Advances in Consumer Research Volume 19, 1992 Pages 299-303
CONSUMER STEREOTYPING: THE COGNITIVE BASES OF THE SOCIAL SYMBOLISM OF PRODUCTS
Eva M. Hyatt, Appalachian State University
This paper provides a framework for the investigation of the cognitive processes underlying the formation of judgments regarding products typically owned by members of different social groups, termed consumer stereotyping. It is an attempt to extend the cognitive perspective to the social realm of consumption. Based on the well-estabilshed notion that products provide symbolic, social meaning about their owners, principles from the social cognition literature are extended to explain how and why such consumer stereotypes are formed. This proposed line of research provides a way to empirically test many of the theoretical concepts that have been developed regarding product symbolism.
Most of the growing number of studies taking the cognitive perspective in consumer research deal with the various cognitive processes (attention, encoding, storage, retrieval and inference-making) as they apply to advertising or product information (e.g., Burke and Srull 1988; Johnson and Fornell 1987; Sujan and Bettman 1989), and not to information about consumers themselves. That is, the cognitive processes driving the formation of social judgments regarding product ownership by consumers have not yet been investigated. This paper brings together findings from various areas of cognitive psychology, social cognition, and consumer behavior to provide a framework for a new line of research termed consumer stereotyping which addresses these issues.
The process of consumer stereotyping may be defined as the formation of generalizations about consumption objects possessed or used by members of a particular social category. More specifically, it is based on the well-established idea that people associate ownership of certain products with membership in a certain social group. Such products take on symbolic qualities associated with the group and become descriptive of the group members. From this perspective, the consumption objects owned and used by consumers are viewed as social traits and behaviors. These consumer traits become organized in memory so as to affect later processing of related information. This viewpoint has been neglected in the field to date by consumer researchers utilizing cognitive theory and methods. Instead, they view the products people consume as inanimate objects or bundles of attributes with little concern for their symbolic significance. This is an attempt to extend the cognitive perspective to the social realm; i.e., to try to understand some of the mental processes driving the social symbolism of products.
Before developing the concept of consumer stereotyping in more detail, a distinction needs to be made between it and the well-known concept of segmentation. Segmentation is the process of dividing a market into consumer groups who merit separate marketing mixes based on some similar characteristic(s). This is similar to consumer stereotyping in that both concepts involve the categorization of consumers into homogeneous groups according to some descriptor variable(s). However, segmentation is a managerial function aimed at gaining competitive advantage, whereas consumer stereotyping is a natural process people engage in due to cognitive limitations. The investigation of consumer stereotyping focuses on the cognitive processes underlying the association of products with various social groups and the biases in information processing that result.
The following literature review brings together findings from various areas of research that are integral to the conceptualization of consumer stereotyping that follows.
There is an established notion in the marketing and consumer literatures that products provide symbolic meaning to and about their owners above and beyond their functional uses (Levy 1959; McCracken 1988; Mick 1986; Solomon 1983). This symbolic meaning is often social in nature. In this sense goods and services are a type of social tool that serve as a means of communication between people (Grubb and Grathwohl 1967). The essence of a product, then, becomes not the physical product itself, but the relation between the product, its owner, and the rest of society. This is especially true of socially visible products that are more central in the process of social interaction among people.
These arguments support a fundamental assumption of the notion of consumer stereotyping--that the products people own comprise items of social information about them. That is, a product is a socially perceivable trait or an identifiable unit of social information about its owner. Hence, consumers have product-ownership "traits" in addition to personality traits. Product-ownership traits represent the "extended self" (Belk 1988). Belk's conceptualization of the extended self posits that consumption objects actually become part of the self, much like personality traits. The extended self research deals with how people encode who they are into external expressions (i.e., product ownership), while consumer stereotyping research is concerned with how observers decode external product ownership information into meaningful impressions of the consumers of those products.
Product ownership, then, can be studied using social perception methods. The study of consumer stereotyping in particular is concerned with investigating the social perception of products that are typically associated with membership in various groups. At this aggregate level, perceived patterns of product usage by members of various social groups can be ascertained and investigated in much the same manner as are the perceived personality and behavior patterns of such groups studied by social stereotyping researchers.
Categorization and Schema Theory
The organization of social information into categories provides the basis for giving it meaning and plays a central role in social perception. Rather than hold each discrete piece of information separately, people make their social world manageable by categorizing (Ostrum, Pryor, and Simpson 1981). People do not see others simply as individuals who perform isolated behaviors, but as group members exhibiting patterns of behaviors. This simplification of a complex world is the major task of category systems: to provide the maximum amount of information with as little cognitive effort as possible (Rosch 1978).
Rosch has posited a model of categorization based on schema theory. Her research involves the semantic meaning of concrete objects, but can shed light on social categorization as well. This model pictures categories as organized around salient prototypes (Rosch 1975). That is, the internal structure of a category is composed of a central meaning or prototype based on the clearest instances of that category, surrounded by other less representative members. This results in varying degrees of membership in every category, with members being somewhat good or bad representatives of the category according to their degree of deviation from the prototype. The prototype itself, which may or may not exist in reality, represents the mean or central tendency of all attributes relevant to category membership, with more salient attributes contributing more to this average. Thus, a network of attributes is formed, with fuzzy edges of categories overlapping with each other and the more prototypical core remaining distinct. Semantically, this implies that the prototype has more associative links to the category label than less representative members (Rosch and Mervis 1975).
Social categories, or groups of people based on some descriptor variable such as gender, race, or social class, are more heterogeneous than are natural object categories. That is, the underlying internal structures and the formation and inference-making processes characterizing social categories are much more wide-ranging (Lingle, Altom, and Medin 1984). Social categories often form overlapping classes which may include some of the same individuals as members in several different categories. Any one person can be categorized into a large number of groups, and any one attribute of that person, such as ownership of a particular product, can be associated with many different groups.
An attribute which is used to assign an individual to a particular category is known as a membership attribute. So when someone performs some behavior or exhibits some trait that is very typical of belonging to a particular group, he or she is quickly compared to the stereotype representing that category and, if the similarity between attributes is high enough, is assigned to that group. Inferences are then made about the likelihood of other associated attributes being present. These are known as inference attributes, and are ascribed on the basis of category membership. Both processes occur in consumer stereotyping.
A category, as described by the prototype model, is a type of schema. Schemas lead to expectations about the structure and content of incoming information (Hastie 1981; Fiske and Taylor 1984). Schema theories deal with the differences between actual incoming stimulus information and its abstract mental representation. Incoming information is interpreted in light of what it should look like according to the schema that is invoked; i.e., a schema is a normative structure against which incoming data are matched (Taylor and Crocker 1981).
Such information is then encoded and stored along with category labels according to this schematic organization. During retrieval, other information in the same schema or in closely related schemas is easily confused with the original information, so that as time goes by, the original facts and details become blurred with schema-generated associations and inferences, and incorrect generalizations or biases can result (Hamilton 1981). This is especially the case when a particular descriptor or category label dominates the processing of information about someone, which seems to be the case with information regarding strangers or unfamiliar persons (Ostrum, Pryor and Simpson 1981). Less congruent or typical characteristics of that person may become less accessible in memory, thus forming the basis for stereotyping. For example, if I see a man wearing a sweat suit and running shoes (membership attributes), my schema for "athletic," along with my stereotype for "jock," may be invoked. Later, when recalling information about this incident, I may incorrectly remember that he was also carrying a gym bag and may forget the color of his hair. I may also infer that he was in good physical condition (inference attribute). Any specific piece of information can be evaluated, then, as either incongruent, congruent, or irrelevant with reference to the schema, and will be differentially recalled accordingly (Hastie 1981).
The concept of a stereotype was originally conceived of as a "picture in our head" of a typical member of some social group that has been handed down to us by our culture and used to rationalize and defend our social position, as well as to simplify the confusion of the outer world (Lippmann 1922). More recently, it has been established that the normal cognitive processes discussed thus far are adequate for explaining the existence of stereotypes. No motivational construct is needed to explain the biases that exist in social information processing (Taylor, Fiske, Etcoff and Ruderman 1978; Taylor and Crocker 1981). They occur simply because people take short-cuts (due to processing limitations) in the interest of cognitive efficiency, and use only the most readily available, consistent, and salient information when making judgments. Incoming social information is inevitably selected, simplified, and categorized according to certain invoked descriptor variables. This labeling produces biases in memory that lead to biased perceptions of various social groups (Allport 1954). A stereotype is, then, simply the social equivalent of the prototype, possessing all the attributes most typically associated with a particular group (Brewer, Dull, and Lui 1981; Fiske and Taylor 1984).
When processing incoming information about a particular group member, that particular descriptor variable or category label dominates the encoding, organization, and retrieval of information about that person as well as the inference-making process, so that certain characteristics of that person may become less accessible in memory, or may even be ignored altogether in the encoding process (Ostrum, Pryor and Simpson 1981). Social information becomes organized around the stereotype of the descriptor category, and the unique features of each group member are ignored. Different members of the same group are treated as equivalent through the use of shared labels, then, and cognitive energy and capacity is thus conserved (Cantor and Mischel 1979); however, this is at the cost of lost information and oftentimes incorrect generalizations (Taylor and Crocker 1981). When discrepant information is present, stereotype beliefs are much more likely to be broadened or new subtypes to be added to the network than they are to be thrown out or changed (Weber and Crocker 1983); i.e., stereotypes are very resistant to change.
Each of the background topics just reviewed focuses on a different aspect of consumer stereotyping, but all are essential for a complete understanding. More specifically, product symbolism theory stresses the socially descriptive nature of products and their relationship to consumers. Categorization theory is concerned with the cognitive organization of information into categories around distinct prototypes. Seeing stereotypes as prototypes reflects the focus on the organization of information and points to the need to investigate what information about category members is accessible, to what degree, and why. Schema theory adds the idea that the mental representation of information within a category or schema is often different from the original incoming stimulus information due to biases in encoding. This is in turn reflected in the biased retrieval of information and in biased inferences made based on such information. Finally, the stereotyping literature looks at the heuristic nature of the stereotyping process, in which a salient descriptor variable dominates the processing of information in order to conserve limited cognitive space. All together we get a well-rounded picture of the cognitive view of the consumer stereotyping process. The major difference between the cognitive investigation of stereotyping in general and that of consumer stereotyping is that the attributes organized around the social stereotype are products or consumption objects owned/consumed by (or inferred to be owned/consumed by) group members rather than personality traits or other behaviors. Thus, the goal in this proposed line of research is to ascertain whether people have product ownership stereotypes of consumers belonging to various social groups and, if so, how the cognitive processes resulting in such stereotypes operate.
Each consumer stereotyping incident involves the following cognitive processes. It begins with exposure to a member or members of a social group. For example, an observer might be waiting in line at the post office and overhear several men in front of him or her that appear to be friends. They are talking about an upcoming turkey shoot and one man is describing his new Winchester shotgun. As they leave they agree to meet Friday night at a local country saloon. The observer deduces that these are typical rednecks, and watches them leave the post office. He or she pays selective attention to traits/behaviors that might further confirm his or her deduction. One man climbs into a Ford pick-up as he reaches into his Levi shirt pocket for a dip of snuff. The other two men get into a Volvo and one man opens a bottle of Coke as his friend lights up a Marlboro cigarette.
Certain behaviors are ignored altogether, while others are encoded into internal representations and stored into long-term memory. All these processes are affected by a pre-existing stereotype of rednecks as a social category. If the observer later attempts to retrieve the stimulus information, certain biases in recall are likely to result, which in turn can affect inferences or further deductions made about the three men. A feedback loop is possible, in which the original stereotype can be affected and modified to better explain the situation.
The study of consumer stereotyping involves the investigation of expected findings regarding the major cognitive processes involved in group stereotyping: encoding of incoming information, storage or organization of information in long-term memory, and inference-making based on this information; i.e., it makes predictions regarding what kinds of biases are likely to occur in a stereotyping incident like the one just described.
The encoding of information is the transformation of external stimuli into internal representations; i.e., the interpretation of incoming stimuli that are attended to. In a stereotyping incident more attention is paid to information that is either congruent or incongruent with the group stereotype. Less cognitive energy is required to encode information that is consistent with prior expectations since it readily fits into the existing cognitive structure, while more is required to encode and make sense of inconsistent information. For example, if I see a yuppie wearing a Polo shirt I can quickly assimilate this fact, whereas if he is drinking Busch beer I might have to think about what circumstances may have led to this atypical event. As a result, both pieces of information will be more accessible in long-term memory than will information that is irrelevant to group membership (e.g., he was also reading a copy of TV Guide).
One issue relevant to the encoding process is whether people engage in category-based (overall comparison with group stereotype) or piecemeal-based (attribute-by-attribute) processing. Category-based processing is more likely as the salience of the category label, the match between incoming product ownership information and category-based expectations, and the degree of short-term memory load increase (Fiske and Pavelchak 1986). Memory is much stronger and faster for category-based processed information. Thus, by presenting experimental subjects with group membership and product ownership information about a consumer, while varying such factors as salience of social group membership, congruence of product ownership profile with pretested expectations, and/or memory load task, and then measuring recall of congruent, incongruent, and irrelevant products, one could better understand the encoding processes underlying the social symbolism of products.
This involves the notion that information is organized into social categories centered around stereotypes. Since typical traits are closer to the stereotypical core, there are more links to them in the long-term memory network and thus should be relied on more heavily when new information is being stored (Cantor and Mischel 1979). For example, I have a node close to the center of my yuppie stereotype for BMW. Closely attached to this node may be places that the typical yuppie might drive to in his BMW, such as the country club to play golf or the symphony. Also close to this node might be the kinds of clothes that the typical yuppie might sport around town in. More typical products should also be retrieved easier, faster, and earlier. By measuring and comparing recall levels, reaction times, and order of recall for products of varying degrees of typicality, as well as collecting cognitive responses regarding which products helped subjects remember other products, some light should be shed on the internal structure of category representations concerning social consumption behavior.
The process of gathering together information, interpreting that information, and making some judgment or inferring additional information is heavily influenced by preexistent expectations (Fiske and Taylor 1984). By knowing the category membership of an individual, people are able to draw inferences about unobservable attributes that may aid in the social judgment task at hand. For example, if I know someone is a yuppie (or categorize him as one), I may confuse information in my yuppie stereotype as being something I know about this fellow in particular so that I have a more complete picture of him. This can be studied by asking experimental subjects to infer additional products owned by profiled group members. Certain stereotyped patterns should emerge. Other types of inferences that are likely (depending on congruence of group membership and product ownership information) include intrusions of stereotype-congruent products in recall measures, generalizations about product ownership from one group member to another, and certain affective responses.
The effects of consumer stereotyping have been proposed to be wide-ranging and diverse. By studying the cognitive processing of product ownership information about group members as has been suggested, a rigorous empirical investigation of the social symbolism of products is possible. This is especially needed in the field of consumer research at present in order to provide a balance with the heavy use of qualitative methods in symbolic research. Understanding the specific cognitive processes driving the social symbolism of product ownership would help to illuminate the valuable descriptive work that has already been done in this area.
Allport, Gordon W. (1954), The Nature of Prejudice, Reading, MA: Addison-Wesley.
Belk, Russell W. (1988), "Possessions and the Extended Self," Journal of Consumer Research, 15 (September), 139-168.
Brewer, Marilynn B., Valerie Dull, and Layton Lui (1981), "Perceptions of the Elderly: Stereotypes as Prototypes," Journal of Personality and Social Psychology, 41, 656-670.
Burke, Raymond R. and Thomas K. Srull (1988), "Competitive Interference and Consumer Memory for Advertising," Journal of Consumer Research, 15 (June), 55-68.
Cantor, Nancy and Walter Mischel (1979), "Prototypes in Person Perception," Advances in Experimental Social Psychology, 12, New York: Academic Press.
Fiske, Susan T. and Shelley E. Taylor (1984), Social Cognition, New York: Random House.
Fiske, Susan T. and Mark L. Pavelchak (1986), "Category-Based versus Piecemeal-Based Affective Responses: Developments in Schema-Triggered Affect," Handbook of Social Cognition: Foundations of Social Behavior, Sorrentino and Higgins (Eds.), New York: Wiley, 167-203.
Grubb, Edward L. and Harrison L. Grathwohl (1967), "Consumer Self-Concept, Symbolism, and Market Behavior: A Theoretical Approach," Journal of Marketing, 31, (October), 22-27.
Hamilton, David L. (1981), "Cognitive Representations of Persons," Social Cognition: The Ontario Symposium, Hillsdale, NJ: Erlbaum, 135-160.
Hastie, Reid (1981), "Schematic Principles in Human Memory," Social Cognition: The Ontario Symposium, Hillsdale, NJ: Erlbaum, 39-88.
Johnson, Michael D. and Claes Fornell (1987), "The Nature and Methodological Implications of the Cognitive Representation of Products," Journal of Consumer Research, 14 (September), 214-228.
Levy, Sidney J. (1959), "Symbols for Sale," Harvard Business Review, (July-August), 117-124.
Lingle, John H., Mark W. Altom, and Douglas L. Medin (1984), "Of Cabbages and Kings: Assessing the Extendability of Natural Objects Concept Models to Social Things," Handbook of Social Cognition, T.K. Srull and R.S. Wyer, Jr. (Eds.), 1, 71-117.
McCracken, Grant (1988), Culture and Consumption: New Approaches to the Symbolic Character of Consumer Goods and Activities, Bloomington, Indiana: Indiana University Press.
Mick, David Glen (1986), "Consumer Research and Semiotics: Exploring the Morphology of Signs, Symbols, and Significance," Journal of Consumer Research, 13, (September), 196-213.
Ostrum, T. M., J. B. Pryor, and D. D. Simpson (1981), "The Organization of Social Information," Social Cognition: The Ontario Symposium, Hillsdale, NJ: Erlbaum, 3-38.
Rosch, Eleanor (1975), "Cognitive Representations of Semantic Categories," Journal of Experimental Psychology: General, 104, 192-233.
Rosch, Eleanor (1978), "Principles of Categorization," Cognition and Categorization, E. Rosch and B.B. Lloyd (Eds.), Hillsdale, N.J.: Erlbaum.
Rosch, Eleanor and Carolyn B. Mervis (1975), "Family Resemblances: Studies in the Internal Structure of Categories," Cognitive Psychology, 7, 573-605.
Solomon, Michael R. (1983), "The Role of Products as Social Stimuli: A Symbolic Interactionism Perspective," Journal of Consumer Research, 10, (December), 319-329.
Sujan, Mita and James R. Bettman (1989), "The Effects of Brand Positioning Strategies on Consumers' Brand and Category Perceptions: Some Insights From Schema Research," Journal of Marketing Research, 26, (November), 454-467.
Taylor, Shelley E. and Jennifer Crocker (1981), "Schematic Bases of Social Information Processing," Social Cognition: The Ontario Symposium, Hillsdale, NJ: Erlbaum, 89-134.
Taylor, Shelley E., Susan T. Fiske, Nancy Etcoff, and Ann Ruderman (1978), "The Categorical and Contextual Bases of Person Memory and Stereotyping," Journal of Personality and Social Psychology, 36, 778-793.
Weber, Renee and Jennifer Crocker (1983), "Cognitive Processes in the Revision of Stereotypic Beliefs," Journal of Personality and Social Psychology, 45, 961-977.
Wyer, Robert S., Jr. and Thomas K. Srull (1981), "Category Accessibility: Some Theoretical and Empirical Issues Concerning the Processing of Social Stimulus Information," Social Cognition: The Ontario Symposium, Hillsdale, NJ: Erlbaum, 161-198.