Effects of Schema Congruity and Involvement on Product Evaluations

ABSTRACT - This study examined how consumer evaluations of products are influenced by two factors: the degree of congruity between products and their associated category schema and the level of consumers' involvement with the evaluation task. It was hypothesized that when product information is congruent with category schema, category-based processing occurs, resulting in assimilation effects; when it is incongruent, contrast or piecemeal processes are used, depending on the level of consumers' involvement. The results generally supported the prediction. Theoretical and managerial implications of the results are discussed.

Citation:

Moonkyu Lee (1995) ,"Effects of Schema Congruity and Involvement on Product Evaluations", in NA - Advances in Consumer Research Volume 22, eds. Frank R. Kardes and Mita Sujan, Provo, UT : Association for Consumer Research, Pages: 210-216.

Advances in Consumer Research Volume 22, 1995      Pages 210-216

EFFECTS OF SCHEMA CONGRUITY AND INVOLVEMENT ON PRODUCT EVALUATIONS

Moonkyu Lee, University of Colorado at Denver

ABSTRACT -

This study examined how consumer evaluations of products are influenced by two factors: the degree of congruity between products and their associated category schema and the level of consumers' involvement with the evaluation task. It was hypothesized that when product information is congruent with category schema, category-based processing occurs, resulting in assimilation effects; when it is incongruent, contrast or piecemeal processes are used, depending on the level of consumers' involvement. The results generally supported the prediction. Theoretical and managerial implications of the results are discussed.

INTRODUCTION

Assimilation-contrast theorists suggest that when people make an evaluative judgment of a social or a nonsocial object, they go through either one of two processes: sometimes they integrate the object into its associated category and make their evaluation on the basis of the category affect, while other times they contrast the object with the category and make their evaluation in the opposite direction from the category affect (Sherif and Hovland 1961; Sherif, Sherif, and Nebergall 1965). On the other hand, categorization researchers propose that sometimes people simply categorize the object and make their judgment on the basis of the likability of the category, while other times they piece together their judgment on the basis of the likability of the individual attributes of the object (Brewer 1988; Fiske 1982; Fiske and Pavelchak 1986; Fiske and Neuberg 1990; Pavelchak 1989).

Although these alternative modes of evaluation processes have been found and reported in numerous studies in the areas of social and cognitive psychology as well as consumer behavior, it is not yet clear when they occur and what types of judgmental consequences they produce. This research focuses on situations where consumers make their evaluative judgments of products, and predicts that the evaluation outcomes will be determined by the degree of congruity between products and their associated category schema, and the level of consumers' involvement with the evaluation task.

THEORETICAL BACKGROUND

Assimilation-Contrast Literature

Past research in psychology and consumer behavior has suggested that some judgments are very sensitive to the context in which they are made. Applied to the present research context, this means that the evaluation of a product is influenced not only by the overall configuration of the product information but also by the characteristics of the cognitive category accessible during the evaluation. The best known of such contextual phenomena are assimilation and contrast (Sherif and Hovland 1961; Sherif, Sherif, and Nebergall 1965). Assimilation takes place when the evaluation is made in a direction toward the current category; contrast occurs when it is made in a direction away from the category. Assimilation and contrast phenomena have been found in many empirical studies (e.g., Herr 1986, 1989; Herr, Sherman, and Fazio 1983; Martin 1986; Martin, Seta, and Crelia 1990; Wilson, Lisle, Kraft, and Wetzel 1989). As an example, Herr et al. (1983) found that assimilation took place when people considered a stimulus as a member of the primed category; otherwise, contrast occurred. They assumed that the primed category served as a standard to which the stimulus was compared. However, this assimilation-contrast paradigm does not take into account the possible occurrence of a piecemeal-based evaluation process, which is described in the following section.

Categorization Literature

Categories are cognitive structures which contain instances or category members that are perceived similar or equivalent (Fiske and Pavelchak 1986; Smith and Medin 1981). Over time and through experience, people develop a number of categories of objects in an effort to organize and understand them. They also develop a set of expectations about the features of a typical category member with respect to a particular category (category schema), and an affective reaction to the category (category affect; Rosch 1978; Mervis and Rosch 1981; Smith and Medin 1981). When making an evaluative judgment of an object, they sometimes base their judgment on the affect associated with the category to which the object belong (category-based process), while other times they use their evaluations of individual attributes of the object (piecemeal process). Then, a question arises as to when people use a category-based and when they use a piecemeal process. This question can be answered by two conceptual models developed in the area, one by Fiske et al. (Fiske 1982; Fiske and Pavelchak 1986; Fiske and Neuberg 1990) and the other by Brewer (1988).

Fiske and her colleagues suggest that informational characteristics of a target stimulus determine which evaluation process is used; category-based processes are used when the stimulus information is descriptively congruent with the category schema, whereas piecemeal processes are used when the information is incongruent with it. Their conceptual model has gained a substantial amount of empirical support in psychology and consumer behavior (e.g., Boush and Loken 1991; Fiske, Neuberg, Beattie, and Milberg 1987; Meyers-Levy and Tybout 1989; Pavelchak 1989; Sujan 1985; Sujan and Bettman 1989). As an example, Sujan (1985) found that the alternative modes of processing occurred depending on the match or mismatch of a product to a category schema, with a match triggering category-based reactions and with a mismatch leading to piecemeal evaluations.

On the other hand, Brewer proposes that a perceiver's level of situational, task-oriented involvement is a primary determinant of different types of processing; category-based processing occurs under low involvement situations, whereas piecemeal processing takes place under high involvement situations. Her model also has been supported by many empirical studies (e.g., Borgida and Howard-Pitney 1983; Celsi and Olson 1988; Howard-Pitney, Borgida, and Omoto 1986; Omoto and Borgida 1988). For instance, Celsi and Olson (1988) found that highly involved consumers exerted more cognitive effort than low involved consumers in processing product information.

One limitation of these two models is that they place a lopsided emphasis upon either the informational or the motivational factor, when in fact, both factors might have a simultaneous influence on the evaluation process (see Wyer and Srull 1989, pp. 305-315, for a further discussion of these models).

In sum, the existing literature identifies different types of processes underlying product evaluations. It also indicates two important factors in determining which process is used: product-category schema congruity and situational involvement. In the next section, a conceptual model is developed along with some specific hypotheses.

A CONCEPTUAL MODEL OF PRODUCT EVALUATIONS

Initial Considerations

Suppose consumers, who are shopping for a color T.V. set, run across the following product: a Zenith color T.V. set with built-in stereo, multi-functional remote control, and automatic shut-off. If they find this product information sufficiently interesting and relevant to their goal, they will first try to categorize the product in an effort to make sense of it (Fiske and Neuberg 1990; Fiske and Pavelchak 1986). In this case, the most dominant cue will immediately bring to the consumers' mind an apparently appropriate category. In the product domain, it is believed that the most salient cues associated with basic level categories include brand name and price, since a substantial amount of research has demonstrated their dominant impact on product evaluations (Zeithaml 1988; Rao and Monroe 1989). These cues will serve as category labels which are used to organize and understand the remaining features of the product. When a category is accessed, the next step for the consumers would be to make a confirmation that the product is really a member of that category, by considering descriptive implications of the rest of the attribute information.

Successful Categorization

To the extent that the consumers perceive the attribute information to be congruent with the current category, regardless of their involvement levels, they will use the category affect as the basis for their overall evaluations, because this type of category-based process is much more efficient than any other types of processes from a cognitive economy standpoint (Fiske and Pavelchak 1986; Fiske and Neuberg 1990; Fiske and Taylor 1991). Even highly involved consumers would not feel the need to re-elaborate on the attribute information, unless they find anything different about the product. Category-based processing will result in assimilation effects; that is, the overall judgment will be made in a direction toward the category affect. Therefore, it is hypothesized that:

H1: Regardless of involvement levels, when the initial categorization is successful, category-based processing and assimilation effects will occur.

Unsuccessful Categorization

When the confirmation check reveals an incongruity between the product and the category expectations, consumers will react to the product information differently depending on their involvement levels.

Under Low Involvement Conditions. When the initial categorization is not successful due to the product-schema incongruity, and the consumers are not highly involved with the judgment task, a piecemeal elaboration upon product attributes is not likely to occur, which would require a considerable cognitive effort. Instead, a "contrast process" will be used; with the current category serving as a standard of comparison, the overall judgment will be made in the opposite direction from the category, simply because this type of process must be a much easier and faster way to reach a judgment than a piecemeal process. Contrast processing will lead to contrast effects, or the overall judgment negatively correlated with the category affect. Thus, it is predicted that:

H2: When the initial categorization is not successful and consumers are not highly involved with the judgment task, contrast processing and contrast effects will take place.

Under High Involvement Conditions. On the other hand, when the consumers are highly involved with the judgment task, they will not only notice the product-schema incongruity but also engage in piecemeal processing. In other words, they will evaluate the product in an attribute-by-attribute manner and integrate their attribute evaluations to reach a final judgment (Anderson 1974; Fishbein and Ajzen 1975; Fiske et al. 1987; Fiske and Pavelchak 1986; Pavelchak 1989; Sujan and Bettman 1989). Consequently, piecemeal processing will produce the overall judgment which is strongly correlated with the evaluations of individual product attributes (called attribute effects, hereafter), but not with the category affect. Therefore, it is predicted that:

H3: When the initial categorization is not successful but consumers are highly involved with the evaluation task, piecemeal processing and attribute effects will occur.

In this study, two types of correlations are used for testing the hypotheses: (1) correlations between overall judgments and category affect, and (2) those between overall judgments and attribute evaluations (cf. Fiske et al. 1987; Pavelchak 1989; Sujan and Bettman 1989). These correlations served as indicators of the different types of processing and judgmental effects proposed in the conceptual model. Specifically, it was predicted that assimilation effects resulting from category-based processing would be evidenced by a significant positive correlation between overall judgments and category affect, contrast effects from contrast processing would be detected by a significant negative correlation between overall judgements and category affect, and attribute effects from piecemeal processing would be indicated by a significant positive correlation between overall judgments and attribute evaluations. The expected pattern of results under the hypothesized conditions is summarized in Table 1.

METHOD

Pretests

A series of pretests was conducted to develop stimulus materials and to see if manipulations would work as intended.

Pretest 1. For a successful manipulation of product-schema congruity, it was very important to select a product category that the subject population was very familiar with. In other words, the product category had to be the one that was associated with a well-established schema. A pretest was conducted with sixty-two students enrolled in business courses at a midwestern university. They were asked to rate their familiarity with several product categories. The results indicated that they were most familiar with the fast-food category. Thus, this category was used for the experiment.

Pretest 2. The purpose of the second pretest was to select a brand name for stimulus materials. The selection criterion was basically the same as was used for selecting the product category: the brand name had to be the one that subjects were very familiar with. Pretest 2 was conducted to find out a brand name that met the criterion. In this pretest, names of fast-food restaurants were considered as brand names of fast-food. An unaided recall measure was used in the pretest, where 68 subjects participated and were asked to write down all the names of fast-food restaurants that they were familiar with. Burger King was chosen for the study because it obtained almost 100% mentioning rate.

TABLE 1

EXPECTED PATTERN OF RESULTS

Pretest 3. The goal of the third pretest was twofold: to develop experimental stimuli for two schema congruity conditions (i.e., Congruent and Incongruent conditions) and involvement manipulation material for two involvement conditions (i.e., High and Low Involvement conditions), and to see if manipulations would work as expected.

Based on a focus group discussion, the congruity was operationalized in the following way. For the Congruent condition, a list of seven fast-food items was created that were typical of Burger King restaurants (e.g., cheese burger, broiler sandwich, French fries, apple pie, etc.). The list included the restaurant name. For the Incongruent condition, another list of seven food items were developed that were at odds with the Burger King image (e.g., personal pan pizza, stuffed pizza slice, garden salad, carrot cake, etc.). In addition, it was ensured that the descriptions within this condition were also varied so they were not suggestive of any particular type of restaurant.

For the involvement manipulation, a cash prize procedure was employed a la Celsi and Olson (1988). Specifically, for the High Involvement condition, subjects were told that they could win a $10 cash prize by answering a few questions correctly at the end of the pretest. Since they were also informed that the questions would be drawn from the main part of the pretest, they were expected to pay more attention to and become more involved with the test procedures.

The effectiveness of the stimulus materials and the involvement manipulation was tested through a 2 (schema congruity) X 2 (situational involvement) between-subjects, factorial design. Seventy nine students who were enrolled in marketing courses were randomly assigned to one of the four conditions. The effectiveness of the schema congruity manipulation was checked by subjects' ratings on 9-point scales on three questions: (1) "How typical are these food items of Burger King?" ("very atypical-very typical"), (2) "How similar are these items to those found in a usual Burger King menu?" ("very different-very similar"), and (3) "How likely will you find these items at Burger King?" ("very unlikely-very likely"). Since these three measures of perceived congruity obtained a high reliability (alpha=.80), simple average ratings on these items were used to check the success of the manipulation. These average ratings were analyzed in a two-way analysis of variance (ANOVA). As expected, schema congruity had a strong main effect on the average ratings of the three items (F(1,75)=538.99, p<.001). The means of the average ratings differed significantly between the two congruity conditions (Ms=6.98 and 1.73 for the Congruent and Incongruent conditions, respectively; t(77)=17.03, p<.001). These results indicated that the manipulation of schema congruity worked as intended. Thus, the same materials were used in the main study.

Success of the involvement manipulation was examined by having subjects indicate how interested, how careful, and how attentive they were in reading the food list and evaluating the restaurant. Specifically, they were asked on 9-point scales: (1) "How interested were you when reading and evaluating the previous menu?" ("not-at-all interested-very interested"), (2) "How carefully did you read and evaluate the previous menu?" ("not-at-all carefully-very carefully"), and (3) "How attentive were you when reading and evaluating the previous menu?" ("not-at-all attentive-very attentive"). Since the ratings on these three items also showed a relatively high reliability (alpha=.81), simple average ratings on the measures were used again to check the success of the manipulation. These ratings were submitted to a two-way ANOVA. The results showed that the involvement manipulation had a significant main effect on the average ratings (F(1,75)=5.85, p<.01). The means of subjects' responses to those two questions differed significantly between the two involvement conditions (Ms=5.72 and 6.55 for the Low and High Involvement conditions, respectively; t(77)=2.34, p<.03). Thus, the manipulation of involvement was successful, and was used again in the main study.

Subjects and Design

One hundred and eleven students enrolled in business courses participated in the experiment for extra course credit. These subjects were randomly assigned to one of four treatment conditions in a 2 (schema congruity: Congruent vs. Incongruent) X 2 (situational involvement: High vs. Low) between-subjects, factorial design. Cell sizes ranged from 23 to 26.

Procedure

A computer laboratory was set up for the experiment. All the research instruments were provided on the computer screen and data were collected on-line. Upon arrival, subjects were randomly assigned to one of the treatment conditions and told that the purpose of the experiment was to help Burger King provide better service and higher quality food. They first provided their general impression of the restaurant by pressing a number on a 9-point bipolar scale on the screen, ranging from "very bad" (1) to "very good" (9). Their ratings served as their original category affect. Then they were told that they would see a list of food items available at a Burger King restaurant and that they should read it with the purpose of forming an evaluation of this particular restaurant. At this point, the cash prize instruction was announced on the screen for the High Involvement conditions. Subjects then received one of the two stimuli that were developed in Pretest 3. They were asked to provide their overall evaluation of the restaurant by pressing a number on a 9-point scale shown on the screen, ranging from "very bad" (1) to "very good" (9). They also indicated their likings for the individual food items listed on the menu (i.e., their attribute evaluations) on 9-point scales from "strongly dislike" (1) to "strongly like" (9). They were then asked, "How important to you is the inclusion of the following food items to the menu of any fast-food restaurant in general (not necessarily Burger King)?" They provided their answers for all the seven food items on 9-point scales from "not-at-all important" (1) to "very important" (9). Then they filled in manipulation checks. The high involvement subjects completed the cash prize questionnaire at this point. Finally, subjects were debriefed and thanked.

RESULTS

Manipulation Checks

Schema Congruity. The same measures used in Pretest 3 were employed to check the effectiveness of the schema congruity manipulation. Since those measures showed a high reliability (alpha=.98), simple average ratings on these measures were used. Again, there was a significant main effect of congruity on the average ratings of the three items (F(1,107)=471.26, p<.001). Neither the main effect of involvement nor the congruity X involvement interaction was significant (F(1,107)=2.75, p>.10, and F(1,107)=.38, p>.54, respectively). Additionally, the means of the average ratings differed significantly between the two congruity conditions (Ms=7.81 and 2.16 for the Congruent and Incongruent conditions, respectively; t(109)=21.61, p<.001). Thus, the manipulation of congruity was successful.

Involvement. Again, the same measures used in Pretest 3 were employed to see if the involvement manipulation was successful. Simple average ratings on the three measures were used again to check the effectiveness of the manipulation, since the ratings on the items showed a high reliability (alpha=.77). Theses ratings were submitted to a two-way ANOVA. For the involvement manipulation, there was a significant main effect on the average ratings (F(1,107)=12.86, p<.001). The main effect for congruity and the congruity X involvement interaction were not significant (F(1,107)=.08, p>.78, and F(1,107)=.00, p>.99, respectively). The means of subjects' responses to those three questions differed significantly between the two involvement conditions (Ms=6.23 and 7.12 for the Low and High Involvement conditions, respectively; t(109)=3.62, p<.001). Thus, the manipulation of involvement was successful.

Judgment Correlations

As mentioned earlier, two sets of correlations were examined: correlations between overall judgments and category affect, and those between overall judgments and attribute evaluations. The use of such correlational measures is based upon the assumption that a high correlation of overall judgments with category affect or attribute evaluations should be indicative of utilization of either of these in arriving at the judgments.

Since category affect could be a function of attribute evaluations to a certain extent in reality, partial correlations were examined in the study. Specifically, they were correlations between overall judgments and category affect with the effects of attribute evaluations controlled for (r_j.ca), and those between overall judgments and attribute evaluations with the effects of category affect partialed out (r_j.att). Attribute evaluations were summarized using the simple average as well as weighted average rules. Since the attribute evaluations integrated by either of the rules were highly correlated (r=.68, p < .01), the results based on these rules were essentially similar. Thus, for the presentation purpose, only the results based on the weighted average rule will be described here (cf. Anderson 1965; Anderson and Birnbaum 1976).

The partial correlations, r_j.ca's and r_j.att's across the experimental conditions are shown in Table 2. H1 predicted that a successful initial categorization would bring about category-based processing and assimilation effects. This hypothesis was confirmed by strong positive r_j.ca's (r=.54, p =.003, and r=.43, p =.012, for the Low and High Involvement conditions, respectively), and nonsignificant r_j.att under the Congruent conditions (r=.10, p =.324, and r=.31, p =.060, for the Low and High Involvement conditions, respectively). H2 proposed that an unsuccessful categorization under a low involvement condition would result in contrast processing and contrast effects. This was directionally supported by the negative r_j.ca under the Incongruent/Low Involvement condition (r=-.10, p=.302), which did not reach significance. A possible explanation for this nonsignificant r_j.ca is that the relatively high category affect (M=5.92) might have had a ceiling effect on the overall judgments. The r_j.att under the same condition was not significantly greater than zero, as expected (r=.39, p=.021). H3 predicted that an unsuccessful categorization under a high involvement condition would induce piecemeal processing and attribute effects. This hypothesis was confirmed by a significant positive r_j.att(r=.62, p=.000) along with a nonsignificant r_j.ca under the Incongruent/High Involvement condition (r=-.24, p=.107). Therefore, the correlational pattern generally supported the hypotheses.

DISCUSSION

The results of the study suggest that consumers engage in different types of information processing when evaluating products, depending on product-schema congruity and involvement levels. According to the results, consumers first check the congruity between incoming product information and their prior knowledge about the product. To the extent that they find a good match between these two, they do not go through an effortful piecemeal process; they simply use their evaluations of the product they formed before. Even when they do not find a good fit between the two, they do not use a piecemeal process unless they are highly motivated to do so. It is only when the information is incongruent with the category schema and they are highly involved with the evaluation situation that they engage in attribute-based piecemeal processing. The study findings have important theoretical and managerial implications, which are discussed below.

Theoretical Implications

First, the study results have an implication for the two categorization models discussed earlier, one by Fiske and her colleagues (Fiske 1982; Fiske and Pavelchak 1986; Fiske and Neuberg 1990) and the other by Brewer (1988). These models focus on either informational characteristics of a target stimulus or motivational aspects of a perceiver as a determinant of different types of cognitive processes underlying overall judgments of objects. The results of this study, however, suggest that both of these factors can influence the judgmental processes. For example, schema incongruity would not automatically elicit a piecemeal process unless it occurs under high involvement conditions. In addition, high involvement by itself would not induce a piecemeal process when the information is schema-congruent. In this sense, the study results provide a more integrative view on evaluation processes in comparison with the previous models.

TABLE 2

PARTIAL CORRELATIONS AS A FUNCTION OF SCHEMA CONGRUITY AND SITUATIONAL INVOLVEMENT

Second, the results have implications for both categorization models and assimilation-contrast theory. Specifically, the possible occurrence of contrast has been overlooked in categorization models while that of piecemeal processes and attribute effects has been relatively neglected in the assimilation-contrast paradigm. The study results indicate that any of the processes and judgmental effects that these two groups of literature have proposed can occur, depending on the informational and motivational factors. In this light, the results suggest that the two groups of literature can be integrated to explain consumer information processing and judgments.

Finally, the findings also have an implication for the consumer behavior literature. One of the most controversial issues in consumer research has been when brand names affect product evaluations and when they do not (e.g., Jacoby, Szybillo, and Busato-Schach 1977; Maheswaran, Mackie, and Chaiken 1992; Peterson and Jolibert 1976). The research findings provide one possible answer to this research question; that is, brand names have a positive influence on product evaluations when product attribute information is congruent with brand category, while it may have a negative impact when attribute information is incongruent and consumers are not involved with the evaluation situation.

Managerial Implications

Marketers make a constant effort to change and improve their current products to survive competition. From a consumer's standpoint, such changes in product characteristics are perceived as "incongruities" with existing beliefs and expectations. Then, the marketers' goal is to shape the incongruities and position the products in their favor. The results of this study have some important strategic implications for the companies with this goal.

Companies launching a product with modified features should use a communication strategy that can elicit a high level of consumer involvement. At the same time, the advertising and promotional campaign should emphasize what has been changed about the product. The rationale underlying this strategy is that the increased level of involvement along with the perceived incongruity would induce consumers to use piecemeal processes, instead of simply relying on some prejudgment or contrast processes.

In practice, however, it would be extremely costly for companies to utilize such massive advertising or promotion tools whenever they introduce a modified product. Thus, companies should take advantage of the image associated with the product before change. Specifically, if their existing product has been enjoying a favorable image, they should emphasize that the new, modified product belongs to the same category of the existing one, and thus minimize the perceived incongruity between the existing and the new one. The logic behind this strategy is that the minimized level of incongruity would trigger category-based processing, resulting in affect transfer from the existing to the modified product. On the other hand, if the current product does not have a good image, companies should make every effort to maximize the perceived incongruity between the current and the new product, encouraging contrast processing, or should try to elicit a high level of involvement, inducing piecemeal processing.

Limitations and Suggestions for Future Research

Interpretation of the results should be tempered by recognition of some limitations of the study. First, only one category and one brand name were examined in the analysis. Although the category and the brand name were selected based on a series of carefully designed pretests, future research should examine the generalizability of the current findings to other categories with different category labels and attributes. Second, only correlations were used as dependent measures in the study. Future studies should use process-oriented measures, in addition to outcome measures, to trace cognitive processes underlying judgmental effects. Third, the research framework was built upon an assumption that categorization would occur at the brand name level, when in reality, it can occur at any level of category (Pavelchak 1989). In other words, a product can be categorized into a number of different categories. For example, in the present study, the restaurant could have been categorized not only as a Burger King restaurant but also as a fast-food restaurant, a restaurant in general, and so forth. The assumption would have posed a problem in the study because, if the restaurant, Burger King, had been categorized into a higher level category, e.g., a fast-food restaurant category, the manipulation of schema congruity would not have been effective, and the expected piecemeal processes might not have occurred. Although the manipulation worked in this particular study, the assumption still remains restrictive on the generality of results. Future research should investigate what determines the level of categorization.

REFERENCES

Anderson, Norman H. (1965), "Averaging versus Adding as a Stimulus Combination Rule in Impression Formation," Journal of Experimental Psychology, 70, 394-400.

Anderson, Norman H. (1974), "Information Integration Theory: A Brief Summary," in Contemporary Developments in Mathematical Psychology, Vol.2, eds. David H. Krantz, R. Duncan Luce, Richard C. Atkinson, and Patrick Suppes, San Francisco: W. H. Freeman, 236-305.

Anderson, Norman H. and Michael H. Birnbaum (1976), "Test of an Additive Model of Social Inference," Journal of Personality and Social Psychology, 33, 655-662.

Borgida, Eugene and Beth Howard-Pitney (1983), "Personal Involvement and the Robustness of Perceptual Salience Effects," Journal of Personality and Social Psychology, 45, 560-570.

Boush, David M. and Barbara Loken (1991), "A Process-Tracing Study of Brand Extension Evaluation," Journal of Marketing Research, 28 (February), 16-28.

Brewer, Marilynn B. (1988), "A Dual-Process Model of Impression Formation," in Advances in Social Cognition, eds. Thomas K. Srull and Robert S. Wyer Jr., Hillsdale, NJ: Erlbaum, 1-36.

Celsi, Richard L. and Jerry C. Olson (1988), "The Role of Involvement in Attention and Comprehension Processes," Journal of Consumer Research, 15 (September), 210-224.

Fishbein, Martin and Icek Ajzen (1975), Belief, Attitude, Intention and Behavior: An Introduction to Theory and Research, Reading, MA: Addison-Wesley.

Fiske, Susan T. (1982), "Schema-triggered Affect: Applications to Social Perception," in Affect and Cognition: The 17th Annual Carnegie Symposium on Cognition, eds. Margaret S. Clark and Susan T. Fiske, Hillsdale, NJ: Erlbaum, 55-78.

Fiske, Susan T. and Steven L. Neuberg (1990), "A Continuum of Impression Formation, From Category-based to Individuating Processes: Influence of Information and Motivation on Attention and Interpretation," in Advances in Experimental Social Psychology, Vol. 23, eds. Mark P. Zanna, San Diego, CA: Academic Press, 1-74.

Fiske, Susan T., Steven L. Neuberg, Ann E. Beattie, and Sandra J. Milberg (1987), "Category-based and Attribute-based Reactions to Others: Some Informational Conditions of Stereotyping and Individuating Process," Journal of Experimental Social Psychology, 23, 399-427.

Fiske, Susan T. and Mark A. Pavelchak (1986), "Category-based versus Piecemeal-based Affective Responses: Developments in Schema-Triggered Affect," in The Handbook of Motivation and Cognition: Foundations of Social Behavior, eds. Richard M. Sorrentino and E. Tory Higgins, New York: Guilford Press, 167-203.

Fiske, Susan T. and Shelly E. Taylor (1991), Social Cognition, Reading, MA: Addison-Wesley.

Herr, Paul M. (1986), "Consequences of Priming: Judgment and Behavior," Journal of Personality and Social Psychology, 51 (December), 1106-1115.

Herr, Paul M. (1989), "Priming Price: Prior Knowledge and Context Effects," Journal of Consumer Research, 16 (June), 67-75.

Herr, Paul M., Steven J. Sherman, and Russell H. Fazio (1983), "On the Consequences of Priming: Assimilation and Contrast Effects," The Journal of Experimental Social Psychology, 19 (July), 323-340.

Howard-Pitney, Beth, Eugene Borgida, and Allen M. Omoto (1986), "Personal Involvement: An Examination of Processing Differences," Social Cognition, 4 (1), 39-57.

Jacoby, Jacob, Olson, Jerry C., and Busato-Schach, J. (1977), "Information Acquisition Behavior in Brand Choice Situations," Journal of Consumer Research, 3, 209-216.

Maheswaran, Durairaj, Diane M. Mackie, and Shelly Chaiken (1992), "Brand Name as a Heuristic Cue: The Effects of Task Importance and Expectancy Confirmation on Consumer Judgments," Journal of Consumer Psychology, 1 (4), 317-336.

Martin, Leonard L. (1986), "Set/Reset: Use and Disuse of Concepts in Impression Formation," Journal of Personality and Social Psychology," 51 (3), 493-504.

Martin, Leonard L., John J. Seta, and Rick A. Crelia (1990), "Assimilation and Contrast as a Function of People's Willingness and Ability to Expend Effort in Forming an Impression," Journal of Personality and Social Psychology, 59 (1), 27-37.

Mervis, Carolyn B. and Eleanor Rosch (1981), "Categorization of Natural Objects," Annual Review of Psychology, 32, 89-115.

Meyers-Levy, Joan and Alice M. Tybout (1989), "Schema Congruity as a Basis for Product Evaluation," Journal of Consumer Research, 16 (June), 39-53.

Omoto, Allen M. and Eugene Borgida (1988), "Guess Who Might Be Coming to Dinner?: Personal Involvement and Racial Stereotyping," Journal of Experimental Social Psychology, 24, 571-593.

Pavelchak, Mark A. (1989), "Piecemeal and Category-based Evaluation: An Idiographic Analysis," Journal of Personality and Social Psychology, 56, 3, 354-363.

Peterson, Robert A., and Jolibert, Alain J. P. (1976), "A Cross-national Investigation of Price and Brand as Determinants of Product Quality," Journal of Applied Psychology, 61, 533-536.

Rao, Akshay R. and Kent B. Monroe (1989), "The Effect of Price, Brand Name, and Store Name on Buyers' Perceptions of Product Quality: An Integrative Review," Journal of Marketing Research, 26 (August), 351-357.

Rosch, Eleanor (1978), "Principles of Categorization," in Cognition and Categorization, eds. Eleanor Rosch and B. B. Lloyd, Hillsdale, NJ: Erlbaum, 27-48.

Sherif, Carolyn, Muzafur Sherif, and R. E. Nebergall (1965), Attitude and Attitude Change: The Social Judgment-Involvement Approach, Philadelphia, PA: Saunders.

Sherif, Muzafur, Carl D. Hovland (1961), Social Judgment: Assimilation and Contrast Effects in Communication and Attitude Change, New Haven, CT: Yale University Press.

Smith, Edward E. and Douglas L. Medin (1981), Categories and Concepts, Cambridge, MA: Harvard University Press.

Sujan, Mita (1985), "Consumer Knowledge: Effect on Evaluation Strategies Mediating Consumer Judgment," Journal of Consumer Research, 12 (June), 31-46.

Sujan, Mita and James Bettman (1989), "The Effect of Brand Positioning Strategies on Consumers' Brand and Category Perceptions: Some Insights from Schema Research," Journal of Marketing Research, 26 (November), 454-467.

Sujan, Mita, James Bettman, and Harish Sujan (1986), "Effects of Consumer Expectations on Information Processing in Selling Encounters," Journal of Marketing Research, 23 (November), 346-353.

Wilson, Timothy D., Douglas J. Lisle, Dolores Kraft, and Christopher G. Wetzel (1989), "Preferences as Expectation-Driven Inferences: Effects of Affective Experience," Journal of Personality and Social Psychology, 56, 4, 519-530.

Wyer, Robert S., Jr. and Thomas K. Srull (1989), Memory and Cognition in Its Social Context, Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.

Zeithaml, Valarie A. (1988), "Consumer Perceptions of Price, Quality, and Value: A Means-End Model and Synthesis of Evidence," Journal of Marketing, 52 (July), 2-22.

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