much cognitive effort. That is, even the few dimensions employed in classifying a trivial product must be the ones that are easily discernible. To illustrate, a possible class structure for table salt is shown for a hypothetical consumer. (See Figure 2.) In this instance, it is assumed that all that matters to this particular consumer is whether or not the salt is iodized. And this is very easy to tell from the package label.

FIGURE 2

COGNITIVE STRUCTURE FOR SALT FOR A HYPOTHETICAL CONSUMER

For subjectively important products, on the other hand, the cognitive structures are expected to be "elaborate." The relatively high motivation for an optimal selection would induce the perceived costs of misclassification to be high. Consequently, product alternatives would be differentiated to a finer degree. One would expect a longer pathway sorting the product forest, and the different portions of the pathway to be varyingly learned. Presumably, the initial portion will have been learned more than the later portions. In the TV example (Figure 1), the B&W/Color dimension would be easier to discern than, say, the color-lock feature whose function and/or utility a consumer may not readily determine.

For subjectively important products, then, some explicit effort would be needed to cognize and traverse the later portions of the pathways. There is support in the literature (e.g., Bettman and Zins 1977) for the fact that some selection criteria are learned only after the consumer has actively acquired and processed information bearing directly upon a specific choice task. This means that when consumers enter a marketplace, so to speak, to acquire an important product, they carry in their heads only a semi-developed cognitive classification structure for that product. Then, as they spend time purposely attempting to learn more about the specific product environment, they are able to sight new pathways, and advance progressively toward their "own categories."

The Second Stage of the Choice Process

For Trivial Products. When products are trivial, the few dimensions employed in delineating the "own category" should be all that matter. In the second stage, therefore, in general, we would expect consumers to be relatively indifferent as to which particular offering to buy from their "own category." To illustrate, a consumer may be indifferent as to which of the "iodized" salts to buy. Given an overall orientation marked by indifference a consumer would be expected to expend little time and effort in making the final selection.

The overall characterization of the strategy employed at this stage is a "minimal effort strategy." This strategy can be implemented through (a) the "availability" heuristic (Tversky and Kahneman 1973), i.e., whichever brand name comes to mind readily; (b) physical convenience (i.e., availability at hand): When none or more than one brand names come to mind, buy the one most conveniently available; and (c) when more than one brand are conveniently available, Single Attribute ComParison, e.g. looking up the prices of a few conveniently reached brands. A reasonable hypothesis would be that a single attribute comparison for only a few brands with externally and readily available attribute information will be the upper bound of information processing undertaken to make a final selection from the "own category."

For Important Products. When products are important, the concern with an optimal choice will continue to be operative beyond the own-category delineation. That is, the consumer would be anxious to select the "best" of the brands from the "own-category." For this purpose, the consumer would want to assess these alternative brands as thoroughly as possible. This would induce him to consider as many attributes as are necessary to deduce the "overall worth" of alternatives. Although several attributes are likely to be used, it would be oversimplistic to suggest that a compensatory or multi-attribute model (Fishbein 1963) would always apply. In circumstances when consumers, owing to their high concern with the optimality of choice, deem as inadequate their own ability to reliably evaluate various attribute information, they would tend to rely on surrogate indicators of-brand worth such as reputation of a brand name, price, store image, etc. and on direct advice from expert and trustworthy sources. The important point here is that consumers will search attribute as well as surrogate information, consider several types of information simultaneously, combine these to deduce the overall worth of brands, and then select a brand with the best overall rating. In sum, this strategy may be termed as an "information-integration" strategy

AFFECTIVE-OVERTONES TO A COGNITIVE PROCESS

So far, the consumer has been portrayed as an explicit-goal seeking cognitive creature, objectively separating

the 'good' from the 'bad', pursuing a conscious and intersubjectively communicable brand evaluation procedure. Self-reflection as well as everyday empirical observations of others suggest that consumers buy many products simply because they feel "emotionally attracted" toward them. In such instances, consumers do not always assess objective features of products and may sometimes be even unaware of the bases for their purchase decisions. In this section are described cognitive class structures and choice processes for the class of products whose purchase is primarily determined by some sort of emotional engagement.

Products as Emotional Satisfiers

All products are sought by consumers for satisfaction of some needs. Here, two broad categories of needs will be distinguished: (i) functional; and (ii) expressive. The former refers to humans' need to maximize gains from their physical environment; the latter concerns their need to symbolically communicate to others some aspects of themselves (Katz 1960). It is well recognized that products serve both functional and symbolic (OF expressive) needs (e.g., Levy 1959). Products directed dominantly at the expressive category of needs will be termed here as "expressive" products. [The term "expressive--defined as, "symbolically communicative of some aspect of oneself"--is used broadly. It includes use of products to express one's deep values as well as surface personality traits such as moods, transitory associations, and emotional surges. If one makes emotional investment in some object and comes to feel affective attraction toward it,one extends part of oneself onto that object. Use of a product owing to an emotional attraction is included, therefore, in the 'expressive' category.] It is difficult at the current state of knowledge to draw up the complete specifications for consumers' orientations toward expressive products. Three elements of it are tentatively suggested below:

(i) Expressive Evaluations Use Global Features: Functional products are evaluated for their physical performance, whereas expressive products are judged for their cit with selected aspects of oneself. The evaluation of expressive products will therefore utilize "image" dimensions, which are extrinsic, holistic and significative of the user personality (Myres and Shocker 1979).

Zajonc (1980) has postulated the existence of a new class of features of objects which do not lend the themselves to content discriminations. Termed "preferenda," these are characterized as "quite gross, vague and global" (p. 159). Whereas performance dimensions are, in the main, intrinsic to the product, image dimensions are "constructed" by the consumer, assessing a product through the "looking glass" of the self-concept. "Style," "prestige," "masculine," etc. are examples of the labels consumers give to such gross, global features.

(ii) The overall affective reaction to product offerings is not tied to specific beliefs about the products. Because affective assessment proceeds from perception of some global features, and content features are not the main bases of the object judgment, the resulting affect is only loosely tied to object related beliefs. (A multi-attribute model of brand attitude is less likely to hold for predominantly expressive products.) Zajonc (1980) has argued extensively that despite a strong research tradition in psychology which regards affect as post-cognitive;-the dispensability of cognitive processes as a prelude to affect formation is unmistakable.

(iii) Awareness of the affect formation process is likely to be low. While there is no one theory of emotional experience, one reasonable view is that people have emotional schemata in their long term memory, and that the perception of some stimulus properties evokes this schemata directly. (See, Leventhal 1980; Wilson 1979.) In everyday life, emotional experiences occur rather "densely." Consequently, the subjective feeling experiences get stored in long term memory. This is why we can relive an emotional experience without an external stimulus, simply by thinking about it. Emotional schemata are also "primed," of course, by external stimulus features. (E.g., seeing a colorful picture may evoke an experience of 'pleasantness'.) Again, because in one's life such priming for each discrete emotion (Izard 1977) occurs repeatedly, the corresponding stimulus properties serving ff as cues and the particular emotional schemata become linked rather instantaneously. That is, the perceptual code is able to make a direct contact with the emotional schemata making cognitive mediation unnecessary and consequently an awareness of affect formation absent or low

Classification Structures and the Choice Process for Expressive Products

Based upon the foregoing three premises concerning affective processes, the cognitive classification structures and choice processes are hypothesized to proceed as follows:

Classification Structures. Since expressive products are assessed, mainly, on global, holistic dimensions, the classifying dimensions can be but few even when products have high subjective importance. Through a generalized learning about the environment, about product features, and about social attributes of users of particular products/brands, some features of products come to be judged as surrogate indicators of particular images. Brand name,price, store name, particular styles, etc., are examples of such image-surrogates. Modern-conservative (as in the architecture of a house), aesthetic appeal (as in a painting), expensive/cheap looks (as in furnishings), "for urban teenagers" (as in branded jeans), or, all inclusively, "my type/not my type," etc. are some examples of evoked images.

Counting both the surrogate features and the evoked image labels, the total number of dimensions employable in classification can still be rather limited. Furthermore, these dimensions do not easily lend themselves to polychotomous divisions. Dichotomous distinctions (modern/ conservative, my type/not my type) are all one can and need make. Both, the dimensionality and dimensional articulation staying low, the cognitive structures for expressive products are expected to be rather "simple" or less elaborate.

Choice Process. When an expressive product offering gets delineated into the "own category" (e.g., designer name jeans), an affinity or affect toward it is(hypothesized to be) experienced while it is being categorized. [No assumption of any temporal relationship between the experience of affect and the identification of a category need be made presently. Sometimes a category forms first, and affect then attaches to it. At other times, category identification is at least partially guided by an experience of affect upon exposure to stimuli.] Cohen (1981) has suggested that categories are often inherently evaluative. Such affect is formed over time, and stored, upon exposure to various brands or communications about them. (This is equivalent to saying that an emotional memory schemata is 'primed', and a link to the particular offering--features of which served as priming cues--is established in that schemata.) When faced with a purchase occasion, the consumer retrieves the affect toward the various brands and selects a brand toward which he finds himself most affect-laden.

It was earlier suggested that the perceptual code directly evokes an affect schemata without any cognitive mediation. Some of these codes would be non-verbal e.g., visual, auditory, olfactory. "Overall appearance" of an object, or a visually encoded imagery of oneself wearing, for example, Gloria Vanderbilt jeans, are illustrative of nonverbal coding. Moreover, the communication of affect relies much more on nonverbal channels (Ekman and Friesen 1969); Schneider, Hastorf, and Ellsworth 1979). Due to spontaneity of affect formation, and/or nonverbalizability of affect-processes, consumers are unaware or unable to articulate (Ericsson and Simon 1980) why they liked a particular offering. However, there is a heightened awareness of the experienced affect itself. Choice outcome is conscious and deliberated one; only its process is less subject to focal attention and consciousness.

The above represents a "pure" type. In practice, products are seldom "purely" expressive; accordingly, the affective processes would be expected to occur as concomitant (rather than as substitute) of the more "cognitive" processes described earlier.

CONCLUSION

The framework presented herein is intentionally metatheoretical (i.e., a way of thinking about consumers' decision processes) rather than a taut set of testable hypotheses. The two have their separate merits, but could not both be accomplished given the space limits. Portions of the framework have, however, been operationalized and empirically tested elsewhere (Mittal 1982).

Much conceptual work is in progress in the domain of consumer cognitions: properties of consumer cognitions (e.g., Gutman and Reynolds 1979; Hirschman and Douglous 1981); product familiarity as cognitive structures (e.g., Russo and Johnson 1980; Mark and Olson 1981); effects of knowledge structure on decision making (e.g., Mitchell 1978; Olson 1980; Brucks and Mitchell 1981); categorization as a basis of substitutability (e.g., Gutman 1981); residual affect not explained by multi-attribute modeling (Olson and Mitchell 1981); the structure of memory representations and their relevance to attitude concept (Cohen 1981), etc. The framework presented here is induced from these more specific ideas, as well as it is deduced from the more general perspectives on human cognition and on human emotions. It is formulated at an intermediate level of generality where it can accommodate diverse and specific developments. As an applied framework, it i is tentative within the limitations of available knowledge in the more basic disciplines upon which it draws. Particularly, the characterization of affect-laden choice processes in this paper has a long way to go before it may be "formalized" (Zaltman. et al 1973).

REFERENCES

Bettman, James R. and Michel A. Zims (1977), "Constructive Process in Consumer Choice," Journal of Consumer Research, 4, 75-85.

Bieri, J. (1971), "Cognitive Structures in Personality," in H.X. Schroder and P. Suefelds, ed., Personality Theory and Information Processing, 178-308.

Brucks, Merrie and Andrew Mitchell (1981), "Knowledge Structures, Production Systems and Decision Strategies," in Kent B. Monroe, ed. Advances in Consumer Research, Chicago: Association for Consumer Research, 8, 750-755.

Bruner, J.S. (1958), "Social Psychology and Perception," in E.E. Maccoby, T.M.Newcomb and E.L. Hartley, eds. Readings in Social Psychology, New York: Holt, Rinehart and Winston.

Burnkrant, Robert E 4 1976), "A Motivational Model of Information Processing Intensity," Journal of Consumer Research. 3, 21-30.

Cantor, Nancy, Walter Mischel (1979), "Prototypes in Person Perception," Advances in Experimental Social Psychology, 12, 3-52.

Cohen, Joel B. (1981), "The Role of Affect in Categorization: Toward a Reconsideration of the Concept of Attitude, paper presented to the Twelfth ACR Conference.

Cohen, Joel B., Paul W. Miniard and Peter R. Dickson (1980), "Information Integration: An Information Processing Perspective, in Jerry C. Olson. ed. Advances in Consumer Research, Ann Arbor: Association for Consumer Research, 161-170.

Craik, Fergus I.M. and Lockhart, Robert S. (1972), "Levels of Processing .A Framework for Memory Research," Journal of Verbal Learning and Verbal Behavior, 11, 671-684.

Craik, Fergus I.X. and Tulving, Endel (1975), "Depth of Processing and the Retention of Words in Episodic Memory," Journal of Experimental Psychology: General, 104, 268-294.

Einhorn, Hillel J. (1970), "Use of Nonlinear, Noncompensatory Models in Decision Making," Psychological Bulletin, 73, 221-230.

Ekman, P. and W.V. Friesen, (1969) "The Repertoire of Nonverbal Behavior: Categories, Origins, Usage and Coding, " Semiotica, 1, 49-98.

Ericsson, K., S. Anders and H.A. Simon, (1980), "Verbal Reports as Data," Psychological Review, 87(3), 215-951.

Fishbein, Martin (1963), "An Investigation of the Relationship Between Beliefs About an ObJect and the Attitude Toward the Object," Human Relations, 16, 233-240.

Gutman, Jonathan and Thomas Reynolds (1979), "An Investigation of the Levels of Cognitive Abstraction Utilized by Consumers," J. Eighmey, ed. Research Under the Sun, Chicago: American Marketing Association.

Gutman, Jonathan (1981), "A Means-End Model for Facilitating AnaLysis of Product Markets Based on Consumer Judgement," in Kent B. Monroe, ed. Advances in Consumer Research, Chicago: Association for Consumer Research, 8, 116-191.

Hirschman, Elizabeth C. and Douglas, Susan P. (1981), 8, 100-105.

Howard, John A. and Sheth, Jagdish N. (1969), The Theory of Buyer Behavior, NY: John Wiley and Sons, Inc.

Izard, C.E. (1977), Human Emotions, NY: Plenum.

Kanwar, Rajesh, Jerry C. Olson and Laura S. Sims (1981), "Toward Conceptualizing and Measuring Cognitive Structures," in Kent B. Monroe, ed. Advances in ConSumer Research, Chicago: Association for Consumer Research, 8.

Katz, Daniel (1960), "The Functional Approach to the Study of Attitudes," Public Opinion Quarterly, 24, 163-204.

Leventhal, Howard (1980), "Toward a Comprehensive Theory in Emotion," Advances in Experimental Social Psychology, 13, 139-207.

Levy, S.J. (1959), "Symbols for Sale," Harvard Business Review, 37, 117-124.

Lynch, John G. and Thomas K. Srull (1982), "Memory and Attentional Factors in Consumer Choice: Concepts and Research Methods," Journal of Consumer Research, Vol G, No. 1, 18-37.

Marks, L.J. and Jerry C. Olson (1981), "Toward a Cognitive Structure Conceptualization of Product Familiarity ," in Kent B. Monroe, ed. Advances in Consumer Research, Chicago: Association for Consumer Research, 8, 145-150.

Mittal, Banwari (1982), "Understanding the Bases and Effects of Involvement in Consumer Choice Process," Ph.D. Dissertation, Pittsburgh: University of Pittsburgh, Graduate School of Business.

Myres, James H. and Allan D. Shocker (1978), Toward a Taxonomy of Product Attributes, WP-320, Working Paper Series, Graduate School of Business, University of Pittsburgh, PA.

Olson, Jerry C. ((1980), "Encoding Process: Levels of Processing and Existing Knowledge Structures," in Jerry C. Olson, ed. Advances in Consumer Research, Ann Arbor: Association f or Consumer Research, 154-160.

Olson, Jerry C. and Mitchell, A.A. (1981), "Are Product Attribute Beliefs the only Mediator of Advertising Effects on Brand Attitude," Journal of Marketing Research, Vol. XVIII, 318-322.

Park, C. Whan (1978), "A Conflict Resolution Choice Model," Journal of Consumer Research, 5.

Payne, John W. (1976), "Heuristic Search Processes in Decision Making," in Beverlee B. Anderson, ed. Advances in Consumer Research, Chicago: Association for Consumer Research, 3, 321-327.

Rosch, E.C., Mervis, W. Gray, D. Johnson and P. Boyes-Braem (1976), "Basic Objects in Natural Categories," Cognitive Psychology, 8.

Russo, J. Edward and Eric J. Johnson (1980), "What do Consumers Know About Familiar Products," in Jerry C. Olson, ed. Advances in Consumer Research, Chicago: Association for Consumer Research, 417-423.

Schneider, D.J., A.H. Hastorf and P.C. Ellsworth (1979), Person Perception, MA: Addison Wesley.

Scott, W.A. (1969), "Structure of Natural Cognitions," Journal of Personality and Social Psychology, 12.

Simon, H.A. (1957), Models of Man: Social and Rational, NY: Wiley and Sons.

Slovic, Paul (1972), "From Shakespeare to Simon: Speculations--and Some Evidence--About Man's Ability to Process Information," Oregon Research Institute Monograph, 12(3).

Tversky, Amos and Daniel Kahneman (1973), "Availability: A Heuristic for Judging Frequency and Probability," Cognitive Psychology, 5, 207-232.

Wilson, W.R. (1979), "Feeling More Than We Can Know: Exposure Effects Without Learning," Journal of Personality and Social Psychology, 37, 811-821.

Wright, Peter (1975), "Consumer Choice Strategies: Simplifying vs. Optimizing," Journal of Marketing Research, 11, 60-671.

Wright, Peter and Frederick Barbour (1977), "Phased Decision Strategies: Sequel to an Initial Screening," in Mart in R. Starr and Milan Zelany, eds. North-Holland TIMS Studies in the Management Sciences, Multiple Criteria Decision Making, Amsterdam: North Holland, 91-109.

Zajonc, R.B. (1980), "Feeling and Thinking: Preferences Need no Inferences," American Psychologist, 35(2) 151-175.

Zaltman, Gerald, Christian Pinson and R. Angelmar (1973), Metatheory and Consumer Research, Hinsdale, IL: Holt, Rinehart and Winston.

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