Presenting Product Improvements Using Verbal and Comparative Communication Formats in Packaging

ABSTRACT - We investigated the effects of three marketing communication variables on consumers' attitude toward improved consumer durables. These variables, feature versus benefit communication, verbal/visual versus visual communication and communication with or without a frame of reference of the product improvement, were executed by manipulating packages of small consumer durables. In experiment 1 (experiment 2) packages of a car radio (a coffee maker) were used.

Citation:

Henry S.J. Robben and Jan P.L. Schoormans (1995) ,"Presenting Product Improvements Using Verbal and Comparative Communication Formats in Packaging", in E - European Advances in Consumer Research Volume 2, eds. Flemming Hansen, Provo, UT : Association for Consumer Research, Pages: 138-143.

European Advances in Consumer Research Volume 2, 1995      Pages 138-143

PRESENTING PRODUCT IMPROVEMENTS USING VERBAL AND COMPARATIVE COMMUNICATION FORMATS IN PACKAGING

Henry S.J. Robben, Delft University of Technology

Jan P.L. Schoormans, Delft University of Technology

[We would like to thank Niki van Houtert and Marlies Wilms Floet for their help in collecting the data.]

ABSTRACT -

We investigated the effects of three marketing communication variables on consumers' attitude toward improved consumer durables. These variables, feature versus benefit communication, verbal/visual versus visual communication and communication with or without a frame of reference of the product improvement, were executed by manipulating packages of small consumer durables. In experiment 1 (experiment 2) packages of a car radio (a coffee maker) were used.

Presenting feature information on the product improvement as opposed to benefit information yielded a significant effect for the car radio but not for the coffee maker, however in the opposite direction. An explanation is provided.

Two other hypotheses were confirmed: A verbal/visual information format proved superior to a visual format, and presenting the improvement with a reference enhanced consumers' attitude toward improved products.

Our findings are relevant for the marketing communication strategies used to introduce improved consumer durables.

BACKGROUND

Consumer markets face a constant stream of new product introductions. Few new products are really 'new': A survey by Booz, Allen, and Hamilton (1982) showed that only 10% of new product introductions are truly innovative or new-to-the-world products. Miller (1993) reports that in 1992, over 16,000 new products were introduced in the United States by consumer packaged goods manufacturers. Of those, true innovations concerning formulation, positioning, packaging, or technology amounted to less than 6 percent.

The introduction of most new products demands a communication campaign to inform the target audience about their existence and the consumer benefits they can provide. Communicating these improvements through mass media channels involves high costs. Nevertheless, these strategies are used to introduce changes in nondurables. Manufacturers typically employ mass media channels to communicate minor improvements in consumer durables to a lesser degree.

In existing markets, many purchase decisions are made in-store, and it seems obvious that to stimulate one's brand or product, the communication focus should not only be on advertising, but also on in-store information processing. Vartan and Rosenfeld (1987) reported that more than 80% of all purchase decisions are made in-store, and Bucklin and Lattin (1991) estimated that two-thirds of the American shoppers make their purchase decision in the store. In-store purchase decisions, both for nondurables and durables, are influenced by communication sources like salespersons, sales promotions, point-of-purchase displays and the product's packaging (Wilkie 1994).

PACKAGING AND CONSUMER BEHAVIOR

When consumers search for and process information in-store, the product's package can contain relevant and useful information for the consumer. Product packaging forms the end of the 'promotion-chain' and is close in time to the actual purchase and may therefore play an important role in predicting consumer outcomes. Packaging has to perform at the point of purchase by attracting attention and through classical conditioning principles, consumers may grow to like the package after they enjoyed the product (Dickson 1994). Especially in low involvement product categories and mature markets are choices made primarily based on recognition responses (Rossiter and Percy 1978), and these are likely to be cued by product packaging.

Packages also deliver brand identification and label information like usage instructions, contents, list of ingredients or raw materials, warnings for use and directives for care of the product (Peter and Olson 1990). For small consumer durables, they frequently display the product and thus can show what has been improved in the product and to what customer benefit the improvement will lead. The present paper addresses the question how the product's packaging design can effectively communicate the product's improvement of small durables to the consumer. We assume that these improvements are inherently valuable to the consumer. The literature suggests that three variables are of particular interest in influencing the information conveyed through and on the package of consumer goods.

Consumer Benefits versus Product Features. The package may emphasize the improved feature or the benefit of the improvement to the consumer. There is empirical evidence suggesting that presenting consumers with the product's benefits generates more positive consumer responses than presenting the product's feature improvement (Corfman 1991). This effect occurs especially when the presentation enhances consumers' ability to imagine themselves using the product or experiencing the improved product (cf. MacInnis and Price 1987). Our first hypothesis is therefore

H1: Emphasizing the product's benefit to the consumer, along with the product itself, enhances consumers' attitude toward the improved product more than emphasizing the product's improved feature.

Verbal/visual versus Visual Presentation. Information on virtual any consumer good package contains visual and verbal stimuli to attract attention and to provide persuasive information for consumers in their search processes. The research literature on the effects of visual and verbal communications (or modality of communication) on consumer outcomes is inconclusive. Pictorial information would be more memorable (Rossiter and Percy 1983, Starch 1966) and more persuasive (Mitchel and Olson 1981, Rossiter and Percy 1978). Three factors relate to this visual dominance effect: The attention arousing character of visual stimuli, the level of information processing, and the ability to make inferences about the new functions of the product from the contents of the communication. Visual dominance is often due to the attention arousing effect of visual stimuli and leads to more favorable product judgments when no other information is available (Costley and Brucks 1992). However, in their review, Taylor and Thompson (1982) report no empirical support for the idea that visual presentation is superior to verbal presentation. Childers and Houston (1984) found a superior effect of visual presentation only when the information was processed at a shallow, sensoric level. Shallow, as opposed to deep processing, occurs in realistic decision situations, especially for fast-moving consumer goods.

The effects of visual and verbal information presentation may also differ between product categories of varying functional nature. In their study on sweaters, Holbrook and Moore (1981) found that subjects evaluated pictorial information on more product attributes than verbal information. In contrast, Domzal and Unger (1985) found that pictorial information on watches was evaluated on fewer attributes and attribute interaction than verbal information. The latter authors explain this difference by assuming that watches are more functional in nature, and sweaters more emotional or aesthetic. It may be that consumers need only verbal information when evaluating the advantages of functional products like watches, but that they would need pictorial information when evaluating emotional and aesthetic features of clothing.

We hypothesize that for the consumer durable products in this study, where the improvements have functional qualities, combining verbal and visual information on the improvement facilitates inference making about the value of a product improvement as opposed to presenting only visual information. More formally stated, the second hypothesis states that:

H2: Verbal and visual information on the improved feature will enhance consumers' attitude toward the improved product more than only visual information on the improved feature.

Frame of Reference. If the new product introduction concerns a product improvement, the new version can be compared to the previous version of the product. For these new products, consumers already have a point of reference (cf. Rowe and Puto 1987) that helps in aiding the product's positioning (Wilkie and Farris 1975). This idea capitalizes on the assumption that consumers work with a consideration set in which the previous version of the product is present. Gorn and Weinberg (1984) conclude that comparative advertising influenced perceptions of similarity between products. By presenting the old and the improved product simultaneously, consumers have an explicit point of reference. This reference point provides them with a decision criterion, as the improvement is the only difference between the old and improved versions. The reference condition should prove superior in highlighting the improvement of the product. This reasoning leads to our third hypothesis:

H3: Presenting a frame of reference enhances consumers' attitude toward the improved product more than not providing a frame of reference.

We tested these three hypotheses by executing two experiments that systematically varied visual and visual/verbal presentation of the improved product, frame of reference, and feature/benefit emphasis of the improvement. In experiment 1 we study the effects of the emphasis on feature versus benefit improvement and of the modality of communication by testing hypotheses 1 and 2. In experiment 2 we study the effects of feature versus benefit emphasis of the product improvement and of the presence of a reference frame in the communication by testing hypotheses 1 and 3. These experiments employed the same subjects and used a nearly identical procedure. The experimental stimuli used form the main difference between the two experiments.

METHOD

Materials

In the two experiments four different product packages were used. In experiment 1 (experiment 2) we worked with car radio (coffee maker) packages. Three reasons guided this decision process. First, to design product packages and use them in the experiments the product packages had to be small. Secondly, because of our objective to investigate the communication of product improvements we needed products belonging to categories with relatively new, but not totally new, consumer-valued features. The removable front of the car radio and the high-speed improvement of a coffee maker were such features at the time of the study. The removable front of a car radio can be taken along when leaving the car, thus diminishing the value of the car radio for possible thieves. High-speed is a coffee maker feature that halves the time needed to brew coffee. The third reason to choose the products was that the selected product improvement could be communicated verbally and visually and in terms of product features and product benefits.

The stimuli designed for experiments 1 and 2 are, like real packages, three-dimensional boxes that show a car radio (coffee maker) in front. The packages show the product improvement, removable front or high-speed. However, the stimuli differ in the way the product improvement is communicated. Experiment 1 communicates the improvement visually or verbally/visually and as a feature or benefit improvement. This design has led to four different stimuli: package 1: verbal/visual and feature ("this car radio contains a removable front"), package 2: verbal/visual and benefit ("this car radio is secured against theft"), package 3: visual and feature ("picture of a hand that holds the removable front"), package 4: visual and benefit ("picture of a smashed car window with a red cross over it suggesting that car burglary is indirectly prevented by the removable front of the car radio").

Experiment 2 emphasizes the feature or the benefit improvement of the product, and with or without a frame of reference. This led to four packages communicating the product improvement high-speed on the front. These stimuli are: package 1: feature without reference ("the words high-speed are shown on the coffee maker"), package 2: benefit without reference ("small clock suggesting a 4-minutes coffee making time"), package 3: feature with reference ("two coffee makers, the old model and the redesigned model showing the text 'high-speed'"), package 4: benefit with reference ("stimulus 3, extended with two small clocks indicating the old (8 minutes) and the new time (4 minutes) needed to brew coffee").

Pilot Study

In a pilot study (nonstudent sample; N=20, ages between 20 and 57) we studied whether the stimuli communicated the messages intended. Each subject was shown two pictures, one of the car radio and one of the coffee maker packages, and two control pictures showing the package of an alarm clock and of an iron. The stimuli were randomly assigned to the subjects. Next, the subjects wrote down the meaning of the product improvement shown on the package. For all four car radio packages, the participants understood the new feature. The pilot led to a slight alteration in the degree of darkness of the visual communication in the visual and benefit stimulus.

Furthermore, the results showed that the text high-speed to communicate the coffee maker improvement was understood as was the small clock used to communicate the new product benefit. Next, the results showed that the time used to indicate the "old" time needed to make coffee corresponded with the time estimated by the subjects to brew coffee (Mean=8.5 minutes). Subjects had a problem with the coffee maker packages that showed both the old and the improved coffee makers. They indicated that they were uncertain which of the two coffee makers was in the package. A text reading "this model in the package" was added below the improved coffee maker on stimuli 3 and 4.

Design of the Experiments

Both experiments employed a 2 x 2 between-subjects design to assess the effects of the independent variables on the attitude toward the improved product (ATTimp). ATTimp was assessed with a six-item semantic differential-type scale.

In experiment 1 the first independent variable is the modality in which the product improvement is communicated. This variable has two conditions: a verbal/visual condition and a visual condition. The second independent variable is the emphasis on feature or benefit improvement.

In experiment 2 the first independent variable is information comparison. This variable has two conditions: no frame of reference (only the new product was shown) and a frame of reference condition (the new and the old product were shown). The second independent variable is the emphasis on feature or benefit improvement.

To assess whether style of processing, an individual difference variable, influenced the effects of the independent variables in the criterion measure, we incorporated a Dutch version of the style of processing scale (Childers, Houston and Heckler, 1985) as a covariate in both experiments.

Subjects

A total of 111 subjects (54 males, 57 females) took part in the investigation. Each subject participated in both experiments. They were randomly assigned to the four experimental cells. A subject's classification on the emphasis variable in the first experiment was unrelated to that in the second experiment. All subjects were nonstudent members of the Delft University of Technology Consumer Research panel. Ages ranged from 22 years to 60 years (Mean=38). Education level ranged from primary school level up to university level; about 45% of the sample attained a higher professional degree or college education, indicating that the sample had an above average education. The subjects received travel expenses and a small gift (value of approximately $3) for their cooperation.

Procedure

Subjects completed a questionnaire containing socio-demografic questions and the style-of-processing scale. Subsequently they viewed four packages, each for about 45 seconds. We controlled for the visual dominance effect (Costley and Brucks 1992) by presenting the stimuli for a longer period, a procedure that mimics the situation consumers would experience when shopping for consumer durables. The four packages were two control stimuli (package of an alarm clock and of an iron) and one of the four randomly selected experimental stimuli, for each experiment. The order of presentation of the stimuli was randomized.

Subjects then completed the second questionnaire to assess the subjective meaning of the product improvements shown on the experimental stimuli. Finally, for experiment 1 (experiment 2) the ATTimp for the car radio (coffee maker) was assessed and the subjects were debriefed subsequently.

RESULTS AND DISCUSSION

Results and Discussion of Experiment 1

Attitude toward the improved product. Subjects indicated their evaluation of each product separately on six semantic differential format items (negative-positive, pleasurable-unpleasurable, boring-interesting, bad-good, irritating-pleasant, repulsive-attractive). Principal components analysis suggested that these items are summarized by a single factor (l=4.4, 73% of the variance explained). This finding warranted the construction of an attitude score based on the six evaluations (n_i=6, a=.92). A higher score on this attitude indicates a more positive evaluation of the improved product; scores ranged from -18 to +18.

Style of processing scale. A principal components analysis of the 22 items of the Childers et al. (1985) style of processing scale revealed eight factors dividing the items into verbal and visual subsets. A new principal components analysis with forced extraction of two factors of 11 items each, accounted for only 29% of the variance (l_verbal=3.6, 16%, a_verbal=.72; l_visual=2.8, 13%, a_visual=.66). Subjects were assigned the summated score of the 11 items of each factor of the style of processing scale.

Subjective experiences and manipulation checks. Subjects indicated on 4-point scales their perception of the meaning of the removable car radio front. A majority (88.2%) gave the correct answer, and they were quite certain about this answer (Mean=4.2; "1" is "very uncertain," "5" is "very certain"). A similarly large proportion (84%) indicated on an open-ended question the intended benefit for the customer of using the removable car radio front. These results show that the subjects knew (or had learned) about the feature and benefit of the removable front.

To further assess the extent to which the subjects perceived the experimental manipulations as intended, they completed several rating scales. Subjects indicated the extent to which the presented information in the car radio experiment was mainly visual ("1" on a 5-point scale) or verbal ("5"). Subjects in the verbal/visual condition agreed more often that the presented information was more verbal (Mean=2.2) than those in the visual condition (Mean=1.9). This difference approached statistical significance in the expected direction (t=1.6, df=106, p=.053). This effect was significant for those subjects who said that they had perceived the improved feature of the car radio (Mean_{verbal/visual}=2.3, Mean_visual=1.9, t=1.8, df=75, p=.037).

Subjects indicated on a dichotomous scale whether the information on the package emphasized the product's features or the product's benefits for the customer. A "0" represented an emphasis on the product's features and a "1" that customer benefits were stressed. In the feature condition subjects indicated that the information emphasized the product's features (Mean=0.2); in the benefit condition subjects indicated that the package stressed customer benefits (Mean=0.5). This difference was significant and in the expected direction (t=4.2, df=100.2, p=.0001).

Summarizing, for all manipulations there is strong evidence that the subjects processed the information in the experimental conditions as intended.

Analysis of variance. Before testing the hypotheses, a check based on the second questionnaire revealed that not all subjects had perceived the product's improvement. The analyses of variance reported below only contained the responses of those subjects who reported to have seen the removable car radio front on the stimulus (N_seen=78; N_notseen=32).

The ATTimp was subjected to a two (emphasis: product feature or product benefit) by two (modality of presentation: verbal/visual or visual) analysis of variance with the verbal and visual factors of the style of processing variable entered as covariates. The style of processing subscales did not have significant effects, suggesting that differences in style of processing were not meaningfully related to ATTimp (F_verbal(1,67)=1.9, p>.10; F_visual(1,67)=0.8, p>.10).

TABLE 1

MEAN ATTITUDE TOWARD THE IMPROVED PRODUCT AS A FUNCTION OF IMPROVEMENT EMPHASIS, MODALITY OF COMMUNICATION, AND INFORMATION COMPARISON

The effect for the emphasis variable approached statistical significance (F(1,67)=3.5, p<.064). This result was contrary to what we expected: Communicating the improvement on the feature level (Mean=6.9) generated a more positive attitude toward the improved product than on a benefit level (Mean=4.5) (See Table 1). This finding rejects hypothesis 1 and supports the opposite prediction. Thus, information presented on the feature level generated a more positive ATTimp than information on the benefit level. The product improvement was presented either as the removable front that car drivers can easily carry with them or as an antitheft mechanism. Although the removable front feature clearly satisfies both operationalizations, the second one may have evoked less positive attitudes. It may have hinted at an unpleasant consumer experience, namely theft of the car radio with damage to the car window. Here, the correct inference was induced with a more or less inappropriate picture, i.c. by using a negative claim (see, e.g., Engel, Blackwell, and Miniard 1995).

The analysis revealed a significant main effect for the modality of presentation variable (F(1,67)=6.9, p<.011). This result was in the expected direction and supported hypothesis 2: Communicating the improvement verbally/visually (Mean=7.2) was more successful than only communicating visually (Mean=4.0). This finding suggests that when communicating product improvements for a consumer durable it is better to tell the consumer what the improvement represents in terms of added value to the consumer. Merely showing the improvement with a picture is probably inadequate to make the product improvement understood as such and consequently to create a favorable attitude toward the improved product.

An additional analysis of variance without the covariates did not yield appreciable differences in the effects of both independent variables.

Results and Discussion of Experiment 2

Attitude toward the improved product. As in experiment 1, subjects indicated their evaluation of each product separately on six semantic differential format items. Principal components analysis suggested that these items are summarized by a single factor (l=4.5, 74% of the variance explained). This finding warranted again the construction of an attitude score based on the six evaluations (n_i=6, a=.92; a higher score indicates a more positive evaluation).

Subjective experiences and manipulation checks. Subjects indicated on 4-point scales their perception of the meaning of the high-speed feature. A majority (93%) gave the correct answer, and they were quite certain about this answer (Mean=4.1; with a "1" indicating "very uncertain," and a "5" "very certain"). A similarly large proportion (88.2%) indicated on an open-ended question the intended benefit for the customer of using the high-speed feature. These results show that the subjects knew (or had learned during the experiment) about the function and benefit of the improved coffee maker.

To assess the effect of presenting a frame of reference, subjects indicated on 5-point scales whether they agreed that the new 1993 model would outperform the 1992 model concerning brewing coffee faster ("5") or that the 1992 model would be faster ("1"). The mean value in the frame of reference condition (Mean=3.8) was significantly larger than that in the no frame of reference condition (Mean=2.2). This difference was highly significant and in the expected direction (t=7.0, df=101.3, p=.0001).

Subjects indicated on a dichotomous scale whether the information on the package emphasized the product's features or the product's benefits for the customer. There was no significantly different perception between the feature and the benefit communication conditions. In the feature condition and in the benefit condition subjects indicated that the package stressed the product's features and customer benefits about equally (Means of 0.46 and 0.48, respectively). This difference was insignificant (t=0.3, df=107, p>.10), even for those subjects who indicated that they recognized the product's enhancement (Means of 0.4 and 0.5, respectively; t=0.7, df=68, p>.10).

Summarizing, these findings indicate that the frame of reference manipulation worked as intended, but that the feature/benefit emphasis manipulation probably did not.

The ATTimp was subjected to a two (emphasis: feature or benefit improvement) by two (information comparison: without reference or with reference) analysis of variance with the verbal and visual subscales of the style of processing scale as covariates. There was a significant main effect for the verbal scale but not for the visual scale (F_verbal(1,57)=6.0, p<.05; F_visual(1,57)=0.2, p>.10).

The emphasis on feature versus benefit improvement did not affect the attitude toward the coffee maker (F(1,57)=0.0, p>.10; see Table 1 for the relevant means). This result rejects our first hypothesis, that information on the benefit level would generate a more positive attitude toward the new product than information on the feature level. Presumably, communicating the high-speed coffee making feature represented the very same thing as the consequence: putting coffee faster on the table.

The main effect of the variable information comparison approached statistical significance (F(1,57)=3.6, p<.062), indicating a more positive attitude toward the improved coffee maker in the presence of a frame of reference than without a frame of reference (see the bottom half of Table 1). In the frame of reference condition, the improved coffee maker generated a more favorable attitude than in the no frame of reference condition. The mere comparison highlighted that the new model coffee maker had been improved over a previous model. This finding supports hypothesis 3, which stated that presenting consumer with a frame of reference would enhance consumers' attitude toward the new improvement. Running the analysis of variance without the covariates showed a stronger and significant effect of the comparison variable (F(1,59)=5.2, p<.05).

CONCLUSIONS AND IMPLICATIONS

We developed and tested three hypotheses on the way to express product improvements through packaging in two experiments. The first hypothesis on the effect of the emphasis on feature versus benefit improvement was tested in both experiments. It stated that emphasizing a product's benefit would create a more positive attitude toward the improved product than emphasizing the improved feature. Unexpectedly, the car radio experiment (experiment 1) yielded results that confirmed the opposite prediction: Communicating the product's improved feature enhanced consumers' attitude toward the improved product. The experiment on the coffee maker (experiment 2) did not support the hypothesis.

These findings suggest that when communicating that a product has been improved, it is better to present concrete information, i.e., to highlight the improved feature, instead of emphasizing abstract information, i.e., to express the product's benefit to the consumer. The research literature on features and benefits shows that consumers are interested in product features only because of the benefits that come with these features (e.g., Haley 1968). From this idea it followed that a communication strategy that highlights this benefit by, for instance, enhancing consumers' ability to imagine themselves using the product or experiencing the benefit would be superior to a strategy that emphasized the improved feature. Our results suggest the opposite has happened.

A possible explanation of this result departs from a consumer learning point of view. We agree with Johnson (1989) that consumers do not relate product or concrete features to benefits or abstract features unambiguously. In addition, consumers have to learn this relationship, a process that takes time. One strategy that may guide this learning process is "hands-on" experience: consumers learn the benefits of a product by using it. It is possible that in the car radio experiment (experiment 1), the subjects were not familiar enough with the possible benefits of a removable front and thus could not connect the benefit to the improved product feature. If this argumentation is correct, it is advisable to use a communication strategy that emphasizes benefits only if consumers can connect these benefits to the associated product features. In the coffee maker experiment (experiment 2), the difference between the improved feature and the resulting benefit may have been too smallCthe high-speed improvement probably implied the benefit, namely brewing coffee fasterCto induce differences in the attitude toward the improved product. This reasoning indicates that in those situations where a direct and clear connection is possible between product features and the associated benefits, there is no reason to expect significant differences in the effects of the communication strategy.

In the car radio experiment (experiment 1) we tested the hypothesis that verbal/visual communication of the improvement would lead to a more positive attitude toward the improved product as compared to visual information only. The data supported this hypothesis, and this suggests the superiority of a verbal/visual information presentation format over a visual format in creating a more favorable attitude. The reason behind this difference is, we think, related to the enhancement of consumers' ability to infer the value of the improved product feature (cf. MacInnis, Moorman, and Jaworski 1991). This ability is higher in verbal/visual communication situations because of the direct and concrete nature of natural language. We therefore suggest communicating product improvements by using a verbal/visual information presentation format, supported by visual information, to create attention for the information on the product's improvement.

Providing consumers with an explicit frame of reference enhanced their attitude toward the improved product. Providing a reference point made it easier to detect and evaluate the product's improvement. The improvement was essential to the product's main function, namely brewing coffee. This finding is consistent with Sanbonmatsu, Kardes, and Gibson (1991), who report that features that are unique to the compared alternative, in our case the high-speed improvement, are highlighted by simply comparing the alternative to another product, here a previous version of the coffee maker.

The relevance of these findings is found in the temporal character of communication strategies. In the introduction phase of totally new products or of products that are functionally improved, marketing communications primarily should inform consumers of the existence and contents of the product improvement. In this phase, presenting information verbally and visually instead of only visually may be the superior strategy. Another potentially valuable communication strategy is presenting the improved product in a reference frame that positively highlights the improvement. After the introduction phase, when the improved feature has become a market standard, communication strategies can aim at persuading consumers. A longitudinal study, incorporating the introduction of the improved product and the point in time where the improvement has become a standard, is needed to investigate the difference in value between the proposed communication strategies and traditional ones that primarily focus on persuading consumers to use new products.

REFERENCES

Booz, Allen, and Hamilton, 1982, New Product Management for the 1980's, New York: Booz, Allen, and Hamilton, Inc.

Bucklin, R.E. and J.M. Lattin, 1991, "A Two-State Model of Purchase Incidence and Brand Choice," Marketing Science, 10, 24-39.

Childers, T.L. and M.J. Houston, 1984, "Conditions for a Picture-superiority Effect on Consumer Memory," Journal of Consumer Research, 11, 643-654.

Childers, T.L., M.J. Houston, and S.B. Heckler, 1985, "Measurement of Individual Differences in Visual Versus Verbal Information Processing," Journal of Consumer Research, 12, 125-134.

Corfman, K.P., 1991, "Comparability and Comparison Levels Used in Choices Among Consumer Products," Journal of Marketing Research, 28, 368-374.

Costley, C.L. and M. Brucks, 1992, "Selective Recall and Information Use in Consumer Preferences," Journal of Consumer Research, 18, 464-474.

Dickson, P.R., 1994, Marketing Management, Forth Worth, TX: The Dryden Press.

Domzal, T.J. and L.S. Unger, 1985, "Judgments of Verbal Versus Pictorial Presentations of a Product with Functional and Aesthetic Features," in Advances in Consumer Research, Vol. 12, eds. E.C. Hirschman and M.B. Holbrook, Ann Arbor, MI: Association for Consumer Research.

Engel, J.T., R.D. Blackwell, and Paul W. Miniard, 1995, Consumer Behavior, eighth edition, Chicago, IL: The Dryden Press.

Gorn, G.J. and C.B. Weinberg, 1984, "The Impact of Comparative Advertising on Perceptions and Attitude: Some Positive Findings," Journal of Consumer Research, 11, 719-727.

Haley, R.I., 1968, "Benefit Segmentation: A Decision-oriented Research Tool," Journal of Marketing, 32, 30-35.

Holbrook, M.B. and W.L. Moore, 1981, "Feature Interactions in Consumer Judgments of Verbal Versus Pictorial Presentations," Journal of Consumer Research, 8, 103-113.

Johnson, M.D., 1989, "On the Nature of Product Attributes and Attribute Relationships," in Advances in Consumer Research, Vol. 16, ed. T.K.-Skrull, Ann Arbor, MI: Association for Consumer Research.

MacInnis, D.J. and L.L. Price, 1987, "The Role of Imagery in Information Processing," Journal of Consumer Research, 13, 473-491.

MacInnis, D.J., C. Moorman and B.J. Jaworski, 1991, "Enhancing and Measuring Consumers' Motivation, Opportunity, and Ability to Process Brand Information From Ads", Journal of Marketing, 55, 32-53.

Miller, C., 1993. "Little Relief Seen for New Product Failure Rate," Marketing News, 27 (13), pp. 1,10.

Mitchell, A.A. and J.C. Olson, 1981, "Are Product Attribute Beliefs the Only Mediator of Advertising Effects on Brand Attitude?," Journal of Marketing Research, 18, 318-332.

Peter, J.P. and J.C. Olson, 1990, Consumer Behavior and Marketing Strategy, 2nd edition, Homewood, IL: Irwin.

Rossiter, J.R. and L. Percy, 1978, "Visual Imaging Ability as a Mediator of Advertising Response," in Advances in Consumer Research, Vol. 5, ed. H. Heith Hunt, Ann Arbor, MI: Association for Consumer Research, 621-629.

Rossiter, J.R. and L. Percy, 1983, "Visual Communication in Advertising," in Information Processing Research in Advertising, ed. R.J. Harris, Hillsdale, NJ: Lawrence Erlbaum Associates.

Rowe, D. and C.P. Puto, 1987, "Do Consumers' Reference Points Affect Their Buying Decisions?," in Advances in Consumer Research, Vol. 14, eds. M. Wallendorf and Paul Anderson, Ann Arbor, MI: Association for Consumer Research, 188-192.

Sanbonmatsu, D.M., F.R. Kardes, and B.D. Gibson, 1991, "The Role of Attribute Knowledge and Overall Evaluations in Comparative Judgment," Organizational Behavior and Human Decision Processes, 48, 131-146.

Starch, D., 1966/1967, "How Does the Shape of Ads Affect Readership," Media/Scope, 10, 83-85.

Taylor, S.E. and S.C. Thompson, 1982, "Stalking the Elusive 'Vividness' Effect," Psychological Review, 89, 155-181.

Vartan, C.G. and J. Rosenfeld, 1987, "Winning the Supermarket Wars: Packaging as a Weapon," Marketing Communications, (August), 33.

Wilkie, W.L., 1994. Consumer Behavior, 3rd edition. New York: John Wiley & Sons, Inc.

Wilkie, W.L. and P.W. Farris, 1975, "Comparison Advertising: Problems and Potential," Journal of Marketing, 39, 7-15.

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