The Effects of Knowledge and Imagery on Advertising Responses to an Innovation

Kathleen Debevec, University of Massachusetts/Amherst
Patricia W. Meyers, University of Massachusetts/Amherst
Kenny K. Chan, University of Massachusetts/Amherst
ABSTRACT - This study attempts to determine if the verbal and visual imagery effects found in past research with traditional products is replicable with a dynamically continuous innovation among subjects with varying levels of product knowledge. To date, innovation research has largely overlooked the issue of communication effectiveness, especially in the earlier stages of product adoption. Results indicate the importance of providing individuals with information (preferably concrete information) and suggest that risk-reducing information can prove valuable, especially as subjects learn more about the innovation. Findings also address recent research on the efficacy of pictures in ads.
[ to cite ]:
Kathleen Debevec, Patricia W. Meyers, and Kenny K. Chan (1985) ,"The Effects of Knowledge and Imagery on Advertising Responses to an Innovation", in NA - Advances in Consumer Research Volume 12, eds. Elizabeth C. Hirschman and Moris B. Holbrook, Provo, UT : Association for Consumer Research, Pages: 273-278.

Advances in Consumer Research Volume 12, 1985      Pages 273-278

THE EFFECTS OF KNOWLEDGE AND IMAGERY ON ADVERTISING RESPONSES TO AN INNOVATION

Kathleen Debevec, University of Massachusetts/Amherst

Patricia W. Meyers, University of Massachusetts/Amherst

Kenny K. Chan, University of Massachusetts/Amherst

ABSTRACT -

This study attempts to determine if the verbal and visual imagery effects found in past research with traditional products is replicable with a dynamically continuous innovation among subjects with varying levels of product knowledge. To date, innovation research has largely overlooked the issue of communication effectiveness, especially in the earlier stages of product adoption. Results indicate the importance of providing individuals with information (preferably concrete information) and suggest that risk-reducing information can prove valuable, especially as subjects learn more about the innovation. Findings also address recent research on the efficacy of pictures in ads.

INTRODUCTION

Innovation has concerned marketers for some time. Techniques for launching innovations, characteristics of innovators and early adopters, and strategies to improve the innovative, new product performance of business units have all received close scholarly scrutiny (Calantone and Cooper 1981, Jacoby 1971, Midgley 1977, Urban and Hauser 1980). Since these studies have addressed varied types of innovations, it is helpful to employ a taxonomy.

In his widely cited efforts to apply diffusion theory to marketing, Robertson discussed three types of innovation classified along a continuum reflecting how "continuous or discontinuous their effects are on established consumption patterns (Robertson 1971, p. 7). The three categories included continuous, dynamically continuous, and discontinuous innovations. A continuous innovation involves alteration of an existing product and causes little behavior change in adopters. A dynamically continuous innovation involves the creation of a new product or of a significantly modified existing product and has some, but not drastic, effect upon consumption behavior. A discontinuous innovation commercializes an entirely new and different product that represents a new product class involving new technologies and radically different consumption behavior patterns.

In their efforts to explicate the diffusion and adoption processes (Brown 1981; Rogers 1983; Rogers and Shoemaker 1971), sociologists and geographers have focused upon the effects of the more radical forms of innovation, i.e., discontinuous and dynamically continuous ones. Many marketing researchers have focused on aspects involving the least radical or continuous form of innovation (Darden and Reynolds 1974; Kotler and Zaltman 1976; Ostlund 1972). This is to be expected since most new consumer products are of this type (Booz, Allen, Hamilton 1982). With the recent commercialization of technological advances in several fields (for example, birth control pills, interferon-based insulin and other planned products of generic engineering, video games, home computers, programmable calculators, microprocessor controls for automobile engines, etc.), marketing interest in the adoption processes involved in more radical innovations will. in all likelihood. increase.

One important research area regarding more radical innovations is the effect of advertising upon the early stages of consumer awareness. These early stages have been described in a number of adoption process models that are now familiar in the marketing lexicon. [Perhaps most widely known among these are the "adoption process" model resulting from rural sociology diffusion studies in the 1940's and 1950's (Ryan and Gross 1943); the "hierarchy of effects" model (Falda 1966; Lavidge and Steiner 1961); and the "innovation-decision process" model (Rogers 1983, 1971). In addition, Midgley has recently proposed a "new model of innovativeness" (Midgley 1978).] Each of these models begins with one or two stages called "awareness," "interest," or "knowledge." Each also hypothesizes that potential buyers begin the adoption process through receiving and processing information about the innovation. The argument can be made that this initial portion of the adoption process is all the more crucial for marketers of more radical innovations because prior knowledge of the product class is limited or--in the most radical cases--nonexistent. Unfortunately, little is known about the types of advertising that prove most effective under this constraint.

Several recent consumer behavior studies have considered the effect of imagery through pictures and verbal statements upon possible adopters of new products of the least radical, or continuous, innovation types. Using a hypothetical new shampoo brands Kisielius and Sternthal (1984) investigated the effect of various vividness manipulations in at claims upon subjects' judgments of the product. They found that vividness effects upon judgments vary by situation. To explain the cognitive elaboration process that they believe mediates judgment, they proposed an availability-valence hypothesis based on the favorableness of information available in memory. Rossiter and Percy (1980) used high vs. low visual emphasis and explicit vs. implicit verbal claims in ads for a fictitious new beer brand to assess effects on subjects' product attitude and intention to try the product. They found that the combination of high visual emphasis and explicit verbal claims had greater positive effect upon both dependent measures than did other ad conditions. Using ads for fictitious new brands of cars, cameras, and calculators, Edell and Staelin (1983) tested their proposed model to explain differences in consumer responses to pictorial and verbal advertising stimuli. They found that "unframed" pictorial ads (i.e., those in which the information conveyed by the picture was not restated verbally) produced less clear recall of brand items and statements and fewer evaluative thoughts--pro or con--about ad claims than did other conditions. In contrast, "framed" pictorial ads and verbal ads in which objective messages were displayed produced more arguments in support of claims than counter arguments. In addition, no significant differences were found between the framed pictorial and verbal conditions. Edell and Staelin note that this finding differs from that of Kisielius (1982), who found that a sentence-only condition produced more positive brand attitudes than did the pictures and sentences condition.

The limited recent research into the effects of verbal and visual ad presentations for continuous innovations indicates a consensus on the importance of visual and verbal effects in advertising. However, it is also revealed that reported results on their relative importance under varied conditions are still equivocal. It is notable that previous studies have dealt with fictitious new brands in known product classes, i.e., with continuous innovations, and thus have presumed prior knowledge of the product type.

Additionally, at least three recent studies provide evidence that attitude toward the ad, as well as attitude toward the product or brand, is an appropriate measure of ad effectiveness (Lutz, MacKenzie, and Belch 1973; Mitchell and Olson 1981: ShimP 1981).

With this background in mind, it appears that a study of the relative importance of visual and verba messa in ad presentations for a more radical innovation, specifically a dynamically continuous one, would be a contribution to both the innovation/diffusion and the advertising imagery literature.

HYPOTHESES

Our purpose is to investigate the effects of visual emphasis and verbal content upon consumers exposed to information about a dynamically continuous innovation (as defined above). Specifically, we examine the effects of experimentally varied visual ad stimuli (picture emphasis or copy emphasis), verbal ad stimuli (concrete or abstract claims), and levels of knowledge about the product class (little or no knowledge of the innovation or experimentally provided prior knowledge). Our hypotheses are as follows:

H1: Pictorial emphasis in ads for a dynamically continuous innovation will produce more favorable attitudes (i.e., toward the at, the product, and information seeking behavior) than will copy emphasis.

Substantial evidence supports the idea that visual factors can have strong positive effects on memory, learning and attitude toward the brand (Paivio 1971; Paivio and Csapo 1973; Shepard 1967; Lutz and Lutz 1977; Alesandrini 1982). In line with recent studies cited previously, attitude measures will be taken not only upon attitude toward the brand or product, but also on attitudes toward the ad. Because of the newness of the dynamically continuous innovation chosen for the study and the consequent assumption that subjects were all in the early stages of the adoption process, intention measures were confined to attitudes toward information-seeking behavior

H2: Concrete (high imagery) verbal claims in ads for a dynamically continuous innovation will produce more favorable attitudes (i.e., toward the ad, the product, and information seeking behavior) than will abstract (low imagery) verbal claims.

Concrete verbal information appears to have higher imagery value, be better understood, be remembered better, and be more influential in decision making than is more abstract verbal information (Paivio 1969; Paivio et al. 1968; Alesandrini 1982; and Nisbett et al. 1976). Imagery value has been defined here and measured in past studies (Lippman 1974; Paivio et al. 1968) in terms of the degree to which words generate mental images among respondents.

H3: Picture emphasis and concrete verbal claims in ads for a dynamically continuous innovation will produce more favorable attitudes (i.e., toward the ad, the product, and information seeking behavior) than will a copy emphasis ad with abstract claims.

Studies discussed in the previous section, particularly Rossiter and Percy (1980) and Edell and Staelin (1983), lend support to this hypothesized relationship.

H4: In the awareness stage of the adoption process, the more knowledge that consumers have about a dynamically continuous innovation, the more likely they will be to expend energy in further information seeking.

We found no studies in the literature that directly support this hypothesis. However, based upon the several currently employed models of the adoption process, we will expect that in the early stages of the adoption process, as potential buyers of a more radical innovation learn more about the new product type, they would be more willing to seek out further information

H5: The more knowledge that consumers have about the innovation, the less risky they will perceive it to be.

This hypothesis, too, is deduced from the various adoption process models currently in use. As consumers progress through the various stages from awareness to adoption, they would logically have to come to terms with whatever risks the innovation holds for them. A decision to adopt would imply that adopters have come to perceive the innovation as less risk-laden over the adoption process time-frame. Dickerson and Gentry (1983) provide some empirical support for this hypothesis. They found that early adopters of home computers had more experience with and knowledge about them than did nonadopters.

METHODOLOGY

Experimental Manipulations

The hypotheses were tested in a 2x2x2 factorial design in which the knowledge level of subjects (present knowledge versus experimentally-induced knowledge), the verbal content of the advertisements (concrete versus abstract information) and the visual content (picture emphasis versus copy emphasis) were manipulated. The product chosen for the experiment was a compact disc (CD) player, a new revolutionary stereo turntable in which a laser beam picks up digital audio signals from discs. The compact disc was selected because it was thought to be a product in which a college audience would be interested and because of its newness on the market (thus resulting in 2 limited product information among subjects as a whole). [Thirty-five out of sixty of the "no knowledge" group had never heard of a CD player before the experiment. No subject in any condition owned a CD player. Although 25 subjects in the "no knowledge" group initially indicated awareness of the existence of CD players, the experimental knowledge treatment did yield significant attitudinal differences between the groups. Additional inquiries into subjects' knowledge level were omitted so as not to contaminate the knowledge manipulations.] The product was also judged to be a dynamically continuous innovation, one which would require some but not total behavioral change as a result of its purchase.

In the knowledge manipulation, half the subjects viewed the ad and completed the questionnaire using only their present level of knowledge about the product during the experiment (no additional information was provided). Before receiving the questionnaire and ad, the other half of the participants were given a sheet of paper describing the products, its benefits and limitations. They were instructed to read the description and write down all the advantages and disadvantages of the product that they could find or think of. Subjects were allowed ten minutes to complete this task. The purpose of this exercise was to give subjects information that they could later recall from memory and to simulate a knowledge level that they would likely have between the introduction and growth stages of the product life cycle.

The verbal content manipulation resulted in subjects receiving ads which conveyed similar information; however, in half the ads, the information was specific (concrete), while in the other half it was nonspecific (abstract). The information was chosen as a result of interviews with an expert panel composed of an independent stereo store owner, two experienced stereo equipment salespeople, and a stereo buff. The content of the ads was a combination of product attributes that the panel had found to be important among prospective buyers and information provided by manufacturers that described the unique benefits of the product.

Lastly, in the visual content manipulation, the ads were constructed so that three-fourths of the ad was dominated by a picture or verbal copy while the other fourth emphasized the opposite. Rossiter and Percy (1978, 1980) employed a comparable manipulation. The picture used was an actual photograph of the compact disc player supplied to dealers by the manufacturer.

Subjects

Subjects were juniors and seniors recruited from undergraduate business courses. One hundred twenty students participated in the experiment, 15 of whom were randomly assigned to each of the eight treatment conditions such that all cells of the design were evenly balanced. Thus, each subject viewed only one advertisement.

Stimulus Materials

The hypotheses were tested utilizing the four print ads pictured in Figure l. Print was selected as the experimental medium because it is the primary medium in which stereo advertising is found and it is consistent with imagery operationalizations in the recent research cited earlier.

Procedure

Subjects were initially told that they would be viewing an ad for a new product and were asked for their impressions of the ad and of the product. They were instructed that the ad was not in finished form but rather a concept mock-up that an ad agency might produce in the first stages of generating possible ads. Those subjects to be provided with preliminary information about the product were then given the sheet of paper describing the product and were asked to List its benefits and limitations (as indicated earlier). Once this exercise was completed, the subjects were handed the ad and a questionnaire in a packet. Subjects in the limited knowledge conditions received the packet only without the initial information manipulation. Prior to viewing the ad, subjects in all groups were first asked if they had ever heard of a compact disc player before and they were instructed to record "yes" or "no" on the cover page of the packet. The purpose of this question was to determine the general level of awareness about the product in the limited knowledge treatment groups.

Next, all subjects were given one minute to view the ad in their packets, after which they completed the questionnaire with the aid of the experimenter. First, subjects were asked for their overall reactions to the ad on a series of seven-point bipolar adjective scales, which would later be factor analyzed to obtain an overall (composite) attitude toward the ad measure. A second question read, "As best as you can judge, based on this ad, how would you rate this product?" Subjects once again responded on a series of bipolar adjective scales which would also be factor analyzed to obtain a composite attitude toward the product measure. The exercise also included three verbal compliance measures, each reflecting varying levels of further information seeking. In line with the early stages of the most common adoption process models, information seeking appeared to be the most appropriate measure of behavioral intention. Subjects indicated their interest in getting more information about the product and seeing a demonstration of the compact disc player on seven-point semantic differential scales bounded by "highly interested" and "not interested at all." A higher level of information seeking would be indicated by subjects' responses to the question, "How likely are you to go to a stereo store and ask for a demonstration for more information about the compact disc player?" Subjects responded on a seven point scale anchored by "highly likely" and "not likely at all."

RESULTS

To measure subjects' overall "attitude toward the ad" and "attitude toward the product," their responses to each construct were factor analyzed. The primary factor representing subjects' attitude toward the ad (ATTAD) was composed of adjective pairs with loadings > 0.6 which included likeable-unlikable, interesting-uninteresting, entertaining-unentertaining, artful-artless, and good-bad. The factor representing subjects' attitude toward the product (ATTPRDT) was characterized by the adjective pairs: good-bad, superior-inferior, pleasant-unpleasant, and risky-riskless. This latter factor is very similar to the product attitude factor employed by Rossiter and Percy (1978). The remaining items on each scale were dropped from subsequent analyses because of low loadings or lack of factor interpretability. Analysis of variance was used to test the primary hypotheses. The results are summarized in Table l.

From hypothesis l we would predict that ads high in visual content (picture emphasis) would produce more favorable attitudes than ads low in visual content (copy emphasis). Therefore, a PICTURE main effect would be expected in the analysis of variance tables for subjects' attitudes toward the ad, the product, and information seeking behavior.

The results indicate an apparently significant main effect for visual emphasis in subjects' attitude toward the ad (ATTAD), which would partially support this hypothesis (F = 15.078, d.f. = 1, p < .001). The high visual content condition induced a significantly more favorable attitude toward the ad (X = 3.25, S.D. = 1.208) than the low visual content condition (X = 4.14, S.D.= 1.316). However, a significant PICTURE x INFO interaction effect (F = 3.741, d.f. = 1, p = .056) deems the PICTURE main effect tenuous. High visual content in ads was more effective in generating a favorable attitude toward the ad (X = 3.19, S.D. = 1.354) than low visual content (X = 4.53, S.D. 1.183) only when abstract verbal claims were made. No significant effects were obtained for attitude toward the product (ATTPRDT) (F = .34, d.f. = 1, p = .561), interest in getting more information (SEEK) (F = .007, d.f. = 1, p = .935), interest in seeing a demonstration (DEMO) (F - .036, d. f . - l, p = .851), or interest in visiting a stereo store for a demonstration (STORE) (F = .840, d.f. = 1, p = .361) at the p < .05 level.

The second hypothesis predicts that concrete verbal claims will produce more favorable attitudes than will abstract verbal claims. It was tested by examining INFO main effects. The hypothesis was supported for subjects' ATTPRDT (F = 3.807, d.f. = 1, p = .054). Subjects who were exposed to concrete (high imagery) verbal claims reported a more favorable attitude toward the innovation (X = 2.9, S.D. = .868) than those exposed to abstract verbal claims (X = 3.22, S.D. = .929). An INFO main effect for DEMO was marginally significant (F = 2.88, d.f. = 1, p = .092), while the main effect for ATTAD IF = 2.158, d.f. = l, p = .145), SEEK (F = .06, d.f. = 1, p = .806), and STORE (F = 1.049, d.f. = 1, p = .308) were nonsignificant. Thus, while subjects' attitude toward the product was affected, their interest in seeking further information was not.

FIGURE 1

AD COPY VARIATIONS USED IN THE EXPERIMENT: PICTURE VS. COPY DOMINANCE (ROWS) AND CONCRETE VS. ABSTRACT - VERBAL CLAIMS (COLUMNS)

The third hypothesis predicts that ads with high visual content (picture emphasis) and concrete verbal claims will produce more favorable attitudes than will ads dominated by verbal copy with abstract claims. A PICTURE x INFO interaction would be expected for each of the attitudinal measures. The hypothesis was supported in subjects' attitude toward the ad (ATTAD) as seen in Figure 2 (F = 3.74, d.f. = 1, p = .056). Subjects who received the concrete verbal claims and high visual dominance treatment displayed a significantly more positive attitude toward the ad than those who received the abstract verbal claims and low visual dominance treatment (t = 3.59, d.f. = 58, p < .001). Although there was a significant PICTURE x INFO interaction for subjects' interest in getting more information, SEEK (F=4.187, d.f.=1, p<.05) as shown in Figure 3, there was no significant difference in means between the picture dominant, concrete verbal claim ad and the copy dominant abstract claim ad (X=2.17 and X=2.13, respectively). A test of simple main effects also revealed that subjects exposed to the ad with abstract verbal claims and high visual content reported a greater interest in getting more information than those exposed to the ad with abstract verbal claims and low visual content (F=4.06, d.f.=57, p<.001).

Hypothesis four predicted that consumers in the experimentally induced knowledge condition would be more likely to expend energy in information seeking than subjects in the "no knowledge" condition. A main effect for KNOWLEDGE would be expected on each of the information seeking measures. The hypothesis received partial support. An EXPERIENCE main effect was found with respect to STORE (F=7.558, d.f.=1, p < .01). Subjects exposed to the "knowledge" manipulation indicated that they would be more likely to go to a stereo store and ask for a demonstration (X=3.3, S.D.=1.67) than subjects in the "no knowledge"manipulation (X=4.15, S.D. 1.735). There was no EXPERIENCE main effect for measures tapping subjects' interest in getting more information, SEEK (F=.06, d.f.=1, p=.806) or interest in seeing a demonstration, DEMO (F=.036, d.f.=1, p=.85).

TABLE 1

ANALYSIS OF VARIANCE SUMMARY TABLES

The expectation in hypothesis 5 was that subjects exposed to the "knowledge" manipulation would rate the new disc player as less risky than those in the "no knowledge" treatment. A KNOWLEDGE main effect for subjects' RISK perceptions would be expected. While a KNOWLEDGE main effect did appear for RISK (F-8.56, d.f.-1, p <.001), the effect was not in the hypothesized direction. The "knowledge" group rated the product as more risky (X-1. 17, S.D.-] .56) than the "no knowledge" group.

SUMMARY AND IMPLICATIONS

Our study examined the effect of various ad presentations on subjects' attitudes toward a dynamically continuous innovation while subjects we-" the early or later awareness stages of the adoption process. Experimental manipulations involved ads with high or low realistic picture dominance, concrete or abstract verbal claims and a knowledge/no knowledge subject condition. The study found no strong influence of picture dominance on any of the three attitudinal measures (attitude toward the ad, attitude toward the product, and attitude toward information seeking). Picture dominance positively affected subjects' attitude toward the ad only when verbal claims were low in information (abstract claims). These results appear to contradict those of Rossiter and Percy (1978), who found a similar effect operating for attitude toward the brand but in a concrete information condition. For a dynamically continuous innovation, the large picture (here a realistic picture of the compact disc player) appears to provide subjects with added information, which abstract claims alone do not convey.

FIGURE 2

INTERACTION EFFECT OF PICTURE AND INFO FOR ATTITUDE TOWARD THE AD-ATTAD

FIGURE 3

INTERACTION EFFECT OF PICTURE AND INFO FOR INTEREST IN GETTING MORE INFORMATION-SEEK

Results did indicate that concrete claims are superior to abstract claims in influencing subjects' attitude toward the product and their interest in seeing a demonstration. It appears that the type of information presented (whether concrete or abstract) is more critical than the visual imagery (picture dominance) in affecting subjects' attitudes. This result compares favorably with those of Kisielius (1982).

In addition, the more knowledge that subjects had about the innovation, the more they tended to actively seek information (in terms of a demonstration). However, with this added knowledge, they were more likely to perceive the product as risky. These results support the notion that as individuals move through adoption process effects models, it is important to address their risk perceptions as well as to provide them with more information. The results of this study do not indicate a significant difference between the verbal and visual treatment conditions employed in terms of subjects' risk perceptions. Since risk appears to be an important variable inhibiting product adoption, future research should address whether there is a preferable method of ad presentation which would effectively reduce the risk perceived by those with increasing amounts of product knowledge.

Finally, if we count the knowledge manipulation of the present study as concrete information, our results indicate that information, whether provided by the ad itself (picture or concrete information) or by natural experience during diffusion, may be the most powerful factor in the earlier stages of consumer adoption processes for more radical innovations.

REFERENCES

Alesandrini, R.L. (1982) "Imagery-eliciting Strategies and Meaningful Learning," Journal of Mental Imagery, 6, 125-140b.

Booz, Allen and Hamilton, Inc. (1982) New Products Best Practices: Today and Tomorrow, New York: Booz, Allen and Hamilton, Inc.

Brown, Lawrence A. (1981) Innovation Diffusion: A New Perspective, New York: Methuen.

Calantone, Roger and Robert G. Cooper (1981) "New Product Scenarios: Prospects for Success," Journal of Marketing, 45 (Spring), 48-60.

Darden, W.R. and R.D. Reynolds (1974), "Backward Profiling of Male Innovators," Journal of Marketing Research, 11, 75-88.

Dickerson, Mary Dee and James W. Gentry (1983), Characteristics of Adopters and Non-Adopters of Home Computers," The Journal of Consumer Research, 10, 225-235.

Edell, J.A. and R. Staelin (1983), "The Information Processing of Pictures in Print Advertisements," The Journal of Consumer Research, 10, 45-61.

Jacoby, J. (1971) "Personality and Innovation Proneness," Journal of Marketing Research, 8 (May), 409-418.

Kisielius, Jolita (1982), "Detecting and Explaining Vividness Effects in Attitudinal Judgment," unpublished doctoral dissertation, Northwestern Univ., Evanston, IL.

Kisielius, Jolita and B. Sternthal (1984) "Detecting and Explaining Vividness Effects in Attitudinal Judgments," Journal of Marketing Research, 21 (Feb.), 54-64.

Kotler, P. and G. Zaltman (1976) "Targeting Prospects for a New Product," Journal of Adv. Research, (Feb.), 7-20.

Lavidge, R.J. and G.A. Steiner (1961) "A Model for Predictive Measurement of Advertising Effectiveness," Journal of Marketing, 25 (Oct.), 59-62.

Lippman, M.Z. (1974), "Enactive Imagery in Paired-Associate Learning," Memory and Cognition, 2, 385-390.

Lutz, K.A. and R.J. Lutz (1977) "The Effects of Interactive Imagery on Learning: Application to Advertising," Journal of APPlied Psychology, 62, 493-498.

Lutz, R.J., S.B. MacKenzie, and G.E. Belch (1983), "Attitude Toward the Ad as a Mediator of Advertising Effectiveness: Determinants and Consequences," in Advances in Consumer Research, Vol. 10, R.P. Bagozzi and A M. Tybout, eds., Ann Arbor, MI: Assoc. for Consumer Research, 532-539.

Midgley, David F. (1977) Innovation and New Product Marketing, New York: John Wiley and Sons.

Midgley, David F. and G.R. Dowling (1978) "Innovativeness: The Concept and Its Measurement," The Journal of Consumer Research, 4, 229-242.

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

Nisbett, R.E., E. Borgida, R. Crandall, and H. Reed (1978) "Popular Indiction: Information is Not Necessarily Informative," in J.B. Carrol and J.W. Payne, eds., Cognition and Social Behavior, Hillsdale, NJ: Lawrence Erlbaum Associates.

Ostlund, L.E. (1972) "Perceived Innovation Attributes as Predictors of Innovativeness," The Journal of Consumer Research, 1, 23-29.

Palda, Kristian S. (1966) "The Hypothesis of a Hierarchy of Effects: A Partial Evaluation," Journal of Marketing Research, 3 (February), 13-24.

Paivio, A. (1969) "Mental Imagery and Associative Learning and Memory," Psychological Review, 76, 241-263.

Paivio, A. (1971) Imagery and Verbal Processes, New York: Holt. Rinehart. and Winston.

Paivio, A., J. C. Yuille and S.A. Madigan (1968) "Concreteness, Imagery, and Meaningfulness Values for 925 Nouns," Journal of Experimental Psychology, 76, 1-25.

Paivio, A. and K. Csapo (1973) "Picture Superiority in Free Recall: Imagery or Dual Coding?" Cognitive Psychology 5, 176-206.

Robertson, Thomas S. (1971), Innovative Behavior and Communication, New York: Holt, Rinehart, and Winston, Inc.

Rogers, Everett M. (1983), Diffusion of Innovations, 3rd ed., New York: Free Press.

Rogers, Everett M. and Floyd Shoemaker (1971), Communication of Innovations, New York: Free Press.

Rossiter, John R. and Larry Percy (1978) "Visual Imaging Ability as a Mediator of Advertising Response," in Advances in Consumer Behavior, Vol. 5, H. K. Hunt, ed., Ann Arbor, MI: Association for Consumer Research, 621-629.

Rossiter, John R. (1980) "Attitude Change through Visual Imagery in Advertising," Journal of Advertising, 9:2, 10-16.

Ryan, Bryce and Neal C. Gross (1943) "The Diffusion of Hybrid Seed Corn in Two Iowa Communities," Rural Sociology, 8, 15-24.

Shepard, R.N. (1967) "Recognition Memory for Words, Sentences and Pictures," Journal of Verbal Learning and Verbal Behavior, 6, 156-163.

Shimp, T.A. (1981) "Attitude Toward the Ad as a Mediator of Consumer Brand Choice," Journal of Advertising, 10:2. 9-15.

Urban, Glen L. and John R. Hauser (1980), Design and Marketing of New Products, Englewood Cliffs, NJ: Prentice-Hall.

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