Printer Friendly

The use of vivid stimuli to enhance comprehension of the content of product warning messages.

The Use of Vivid Stimuli to Enhance Comprehension of the Content of Product Warning Messages

Marketers are under a legal mandate to communicate accurately the hazards associated with product use to consumers (Ross 1981). Yet, thousands of consumers are injured annually using consumer products (McIuturff 1981). Moreover, some data indicate product warnings themselves may not be communicated effectively (Jacoby and Hoyer 1982; Morris et al. 1986). Clearly, improving a marketer's ability to communicate the hazards associated with product use could have significant consequences.

Some firms are beginning to use concrete language and pictorial symbols in an effort to make the content of warning messages more vivid and, hence, more understandable to consumers. It is intuitively compelling to argue that these vivid product warnings will be more memorable and have greater impact than more pallid, abstract warnings. Vivid messages, by being more concrete or image-provoking, are popularly believed to be more memorable and to have a greater impact on judgments than abstract material. However, research in the vividness literature is replete with failures to demonstrate reliably that vivid material will have any consistent impact on memory or judgments (e.g., Taylor and Thompson 1982).

More recently the vividness literature (Taylor and Thompson 1982; Fiske and Taylor 1984; Kisielius and Sternhal 1984, 1986) suggests that much of the early research may have been methodologically flawed and guided by an inadequate conceptual model of how consumers respond to vivid stimuli. In this revised view, vivid material may serve two necessary communication functions better. Vivid material may be more succesful in attracting consumers' attention and may be more likely to stimulate consumers' thoughts, or cognitive elaborations, on a message. The nature and content of these thoughts, in turn, may serve as the basis for vividness effects on memory or judgment (Kisielius and Sternthal 1984, 1986; Mitchell 1983).

This approach may offer the best conceptual method to study the impact of using vivid stimuli to increase comprehension of product warnings. Vivid product warnings may be more successful in attracting consumers' attention to a warning message and may stimulate a greater amount of cognitive elaboration than would be generated by a more pallid, abstract presentation of the warning message. It may be particularly important to identify the nature of the evaluations generated in response to vivid product warnings. The use of warning messages may improve communication of the hazards associated with product use. Yet use of vivid product warnings might also stimulate the production of negative evaluative thoughts that will serve to discourage purchase.

This paper makes several contributions to the consumer affairs literature. First, it shows that vivid product warnings are a more effective communication tool to use to communicate the hazards associated with product use to consumers. Second, it identifies the types of elaborations that are most likely to be produced by vivid product warnings. Third, it identifies some of the limitations of the vividness effect by showing the impact vivid product warnings have on consumers as they are increasingly more motivated to process product-related information. This is notable as it is not known if the vividness effect is relatively constant or if it will disappear as consumers devote more energy to processing product-related information.


Demonstration of the Vividness Effect

Three legal criteria are used to judge the adequacy of a consumer product warning. In general, courts have held that a consumer product warning must clearly communicate the following information to the potential user (Ross 1981): it must describe the nature of the risk facing consumers, it must identify the magnitude of that risk, and it must describe how to avoid that risk. Vivid product warnings may be more likely to achieve these legally mandated goals than nonvivid product warnings.

Vividness has been conceptually defined as a communication characteristic achieved by qualities of a message that enhance cognitive elaboration (Taylor and Thompson 1982). This effect is usually achieved by the use of pictorial symbols, which may have a superior ability to attract the attention of consumers, and concrete language, which serves to communicate the essence of a communication in a very specific manner. Both qualities, i.e., superior attention-getting power and enhanced specificity in communication of the gist of a message, appear to be necessary for observation of the vividness effect (Taylor and Thompson 1982).

In principle, the use of vivid stimuli may improve consumers' comprehension of product warnings in two ways. First, the use of pictorial symbols might serve an attention-getting function. The attention-getting power of manipulations of size, color, and contrast are well documented elsewhere (Engel and Blackwell 1982; Hawkins, Best, and Coney 1986). Moreover, many of the symbols used in warning messages (i.e., red borders) may also provide cues that help communicate part of the content of the textual material used in a warning message.

Second, the use of concrete language may communicate the content of a warning message better to consumers by being more specific and easier to understand than abstract material. Concrete language has the ability to communicate more accurately specific attribute information of the type that product warning messages are legally required to communicate (Johnson and Fornell 1987). In essence, abstract language may leave some ambiguity in consumers' minds as to the nature or magnitude of the risk they face from product use. Yet, this ambiguity may be reduced by the use of simpler, more specific language.

These two functions may be performed by separate components of a product warning message. Pictorial devices, such as contrasting borders or symbols, may serve to attract the attention of the consumer to the warning message. Concrete language may serve to communicate the specific information product warning messages are legally required to communicate. Some research indicates that it may be necessary to use both symbols and concrete language to generate enhanced levels of consumer understanding of the content of a product warning message (Morris et al. 1986; Kelley 1989). These considerations lead to the hypothesis that vivid product warnings generate more accurate memory for message content than nonvivid messages.

The Nature of the Elaborations Induced by Vivid Product Warnings

Early research in the vividness literature assumed vivid stimuli would always spark cognitive elaborations in favor of the position advocated in a vivid communication. Yet, repeated failures to find any vividness effect on memory or judgments showed this view to be somewhat naive (Taylor and Thompson 1982). Currently, the availability valence view cautions that a wide range of elaborations can be generated by vivid stimuli, some of which might run counter to the position advocated in a vivid communication (Kisielius and Sternthal 1984). The essence of the availability valence view is that vivid stimuli may stimulate either positive or negative thoughts in response to a communication. These thoughts may later be more available in memory when a purchase is made. The impact of these thoughts on purchase will be determined by their favorability, or valence, regarding the vivid communication.

Product warnings are hypothesized to generate two basic types of elaborations. The first, termed lexical cognitions, focuses exclusively on extraction of the content of the warning message. The second, termed semantic cognitions, reflects any additional extensions made of the warning message, any implications drawn by consumers based on the warning message, or any evaluative thoughts generated by consumers about the safety of the product in question (Lachman, Lachman, and Butterfield 1979).

Lexical analysis focuses on repetition of the points made in a product warning without embellishment by consumers. For example, consumers might mentally repeat the point that a product should not be used by young children without supervision (Mitchell 1983). Enhancement of the number, or strength, of lexical elaborations should create more accurate memory for the content of the warning message. These elaborations are most likely to be enhanced by the use of very specific, concrete language in the text of the warning message coupled with the use of attention-getting cues (e.g., a red warning symbol) whose meaning is partially redundant with the content of the warning message (Johnson and Fornell 1987; Kelley 1989). A second hypothesis is that vivid product warnings will induce a greater number of lexical thoughts than nonvivid product warnings.

It is currently an open question whether the use of vivid stimuli per se prompts increased thinking at a semantic level. A traditional view of vividness suggests that vivid product warnings enhance semantic analysis because such messages are more interesting and more arousing to consumers, but such intuitively plausible effects have been notoriously difficult to demonstrate in the vividness literature (Taylor and Thompson 1982). It may be very unlikely that vivid stimuli per se will have an impact on semantic analysis of a message. Vivid product warnings may facilitate lexical analysis and provide an enhanced opportunity for semantic analysis to occur, but the use of vivid material per se may not make semantic analysis inevitable. The primary determinant of whether consumers will engage in semantic analysis may be the presence of other situational or receiver characteristics that prompt semantic elaboration, such as processing goals (Andrews 1988). In the absence of such factors, individuals may be unlikely to engage in semantic analysis of a communication solely because of the use of vivid stimuli.

The Impact of Increases in Processing Effort

on the Processing of Vivid Product Warnings

Two plausible, yet contradictory, positions have been taken in the vividness literature regarding how the processing of vivid product warnings would be affected by increases in the amount of cognitive effort consumers devote to processing product warning messages (Taylor and Thompson 1982). One posits an interaction between motivation to process and vivid stimuli. The other suggests that there may be no relationship between motivational increases and the processing of vivid stimuli.

At low levels of motivation consumers may not process any information at all, while at high levels they may engage in extensive semantic elaboration of both vivid and nonvivid material (Johnson-Laird 1974). This point seems intuitively plausible in a product warning context. For example, consumers with a life threatening condition may be very alert for information regardless of its vivid or nonvivid nature. Vivid information may best serve to highlight information and to induce elaborations that may not have occurred otherwise in a moderately motivated group only. Yet, no supporting data have been offered to support this notion, and a strong case can be made for an alternative position.

An alternative position posits that the processing of vivid warning messages should not be affected by motivational increases (Richardson 1980). Vivid stimuli may serve to make it easier for consumers to understand the gist of a warning message by enhancing lexical analysis. Yet, it must be remembered that product warnings are relatively brief communications. Lexical analysis of vivid product warnings may be very straightforward once consumers exceed the minimum threshold level of motivation needed to induce them to process a warning message. If such is indeed the case, one would expect improvements in lexical analysis to be associated solely with the use of vivid warning messages while, somewhat counterintuitively, motivational increases would have little impact.

The counterintuitive nature of this position will be alleviated if care is taken to avoid confusing the impact of increases in processing effort with the impact of other cues designed to trigger semantic elaboration. Without such confounding cues, increases in motivation may have little impact on semantic analysis. In the absence of semantic prompts, processing may remain focused at a lexical level of analysis even though individuals do allocate more effort to processing a communication. Motivational increases are hypothesized to have no impact on the processing of vivid product warning messages.

To date, no research has examined which of these two conflicting views best represents the impact motivational increases have on the processing of vivid product warnings. Testing these effects requires manipulation of both vivdness and subjects' motivations to process a communication, with attendant manipulation checks, if meaningful comparisons are to be made between how vivid and nonvivid product warnings affect cognitive responses across levels of motivation. Taylor and Thompson (1982) have noted that such methodological refinements are conspicuously absent in vividness research.


Vividness and motivation to process information were manipulated in a 2 x 3 factorial design with two levels of vividness (vivid and nonvivid) and three levels of motivation (low, medium, and high). Subjects were recruited under the guise of participating in a study of how people react to new products. Each subject was exposed to four product labels. One contained a vivid (or nonvivid) product warning message. Three neutral, nonexperimental product labels were included in the study to disguise the purpose of the experiment. These contained a nonvivid product warning message. Four product labels were used in the experiment, as pretests indicated that using more labels caused fatigue with the experimental task.

All of the labels used in the experiment had the same design. Fictitious brand names were used. The product warning was placed in the same position on each label. The nature of the claims made in the nonwarning text of each label was similar, but the wording was changed due to the use of different products. Approximately 75 percent of the label was devoted to this material, with the remainder being allocated to the product warning. The three nonexperimental product labels and either the vivid or nonvivid experimental product label were randomly placed in a booklet to guard against order effects. Analysis later indicated that the order of placement had no impact on memory or cognitive responses. A detailed discussion of the experimental procedure is provided below.

Development of the Manipulation of Vividness

A fictitious product was used in this research. Using familiar brands in vividness research may introduce a confound if subjects respond on the basis of prior knowledge rather than from the experimental stimuli. For much the same reason, it is important to use a neutral experimental product to prevent the gender of the subjects from influencing their responses (Kisielius and Sternthal 1984). Pretesting indicated that subjects believed a fictitious fluoride dental rinse would, in principle, be safe for use in the home. Based on the results of this pretest and the fact that it was not a sex-typed product, a fluoride dental rinse was selected as the experimental product.

Many of the warning messages used on fluoride dental rinses caution that the product should not be used by children under the age of six. Some state that the product should not be swallowed and that milk should be given if the product is swallowed. The product warning message created for the fictitious product discussed different attributes than these to guard against product familiarity, but the message approximated the same threat level to the consumer from misuse of the product.

Extensive pretesting was conducted to develop equivalent vivid and nonvivid warning messages for the experimental product. A seven-item vividness scale was constructed from previous research and was used in these pretests. The vividness scale is shown in Appendix 1. The vivid warning message was compared to the nonvivid warning message by summing the item values to obtain an overall vividness score.

Both vivid and nonvivid warning messages for the fictitious product were developed. Color, size of print, and location of the warning message on a label are used currently by manufacturers to attract attention to product warnings. Some manufacturers include a symbol or a picture in the product warning (Federal Hazardous Substance Act 1960) or a contrasting border (Funkhouser 1984). Given these examples, a warning symbol (i.e., a circle with a slash across it) with a red contrasting border drawn around the warning message was selected to operationalize the attention-getting portion of the vivid warning message. The textual portion of the warning message was written in either concrete language (vivid condition) or in abstract language (nonvivid condition). As previous research indicates (Kelley 1989), both portions are necessary to generate a vividness effect.

The vivid product label is shown in Appendix 2. The nonvivid product label made the same product claims but did not contain a red border or the warning symbol. The nonvivid warning was written in more general, or abstract, language. The abstract warning read as follows: "Warning: Ingesting one ounce will result in gastrointestinal irritation due to a shift in gastric pH. Seek advice from a medical professional if ingested. Do not empty gastric contexts after ingesting." Pretests indicated subjects were unfamiliar with the use of medical terminology and felt the meaning of the message was ambiguous. Subjects indicated words commonly found in everyday communication were more specific and easier to understand. This use of the abstract-specific dimension reflects current thinking that equates concreteness with specificity (Johnson-Laird 1974; Johnson and Fornell 1987; Richardson 1980).

Before the labels were used in the experiment, three judges evaluated whether the vivid and nonvivid warnings were equivalent in meaning. The judges were asked to respond to a short questionnaire designed to evaluate the warning messages relative to the legal criteria used to judge the adequacy of a warning message. Warning messages are deemed equivalent if they describe the nature of the risk facing consumers, if they identify the magnitude of that risk, and if they describe how consumers can avoid that risk (Ross 1981). The judges unanimously indicated that the vivid and nonvivid messages were equivalent in terms of meaning.

Manipulation of Motivation to Process

Motivation to process message information is conceptualized in the literature as the amount of cognitive effort consumers invest in processing a message (Petty and Cacioppo 1986). Examining the motivation-to-process issue required the use of a subject pool with three distinct levels of motivation to process the experimental stimuli. In the early stages of this research, it was obvious that efforts to use many preexisting groups (e.g., individuals on sodium restricted diets) would be frustrated by the confounds this procedure would introduce into the data. Such individuals had greater knowledge pertinent to their specific contradiction than the general public. Those with severe contradictions had been thoroughly educated by health care professionals about the hazards associated with ignoring their contradictions. Both of these (greater product knowledge and the hazards associated with the contradiction) would have served as confounding prompts to engage in semantic elaboration.

However, many product warnings do not deal with specific medical contradictions. Many warnings advise of harm associated with products used by the public at large, such as the fluoride dental rinse chosen for this study. With such products, it would be possible to manipulate experimentally processing effort while avoiding the simultaneous introduction of confounds prompting semantic analysis (Andrews 1988).

Several pretests indicated that processing effort could be manipulated by varying (1) the instructions subjects received, (2) the compensation they would receive, and (3) the percentage of correct answers they would need to earn their compensation for participating in the study. (1)

In the high motivation situation, subjects were asked to imagine that they were actively seeking information about the experimental products when they were reading the product labels. In addition, subjects were informed that they could boost their compensation for participation in the experiment by 15 percent if they were able to answer correctly 80 percent of the questions that they would be subsequently asked about the product labels.

In the moderate situation, subjects were told to imagine that they were only somewhat interested in gathering information about these products when they read the labels. They were told that they could boost their compensation for participation in this study by 4 percent if they were able to answer 30 percent of the product label questions correctly.

In the low setting, subjects were told to imagine they were not interested in gathering information about these products when they read the labels. They were told that they only needed to complete the study in order to earn their compensation. They were told that it did not matter whether they answered subsequent questions correctly or not since their answers would not be scored.

A five-item scale was developed to serve as a measure of subjects' motivation to process message information. Each item consisted of a 0-100 point scale segmented into increments every five points. The items were anchored at intervals of every 20 points (Appendix 3).

Dependent Variables

Memory accuracy was measured by questions with "yes," "no," or "do not remember" responses. Multiple choice items were considered as a mode of questioning. However, pretests indicated the limited content of the warning messages precluded construction of more than two or three nonredundant multiple choice questions (Jacoby and Hoyer 1982; Jacoby, Nelson, and Hoyer 1982). The "do not remember" responses were included to preclude forcing subjects to judge the accuracy or the inaccuracy of any particular question (Ford and Yalch 1982).

A total of ten questions was asked for each product label. Six of these ten questions were related to the warning message. These questions asked if the warning label stated: (1) if the product should be swallowed, (2) if you should induce vomiting if swallowed, (3) if you should call a physician if swallowed, (4) if small children should be watched while using the product, (5) if swallowing will change the amount of acid in the stomach, and (6) if swallowing would irritate the stomach.

The other four questions were related to information contained in other parts of the product label. These were included to help disguise the true purpose of the experiment. Questions were also asked for each of the three nonexperimental labels to enhance the experimental camouflage.

Cognitive elaboration was measured by subjects' cognitive responses. These were obtained by having subjects list all thoughts that occured to them while they read the product label. Subjects had three minutes in which to list their thoughts, consistent with Wright's (1980) thought-listing procedure. The coding scheme used to categorize these data is discussed in the analysis section.

Subjects and Experimental Procedure

A total of 90 students enrolled in undergraduate marketing classes at a western university participated in the experiment. Fifteen subjects were randomly assigned to each treatment. Small groups were used. Subjects were recruited under the guise of participating in two tests of how consumers would react to new products.

In the first session, subjects entered a small room and booklets containing all experimental material were distributed. A confederate started the session by reading the following cover story to subjects: "Thank you for participating in this study. Assume that you have been asked to participate in a test of your reactions to a number of new products that are currently being developed by some companies. The pages that follow contain the labels for these products." Then the instructions to manipulate level of need were read. Subjects then read the first product label and performed the first thought-listing task. The subjects were then instructed to read the second label and perform a second thought-listing task. This procedure continued until the end of the fourth thought-listing task.

The booklets containing the labels and thought listings were then collected. A second booklet containing a short distraction task was then distributed. The purpose of this distraction task was to erase any recency effect from reading the last label and consisted of some short mathematical problems (Childers and Houston 1984). Following the distraction task, the measures of memory accuracy and processing motivation were administered. This ended the first session, which lasted approximately 40 minutes.

All subjects returned for a second session two days later. Subjects completed Mark's (1973) Visual Vividness Imagery Questionnaire (VVIQ) to serve as a control for differences in subjects' abilities t process visual material (Childers and Houston 1984). Next, the subjects were exposed to the particular experimental product label they had seen in the first session and were asked to complete the vividness scale developed in the pretests. The VVIQ test and the manipulation check for vividness were done in a second experimental session to prevent subjects from revealing the purpose of the study to others who had not yet completed their first session. This session, which lasted approximately 20 minutes, ended with subject debriefing.

Later analysis revealed that ANOVA results are not statistically different with or without adjusting memory scores for individual differences in visual imagery abilities as measured by VVIQ. Hence, ANOVA results are reported here without using VVIQ scores as a covariate. Apparently, the visual symbols used in the manipulations of vividness are of such a simple nature that all subjects are equally able to process them. Individual differences in ability to process visual material might have been a factor with more complicated visual stimuli.


Manipulation Checks

The unidimensionality of the global measure of vividness used as a manipulation check was examined by means of the confirmatory factor analysis procedure suggested by Gerbing and Anderson (1988) using the recommended maximum likelihood procedure of LISREL VI (Joreskog and Sorbom 1983). Each of the seven items in the scale loads significantly on one factor (p [is less than] .01). While this does not exclude the possibility that vividness is a multidimensional construct, it provides strong evidence that one dimension underlies responses to the global measure of vividness used as a manipulation check. Hence, subjects' scores can be summed on the items in this scale to reach a total vividness score.

Coefficient alpha was used to assess the reliability of the vividness scale used as a manipulation check. Coefficient alpha (.86) indicates a high degree of reliability across the seven items. An independent t-test for the summed vividness scores confirms that the vivid [Mathematical Expression Omitted] and nonvivid warnings [Mathematical Expression Omitted] are significantly different (t = 8.87, p [is less than] 0.001). Therefore, the vividness manipulation is successful.

Coefficient alpha of the five-item motivation-to-process scale is .95, which indicates that the scale possesses a high degree of reliability. An overall ANOVA model containing the high [Mathematical Expression Omitted], medium [Mathematical Expression Omitted], and low [Mathematical Expression Omitted] levels of motivation is significant (F = 79.676, p [is less than] 0.001). A Tukey multiple comparison test indicates that the three levels are significantly different at the 0.05 level. Therefore, the manipulation of motivation is successful.

Cognitive Responses

Thought verbalizations were coded into the four categories depicted below. The codes used reflect different degrees of elaboration of message content ranging from lexical analysis of the symbol/words used in the label or the product warning to semantic elaboration on individual message elements. Transcribed thoughts were listed by subject number and coded independently without reference to experimental condition by two judges who agreed on the classification of 75 percent of all thoughts. Differences were resolved by discussion.

(1) Product warning-related lexical thoughts that reflect merely a restatement of some portion of the content of the warning message.

(2) Product warning-related semantic thoughts that include thoughts reflecting an inference, a conclusion, or an extension of the content of the warning message.

(3) Label-related lexical thoughts reflecting merely a restatement of some portion of the content of the product label.

(4) Label-related semantic thoughts reflecting an inference, a conclusion, or an extension of the content of the product label.


Subjects' responses to the six memory questions were categorized as "correct," "incorrect," and "do not remember." Before the hypotheses were tested, the number of correct responses was adjusted for guessing. To adjust the raw data for guessing, each subject's number of correct responses was converted to a proportion as shown below (Sax 1980).

[(number of correct responses) - (number of incorrect responses)]/6

This adjustment omitted the "do not remember" responses to reflect more accurately subjects' knowledge of the material manufacturers are legally mandated to communicate in a warning message. Including the "do not remember" responses would overestimate knowledge of this material (Funkhouser 1984; Jacoby and Hoyer 1982). As a practical matter, the subsequent analysis was repeated using other methods of correcting scores for guessing as well as with the raw scores, and results are the same. Scores for each treatment condition are shown in Figure 1.

Effects on Memory

The impacts of the manipulations of vividness and motivation on memory scores were tested with univariate analysis of variance. The vividness main effect is significant (F = 3.164, p [is less than] 0.001). Hypothesis one is supported. Use of vivid messages generates more accurate memory for the content of the warning message. The interaction between vividness and motivation is not significant (F [is less than] 1). In addition, the motivation main effect is not significant (F = 1.03, p [is less than] .36). A Tukey multiple comparison test (p = 0.05) indicates that each vivid treatment is significantly different from the corresponding nonvivid treatment at a given level of motivation. These data are displayed in Figure 1.

Use of vivid product warnings consistently results in more accurate memory for the content of the product warning message. This effect is observed regardless of motivational level, clearly indicating the superiority of vivid product warnings in accurately communicating the hazards associated with product use. These data support the hypothesis that vivid warnings have a positive impact on the processing of warning messages regardless of motivation level. This helps reject the contention that vivid stimuli have their greatest impact when consumers are moderately motivated to process a communication. This issue can be addressed more fully by consideration of the cognitive response data.

Effects on Cognitive Responses

Manipulations of vividness and motivation may also increase thoughts about the other elements of the product label in addition to increasing production of thoughts about the product warning message. The average number of total product warning and total label-related thoughts for each subject in each experimental condition is shown in Figure 2. These data indicate that the vividness manipulations increase thoughts only about the product warning message. Production of thoughts about other parts of the product label is not affected. Analyses of these other thoughts confirms that there is no difference across all treatments. Since increased thought production is solely product warning related, these other cognitions will not be discussed further.

Production of lexical and semantic product warning thoughts is shown in Figure 3. Hypothesis two is supported. Vivid stimuli enhance production of lexical cognitions (F = 6.68, p [is less than] 0.001), while production of semantic cognitions is not affected (F [is less than] 1). Vivid product warnings generate a greater number of lexical thoughts than nonvivid warnings. The vividness manipulation has no significant impact on production of semantic cognitions, although inspection of Figure 3 reveals a small change in that direction. Hypothesis three is also supported. The interaction between motivation and vividness has no impact on production of either lexical (F [is less than] 1) or semantic (F [is less than] 1) cognitions. This counts as strong evidence against the contention that vivid product warnings have their greatest impact when consumers are moderately motivated to process information. Motivational increases have no impact on the production of lexical (F [is less than] 1) or semantic (F [is less than] 1) cognitions. This supports the hypothesis that motivational increases have no impact on the processing of vivid product warnings.

In summary, vivid product warnings generate more accurate memory for message content and a greater number of lexical thoughts than nonvivid warnings. Vivid warnings do not increase semantic cognitions. Level of motivation has no impact on cognitions or memory, supporting the contention that the brevity of product warnings may render them relatively immune to increases in motivation to process message information once some minimal level of motivation to process a warning message is achieved. Variations in motivation to process information may effect lexical analysis of longer, more complex textual material.


This research presents a somewhat narrower view of the impact vivid product warnings might have on consumers than has traditionally been taken in the vividness literature. The data in this experiment indicate that vivid product warnings may serve a lexical enhancement function. The multiple lexical cues built into product warning messages may serve to enchance consumers' understanding of the content of the product warning message. In the present case vivid product warnings are clearly superior in communicating the hazards associated with product use to consumers. This advantage is observed regardless of how much cognitive effort consumers put into processing label information. The cognitive response data indicate that the vividness manipulation serves to increase the average number of lexical thoughts subjects have regarding the warning message. It is noteworthy that this communication advantage does not, in and of itself, generate negative semantic evaluations about the product in question. This suggests the tentative conclusion, worth replicating elsewhere, that the increased lexical analysis induced by vivid product warnings may not necessarily stimulate production of semantic elaborations that serve to discourage purchase. Future research should manipulate the presence or absence of semantic prompts and the use of vivid or nonvivid product warning messages. Such a study would make it possible to observe the separate effects of these factors on consumers' cognitive elaborations.

The results of this research provide those who are interested in protecting the consumer with some additional insight on how to facilitate the communication of consumer information. Federal agencies, such as the Consumer Product Safety Commission, whose purview includes the promulgation of product warning guidelines, should begin to examine the appropriateness of requiring the use of vivid stimuli in consumer product warning messages. The present data suggest that vivid product warnings might best enhance the consumer's understanding of the nature of the risks associated with product use, the magnitude of this risk, and how to avoid this risk. Replication of these effects may ultimately help the courts develop a judicial test based on the use of vivid stimuli for use in failure-to-warn cases.

This research also suggests that some important nuances should be added to the conceptualization of the vividness construct. First, vividness should be regarded as a stimulus characteristic that can be manipulated and measured across subjects independent of other message characteristics, such as the amount of information or the source of the message. Second, vivid stimuli per se may serve only to enchance thought production at a lexical level of analysis. Vivid stimuli by themselves may not necessarily serve to increase semantic elaboration. The presence of an additional factor, such as a semantic prompt, may be necessary to stimulate semantic analysis.

Of course, the present study is not without limitations. This study examined the reactions of a small sample of undergraduates to a fictitious product warning. Data were obtained in a laboratory setting using an induced motivation manipulation. This is appropriate given the theoretical purpose of this experiment, but it does necessitate the replication of these findings in an applied setting. Such applied work would benefit by incorporating Houston and Rothschild's (1980) detailed recommendations for the conduct of public policy research in applied settings. In particular, incorporation of their recommendations regarding conditions of exposure, scope of treatments, and interactive effects with subgroups of consumers who differ in terms of product knowledge will require that significant efforts be devoted to stimuli development and use of an appropriate subject pool.

Other research that examines the impact of vivid material on comprehension, memory, or judgment could benefit by using the methodological refinements built into this study. It is beneficial, albeit very difficult, to use manipulation checks in such work. Few studies in the vividness literature have used manipulation checks but such use is essential to have some confidence in the internal validity of these experiments.


Vividness may serve to assist consumers in accurately encoding the meaning of product warning messages. As such, the use of vivid product warnings can be an important tool to help accomplish the legal obligation of clearly communicating the hazards associated with product use to consumers. However, development of messages that are truly vivid may be difficult. In this research, several pretests were necessary to develop messages in which subjects perceived differences in terms of vividness. One should not assume that material will be vivid in the absence of such evidence. Marketers wishing to affect memory or judgment may wish to include additional processing cues designed to enchance semantic analysis. Yet, in such instances semantic elaborations may be due to the impact of these additional cues rather than to the use of vivid product warnings.


Andrews, J. Craig (1988), "Motivation, Ability, and Opportunity to Process Information:

Conceptual and Experimental Manipulation Issues" in Advances in Consumer Research,

Vol. 15, Michael J. Houston (ed.), Provo, UT: Association for Consumer Research:

212-225. Childers, Terry and Michael Houston (1984), "Conditions for a Picture Superiority Effect on

Consumer Memory," Journal of Consumer Research, 11 (September): 642-654. Engel, James F. and Roger D. Blackwell (1982), Consumer Behavior, 4th edition, New York,

NY: The Dryden Press. Federal Hazardous Substance Act (1960), Public Law 86-613, July 12, 1960, 86th Congress. Fiske, Susan T. and Shelley E. Taylor (1984), Social Cognition, Reading, MA: Addison-Wesley Publishing Company. Ford, Gary T. and Richard Yalch (1982), "Viewer Miscomprehension of Televised Communication:

A Comment," Journal of Marketing, 46 (Fall): 31-37. Funkhouser, G. Ray (1984), "An Empirical Study of Consumers' Sensitivity to the Wording

of Affirmative Disclosure Messages," Journal of Public Policy and Marketing, 3: 26-37. Gerbing, David W. and James C. Anderson (1988), "An Updated Paradigm for Scale Development

Incorporating Unidimensionality and Its Assessment," Journal of Marketing

Research, 25 (May): 186-192. Hawkins, Del I., Roger J. Best, and Kenneth A. Coney (1986), Consumer Behavior: Implications

for Marketing Strategy, 3rd edition, Plano, TX: Business Publications, Inc. Houston, Michael J. and Michael L. Rothschild (1980), "Policy-Related Experiments on

Information Provision: A Normative Model and Explication," Journal of Marketing

Research, 17 (November): 432-459. Jacoby, Jacob and Wayne Hoyer (1982), "Viewer Miscomprehension of Televised Communication:

Selected Findings," Journal of Marketing, 46 (Fall): 12-26. Jacoby, Jacob, Margaret C. Nelson, and Wayne D. Hoyer (1982), "Corrective Advertising

and Affirmative Disclosure Statements: Their Potential for Confusing and Misleading the

Consumer," Journal of Marketing, 46 (Winter): 61-72. Johnson, Michael D. and Claes Fornell (1987), "The Nature and Methodological Implications

of the Cognitive Representation of Products," Journal of Consumer Research, 14

(September): 214-228. Johnson-Laird, Phillip N. (1974). "Experimental Psycholinguistics," Annual Review of

Psychology, 25: 135-160. Joreskog, Karl G. and Dag Sorbom (1984), LISREL VI: Analysis of Linear Structural Relationships

by the Method of Maximum Likelihood, Chicago, IL: National Educational

Resources, Inc. Kelley, Craig A. (1989), "A Study of Selected Issues in Vividness Research: The Role of

Attention and Elaboration Enhancing Cues" in Advances in Consumer Research, Vol. 16,

T. Srull (ed.), forthcoming. Kisielius, Jolita and Brian Sternthal (1984), "Detecting and Explaining Vividness Effects in

Attitudinal Judgments," Journal of Marketing Research, 21 (February): 54-64. Kisielius, Jolita and Brian Sternthal (1986), "Examining the Vividness Controversy: An Availability-Valence

Interpretation," Journal of Consumer Research, 12 (March): 418-431. Lachman, Roy, Janet Lachman, and Earl Butterfield (1979), Cognitive Psychology and Information

Processing: An Introduction, Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Marks, David (1973), "Visual Imagery Differences in the Recall of Pictures," British Journal

of Psychology, 64: 11-24. McIuturff, Peter S. (1981), "Product Liability: The Impact of California Manufacturers,"

American Business Law Journal, 19: 343-360. Mitchell, Andrew A. (1983), "Cognitive Processes Initiated by Exposure to Advertising," in

Information Processing Research in Advertising, Richard Jackson Harris (ed.), Hillsdale,

NJ: Lawrence Erlbaum Associates, Inc. Morris, Louis A., David Brinberg, Ron Klimberg, Carole Rivera, and Lloyd Millstein (1986),

"Miscomprehension Rates for Prescription Drug Advertisements," Current Issues and

Research in Advertising, 9 (1): 93-118. Petty, Richard E. and John T. Cacioppo (1986), "The Elaboration Likelihood Model of

Persuasion," in Advances in Experimental Social Psychology, Vol. 19, Leonard Berkowitz

(ed.), New York, NY: Academic Press, Inc. Richardson, John T. (1980), Mental Imagery and Human Memory, New York, NY: St.

Martin's Press, Inc. Ross, Kenneth (1981), "Legal and Practical Considerations for the Creation of Warning

Labels and Instruction Books," Journal of Products Liability, 4: 29-45. Sax, George (1980), Principles of Educational and Psychological Measurement and Evaluation,

Belmont, CA: Wadsworth Publishing. Taylor, Shelley E. and Suzanne C. Thompson (1982), "Stalking the Elusive Vividness Effect,"

Psychological Bulletin, 98 (2): 155-181. Wright, Peter (1980), "Message Evoked Thoughts: Persuasion Research Using Thought Verbalizations,"

Journal of Consumer Research, 5 (September): 256-269. The Journal of Consumer Affairs, Vol. 23, No. 2, 1989 0022-0078/0002-243 $1.50/0 (R) 1989 by The American Council on Consumer Interests

PHOTO : FIGURE 1 Adjusted Memory Scores for Each Treatment Condition

PHOTO : FIGURE 2 Average Number of Total Product Warning and Total Label Related Cognitions for Each Treatment Condition

PHOTO : FIGURE 3 Production of Lexical and Semantic Warning Related Thoughts

Craig A. Kelley is a Professor, Department of Marketing, California State University, Sacramento, California. William C. Gaidis is an Assistant Professor, Department of Marketing, Marquette University, Milwaukee, Wisconsin. Peter H. Reingen is a Professor, Department of Marketing, Arizona State University, Tempe, Arizona.
COPYRIGHT 1989 American Council on Consumer Interests
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1989 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Kelley, Craig A.; Gaidis, William C.; Reingen, Peter H.
Publication:Journal of Consumer Affairs
Date:Dec 22, 1989
Previous Article:The distribution of U.S. income and food expenditures.
Next Article:Consumer representation and local telephone rates.

Related Articles
Developing a standard approach for testing new traffic control signs.
Consumer comprehension of environmental advertising and labeling claims.
NIKE Selects vivid studios as Web Architect of Record
Framing your masterpiece.
vivid studios Develops Information Design and Architecture for PBS Online.
Uncommon Understanding: Development and Disorders of Language Comprehension in Children.
Comprehension of Warnings and Resulting Attitudes.
Uncommon Understanding: Development and Disorders of Language Comprehension in Children.
Teaching college level content and reading comprehension skills simultaneously via an artificially intelligent adaptive computerized instructional...

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters