How do you think the use of humour in fear advertising may differ in interpretation and/or effectiveness for those who are younger vs older?

I’m working on a psychology discussion question and need an explanation and answer to help me learn.

Answer one of the following questions:

How do you think the use of humour in fear advertising may differ in interpretation and/or effectiveness for those who are younger vs older?

Do you think this type of marketing would be “ethical” or “approved” in the politically-charged/ social justice focus of todays times?

Journal of Consumer Behaviour, J. Consumer Behav. 11: 147–161 (2012) Published online in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/cb.389 Mixing emotions: The use of humor in fear advertising ASHESH MUKHERJEE* and LAURETTE DUBÉ Desautels Faculty of Management, McGill University, 1001 Sherbrooke Street West, Montreal, QC H3A 1G5, Canada ABSTRACT Fear is used to advertise many products, services, and causes such as antismoking, sunscreen usage, and safe driving. Past research indicates that high levels of fear tension arousal can prompt defensive responses in the audience, which, in turn, can reduce the persuasive effect of the ad. We show in two studies that humor can reduce these defensive responses and hence increase the persuasiveness of fear advertising. Specifically, we show that increasing the level of fear tension arousal decreases persuasion when humor is absent but increases persuasion when humor is present. Further, this interaction of humor and fear tension arousal is mediated by defensive responses related to message elaboration and vulnerability to threat. Our results suggest that the effectiveness of fear advertising can be increased by adding an element of humor to the ad. Copyright © 2012 John Wiley & Sons, Ltd. INTRODUCTION Fear appeals are used to advertise many products, services, and causes such as antismoking, sunscreen use, home alarm systems, life insurance, and condom use. For example, antismoking ads show pictures of diseased lungs, sunscreen ads show the ravages of skin cancer, highway safety films show the gruesome consequences of driving drunk, and insurance ads depict the dangers of not being covered when disaster strikes. Past research suggests that high levels of fear tension arousal are disturbing to the audience and can induce defensive responses such as reduced message elaboration and perceived vulnerability to the threat. These defensive responses, in turn, can reduce the persuasive impact of the ad (Rippetoe and Rogers, 1987; Tanner et al., 1991; Keller and Block, 1996). For example, individuals seeing a sunscreen lotion ad with pictures of a scarred skin cancer patient could cope with their feelings of stress by turning the page and minimizing thoughts about the ad. Further, these individuals can discount their own chances of contracting skin cancer and thus distance themselves from the threat in the ad. To the extent people think less about the message or discount their vulnerability to threat, the lower will be the effectiveness of the ad. A key question in this context is “What can be done to minimize defensive responses and hence increase the persuasive effect of fear advertising?” In the present research, we show that mixing humor into fear can reduce defensive responses and hence increase the persuasiveness of fear advertising. The results of two studies indicate that increasing the level of fear tension arousal from moderate to high increases persuasion when humor is present but decreases persuasion when humor is absent from the ad. Further, this effect of humor is mediated by reduction in defensive responses to the ad such as message elaboration and perceived vulnerability to threat. Notably, these results *Correspondence to: Ashesh Mukherjee, Associate Professor of Marketing, Desautels Faculty of Management, McGill University, 1001 Sherbrooke Street West, Montreal, QC H3A 1G5, Canada. E-mail: [email protected] Copyright © 2012 John Wiley & Sons, Ltd. give insight into the mechanism underlying real-world ad campaigns such as the Lynx antifur ads (see Appendix A) that have combined humor with fear in advertising. To summarize, the contribution of the present research is twofold. First, unlike past research that has studied fear advertising in isolation, we examine the effectiveness of fear advertising in the presence or absence of humor. Second, we clarify the process underlying humor in fear advertising, namely, that humor reduces defensive responses and hence increases persuasion. CONCEPTUAL FRAMEWORK Fear appeals Fear is a negative emotional response to threats in the environment (Tanner et al., 1991; LaTour and Rotfeld, 1997). Theorists investigating fear appeals have made an important distinction between threat and fear (LaTour and Rotfeld, 1997; Voss, 2009). While threat is the presentation of negative consequences that the audience wants to avoid, fear is the emotional response that can change attitude or behavior intent. The relationship between threat and fear is said to be idiosyncratic because the same threat stimulus can result in different levels of fear depending on factors such as the audience, topic, and context (Burnett and Oliver, 1979; Rotfeld, 1988). For example, the threat of unpopularity due to poor dental hygiene is likely to be more fearful for teenagers than for older adults. Based on such arguments, theorists have recommended focusing on fear arousal— rather than threat—as the antecedent variable in research on fear appeals. Another conclusion in past research is that fear arousal can be decomposed into two elements, namely, tension arousal and energy arousal (Henthorn et al., 1993; LaTour and Rotfeld, 1997). While tension arousal is marked by agitation and stress, energy arousal is characterized by vigor and pep. Past research indicates that these two elements of fear arousal are conceptually distinct and can separately 148 A. Mukherjee and L. Dubé influence downstream variables such as attitude toward the ad (Aad) and attitude toward the brand (Abr) (LaTour and Pitts, 1989; Henthorn et al., 1993; LaTour and Rotfeld, 1997). In the present research, we focus on the tension arousal dimension of fear. Past research suggests that fear tension arousal may be productive up to a point by highlighting a problem that needs to be addressed (e.g., Schwarz, 1990). However, as fear tension arousal increases to high levels, the arousal can become aversive, prompting defensive responses such as lower message elaboration and discounted vulnerability to threat (Keller and Block, 1996; Steenkamp et al., 1996). Such defensive responses, in turn, can reduce the persuasive effect of the message. Reduction of message effectiveness from moderate to high levels of fear tension arousal is consistent with several theories of fear such as the fear drive model (Janis, 1967), the arousal theory (Henthorn et al., 1993; LaTour and Tanner, 2003), the protection motivation theory (Rogers, 1985; Tanner et al., 1991), and the parallel response model (Leventhal, 1970; Witte and Allen, 2000). The fear drive model posits that high levels of fear create negative tension, and this negative tension can inhibit individuals’ response to the message. The arousal theory posits that high level of fear arousal generates tension, which can lead to a negative attitude toward the stimulus. The protection motivation theory posits that individuals show maladaptive defensive responses to high levels of fear. The parallel response model indicates that persuasion drops at high levels of fear and that the drop in persuasion is magnified when self-efficacy and response efficacy are low. A link between high fear tension arousal and defensive responses is also suggested by other related research. The approach-avoidance framework posits that avoidance is often the dominant response to high fear (Rippetoe and Rogers, 1987; Tanner et al., 1991). The literature on coping with life crises indicates that the immediate reaction to traumatic real-world events is often one of denial (Horowitz, 1983; Lazarus, 1983). And, according to Folkman and Lazarus (1984), high levels of fear engage a primitive, short-term avoidance schema that inhibits further processing of threat information (see also Passyn and Sujan, 2006). Consistent with the above theoretical frameworks, empirical investigations also indicate that high fear tension arousal reduces persuasion by prompting defensive responses in the audience. For example, Keller and Block (1996) reported lesser intention to use a brand of nicotine patch (“Wonder Patch”) at higher compared to lower levels of fear in an antismoking ad. This effect was more pronounced when the ad included a self-referential statement (“Cigarette smoking is dangerous to your health”) and the audience was prompted to visualize negative health consequences of smoking. Similarly, Rossiter and Thornton (2004) found that high-fear antispeeding ads without any relief at the end of the ad (i.e., “shock” ads) actually led respondents to drive faster in a simulated driving test. In an advertising context, Strong and Dubas (1993) found that increasing the level of fear tension from moderate to high reduced intent to use sunscreen lotion. From a broader perspective, meta-analysis of the fear appeal literature Copyright © 2012 John Wiley & Sons, Ltd. indicates that effectiveness of fear advertising tends to fall off at high levels of fear (cf. Witte and Allen, 2000). In summary, several streams of research suggest that high fear tension arousal prompts defensive responses in the audience, which in turn reduces the persuasive effect of the ad. In the present research, we propose that humor can minimize these defensive responses and hence increase the effectiveness of fear advertising. Humor in fear advertising Humor is a fundamental ingredient of social communication (Chapman, 1996) and is widely used in advertising around the world (Alden et al., 1993; Kellaris and Cline, 2007). Past research in advertising has examined the antecedents of humor, that is, what makes an ad funny (Alden et al., 2000; Woltman-Elpers et al., 2004), as well as the consequences of humor, that is, when does humor in an ad benefit the brand (Chattopadhyay and Basu, 1990; Zhang and Zinkhan, 1991; Weinberger and Gulas, 1992; Cline and Kellaris, 2007). In the present research, we examine the use of humor in fear advertising. How might humor work in fear advertising? According to theories of humor, a key characteristic of humor is that it first presents an incongruity, that is, a juxtaposition of events that do not normally occur together. This incongruity is resolved in a playful or nonserious manner by the punchline, thus generating a humorous response (Raskin, 1985; Speck, 1991; Wyer and Collins, 1992; Alden et al., 2000). Playfulness is said to be a crucial aspect of humor because it provides a safe context in which individuals can resolve the mental challenge posed by incongruity (Wyer and Collins, 1992; Alden et al., 2000). Extending this idea, we argue that the playfulness of humor can also be useful in fear advertising. In brief, we argue that the playfulness of humor provides a safe context that reduces the need for defensive responses to fear advertising. And the lesser the defensive responses, the greater would be the effectiveness of fear advertising. First, consider a fear ad where humor is absent. In this case, as discussed earlier, past research indicates that increasing the level of fear tension arousal from moderate to high is likely to reduce persuasion (Rogers, 1985; Tanner et al., 1991; Witte and Allen, 2000). Next, consider a fear ad where humor is present. In this case, as fear tension arousal increases, the playful nature of humor could provide a margin of safety to the audience—within which individuals can elaborate on the threatening message and acknowledge their personal vulnerability to the threat. Such audience responses, in turn, would increase the persuasive effect of the ad. Notably, our prediction extends the findings of Conway and Dubé (2002), who showed that humor in a moderate-fear ad for condom usage improved persuasion for individuals high (but not low) on the personality trait of masculinity. We argue that, in addition to personality factors such as masculinity, stimulus factors such as the level of fear tension (moderate vs. high) can also influence the effectiveness of humor in fear advertising. Our prediction also extends research indicating that humor can be an effective addition to advertising that shows negative consequences without arousal, especially when audience comprehension J. Consumer Behav. 11: 147–161 (2012) DOI: 10.1002/cb Fear and humor of the ad is low (Voss, 2009). In contrast to negative consequences without an arousal response, we focus on ads that create fear tension arousal, and we show that humor can be effective in such ads by reducing defensive responses to the ad. Our arguments are also consistent with research indicating that humor helps individuals cope with stressful situations in daily life. For example, it has been reported that individuals with a strong sense of humor show lesser distraction and denial in the face of threatening events such as job losses (Martin and Dobbin, 1988; Kuiper et al., 1993). This advantage of a sense of humor, in turn, is said to arise from the evolutionary role of humor as a coping tool for social and cognitive stressors encountered in everyday life (Dixon, 1980). We argue that, in addition to sense of humor as an internal trait, external humor embedded in fear advertising can also reduce defensive responses and hence improve persuasiveness of the ad. Our arguments are summarized in the following hypotheses: H1: Humor and fear in advertising interact to influence persuasion. Specifically, (i) when humor is absent, high fear tension arousal leads to lower persuasion than moderate fear tension arousal, and (ii) when humor is present, high fear tension arousal leads to greater persuasion than moderate fear tension arousal. H2: The interactive effect of humor and fear in H1 is mediated by reduction in defensive responses to the ad, such as message elaboration and perceived vulnerability to threat. We conducted two studies to test H1 and H2. Study 1 used an ad for sunscreen lotion that manipulated fear tension arousal and humor, and assessed brand attitude as a measure of persuasion. Study 2 replicated the results with a contextual manipulation of fear tension arousal and a behavior intent measure of persuasion. We conducted mediation analyses in both studies to validate the psychological mechanism underlying humor in fear advertising. STUDY 1 Design and participants This study was designed as a 2 (fear tension arousal: moderate vs. high)
2 (humor: absent vs. present) betweensubjects factorial. A convenience sample of 124 undergraduate students (31 per group) at a large Canadian university was used in this study. Students were used because we used the threat of skin cancer in a sunscreen lotion ad to manipulate fear tension arousal. Skin cancer is likely to be perceived as a relevant health issue by student respondents. The incidence of skin cancer has increased significantly among young adults over the last three decades (Christenson, 2005). In absolute numbers, the annual incidence of invasive melanoma—the most malignant form of skin cancer—among young adults has increased from 5.5 cases per 100,000 persons in 1973 to 13.9 per 100,000 in 2004. This trend has been further accentuated by the use of tanning beds by young adults, which Copyright © 2012 John Wiley & Sons, Ltd. 149 increases exposure to harmful ultraviolet (UV) rays and magnifies chances of getting skin cancer (Time, 2009). As a result of these trends, young adults of college age are often targeted by public service ads promoting sun safety (see Appendix B). Hence, we judged that undergraduate students were appropriate respondents for our study. Our sample included 66 females and 58 males, with an age range of 19–24 years. Procedure Participants were told that they were taking part in a marketing survey for a new sunscreen lotion. As part of the survey, they were shown an ad for sunscreen lotion that manipulated the independent variables of fear tension arousal and humor. Participants then responded to questions on persuasion and reported their thoughts and feelings while reading the ad. Finally, participants completed the manipulation checks and were debriefed and dismissed. Independent variables Fear tension arousal and humor were manipulated in an ad for a fictional brand of sunscreen lotion (see Appendices C–F). All versions of the ad contained a heading mentioning skin cancer (i.e., “Skin Cancer: It Need Not Happen to You!”), a picture of the brand (i.e., “Sail Sunscreen Lotion”), a slogan (i.e., “The Sunscreen for Today’s Generation”), and five bulleted items of brand information. Fear tension arousal was manipulated by a picture in the ad that varied the consequences of skin cancer (McMath and Prentice-Dunn, 2005; Keller, 2006). The high fear tension ad showed a grievous consequence in the form of a scarred face, while the moderate fear tension ad showed the same face without scars. To increase relevance to participants, the face in the ad was male for male participants and female for female participants (Appendices C–F show male versions; female versions are available from the authors). A professional makeup artist and photographer developed the pictures using student models. The presence/absence of humor was manipulated by adding/not adding a cartoon that showed humorous use of sunscreen lotion on a beach. This beach-related cartoon was integrated with the ad copy by linking it to a sentence in the ad that read as follows: “Always keep it handy at beaches, pools, and parks.” Notably, sunscreen lotion ads with the threat of skin cancer have been used in previous research on fear advertising (Strong and Dubas, 1993). The fear tension arousal manipulation was checked in a pilot study where 162 students from the same population rated the high- or moderate-fear ad (without the cartoon) on a fear tension arousal scale validated by Keller and Block (1996) and anchored by very unafraid/afraid, relaxed/tense, calm/agitated, and restful/excited (a = 0.85). Similar items for fear tension arousal have been used in previous research (e.g., LaTour and Rotfeld, 1997; Potter et al. 2006). Supportive of the manipulation, scale means differed significantly between the high and moderate fear tension arousal conditions (M = 6.93 vs. 5.80, t(160) = 6.08, p < 0.001). Moreover, the rating in the moderate fear tension arousal condition was significantly higher than the neutral midpoint of the scale (M = 5.80 vs. 5.0, t(79) = 5.78, p < 0.001). We also checked J. Consumer Behav. 11: 147–161 (2012) DOI: 10.1002/cb 150 A. Mukherjee and L. Dubé the fear tension arousal manipulation in the main study, the results of which are reported hereafter. The humor manipulation was checked in the main study using a three-item humor scale (Chattopadhyay and Basu, 1990) anchored by not at all/ very funny, humorous, and amusing (a = 0.88); results of this manipulation check are reported hereafter. Dependent and mediating variables The dependent variable of persuasion was measured by attitude toward the brand (Abr), with items good/bad, like/ dislike, and desirable/undesirable (a = 0.88). We had conceptualized the mediating variable of defensive response in terms of message elaboration and perceived vulnerability to threat. In the present study, we measured message elaboration, that is, positive thoughts about the advertised brand; in the next study, we measure both message elaboration and perceived vulnerability. Two coders blind to experimental conditions counted the number of positive, negative, and neutral thoughts about the advertised brand in the openended thought listings. An example of a positive brand thought was “I might try this product if I see it at the pharmacy”; negative brand thought was “I’m hesitant about using an unknown product on my face”; and neutral thought was “I wonder if this lotion is by L’Oreal.” Intercoder agreement was 86 per cent, and disagreements were resolved within coders. Net positive brand thoughts, that is, the difference between positive and negative brand thoughts, were used as the measure of message elaboration (Agrawal and Maheshwaran, 2005). We also measured self-efficacy and response efficacy because fear theories such as protection motivation theory have argued that these variables can influence fear response (e.g., Tanner et al., 1991; Witte and Allen, 2000). Selfefficacy was measured by “It is difficult for me to use sunscreen lotion when I am exposed to strong sunlight (strongly agree/disagree),” and response efficacy was measured by “Using sunscreen lotion with the right SPF factor can protect my skin against the harmful effects of strong sunlight (strongly agree/disagree).” Results showed no significant differences of self- and response efficacy across experimental conditions; hence, they are not discussed further. All items were measured on nine-point scales. Results As in the pilot study, the manipulation of fear tension arousal was checked in the main study by comparing moderate and high fear tension conditions in the absence of humor. The manipulation was successful, with a significant difference in fear tension arousal between the moderate- and high-fear conditions (M = 5.67 vs. 6.91, t(60) = 4.59, p < 0.001). Moreover, the level of fear tension arousal in the moderatefear condition was greater than the neutral midpoint of the scale (M = 5.67 vs. 5.0, t(30) = 3.35, p < 0.002). Similarly, the humor manipulation was successful, with the mean humor rating higher than the neutral midpoint of the scale (M = 6.02 vs. 5.0, t(61) = 6.80, p < 0.001). Notably, the humor scale was administered in the humor-present condition because participants could rate humor only when it was present. Means and standard deviations of the outcome variables are summarized in Table 1 and Figure 1. Hypothesis 1 posited that high (compared to moderate) fear tension arousal decreases persuasion when humor is absent but increases persuasion when humor is present. This hypothesis was tested by the interaction of humor and fear tension arousal, with Abr as the dependent variable. Analysis of variance showed a significant interaction of humor and fear tension arousal (F(1, 120) = 10.70, p < 0.001), along with a main effect of humor (F(1,120) = 18.33, p < 0.001) and no main of fear tension arousal (F(1,120) = 0.48, p < 0.48). Consistent with H1, increasing fear tension arousal from moderate to high decreased Abr when humor was absent (M = 5.51 vs. 5.07, t(60) = 2.08, p < 0.04) but increased Abr when humor was present (M = 5.68 vs. 6.36, t(60) = 2.52, p < 0.01). H2 posited that the interaction of humor and fear tension arousal is mediated by reduction in defensive response to the ad. In this study, we measured defensive response by net positive thoughts about the advertised brand. H2 was supported by a four-step mediation Interaction of Humor and Fear: Study 1 Attitude towards the Brand (Abr) Humor Present 6.36 Humor Absent 5.68 5.51 5.07 Fear Tension Arousal High Moderate Figure 1. Interaction of humor and fear: Study 1. Table 1. Humor in fear advertising: Study 1 Humor absent Humor present Outcome variable Moderate fear High fear Moderate fear High fear Attitude toward the brand Positive brand thoughts Fear tension arousal Humor 5.51 (0.78) 1.93 (1.18) 5.67 (1.12) 5.07 (0.87) 1.35 (0.83) 6.91 (0.98) 5.68 (1.01) 1.87 (1.08) 5.39 (1.16) 6.09 (1.11) 6.36 (1.09) 2.41 (1.28) 6.37 (0.90) 5.94 (1.25) Note: Values in cells are means (standard deviations). Copyright © 2012 John Wiley & Sons, Ltd. J. Consumer Behav. 11: 147–161 (2012) DOI: 10.1002/cb Fear and humor analysis (Baron and Kenny, 1986) (see Table 1 for means and standard deviations). First, as reported earlier, humor and fear tension arousal interacted to influence Abr (F(1, 120) = 10.70, p < 0.001). Second, humor and fear tension arousal interacted to influence net positive brand thoughts (F (1, 120) = 8.01, p < 0.005). Third, net positive brand thoughts had a positive effect on Abr (F(1, 122) = 35.38, p < 0.001). Fourth, the interactive effect of humor and fear tension arousal on Abr was reduced when net positive brand thoughts was added as a covariate (F(1, 119) = 5.12, p < 0.02); a Sobel test indicated that this reduction was significant (Sobel z = 2.55, p < 0.01). Thus, in summary, H1 and H2 were supported by the data. An alternative mechanism for humor in fear advertising is suggested by research on humor as a desensitizer of fear (e.g., Ventis et al., 2001). According to this view, humor could reduce fear tension arousal in our high-fear condition to a moderate level, and in our moderate-fear condition to a low level. Such results could explain our observed positive effect of fear on persuasion, based on the inverted U theory of fear appeals (Janis, 1967; Sternthal and Craig, 1974). We tested this alternative mechanism by analyzing responses to the fear tension arousal scale, in the presence and absence of humor (see Table 1). As shown in Table 1, the fear tension arousal means in the presence of humor (M = 5.39 vs. 6.37, t(60) = 3.68, p < 0.001) followed a similar pattern as the corresponding means in the absence of humor (M = 5.67 vs. 6.91, t(60) = 4.59, p < 0.001). Thus, the high-fear condition in the presence of humor continued to be relatively high on fear tension arousal; similarly, the moderate-fear condition in the presence of humor continued to be moderate on fear tension arousal. Furthermore, mediation analysis did not support a fear reduction mechanism because the interaction of humor and fear tension on the fear tension arousal scale was not significant (F(1,120) = 0.46, p < 0.49). These results do not support a fear reduction mechanism underlying the persuasive effect of humor in fear advertising. We analyzed the fear reduction mechanism again in study 2, with similar results as in the present study. A possible risk with humor in fear advertising is that the audience could consider humor to be inappropriate in advertising containing health threats such as skin cancer. We investigated this by measuring perceived inappropriateness of humor in the ad, via respondents’ thought listings. Two coders counted the number of respondents who expressed disapproval of using humor in an ad featuring skin cancer. Coders consensually identified 4 respondents, out of 62 in the humor-present condition, who felt that humor was inappropriate in the ad. When the data were reanalyzed excluding these respondents, the results were essentially unchanged. Based on this result, we concluded that inappropriateness of humor was not a significant issue in the present study. Notably, past research has also used humor in advertising with a health threat, e.g., humorous cartoons showing condom use as protection against AIDS (Conway and Dubé, 2002). We further discuss the issue of appropriateness of humor in the future research section. The next study (study 2) was designed with two objectives in mind. First, study 2 manipulated fear tension Copyright © 2012 John Wiley & Sons, Ltd. 151 arousal differently, this time in the context of the target ad. Recall that fear tension was manipulated in study 1 within the target ad, by showing either a scarred or unscarred skin cancer patient. However, real-world advertising is often seen in the context of television programming or editorial content that can influence audience responses to the ad. For example, an ad for sunscreen lotion may be seen during a television program on skin cancer and/or in a magazine with a cover story on UV ray exposure. Such contexts can prime individuals to be more sensitive to fear appeals in the target ad, thus heightening the level of fear tension arousal (Miniard et al., 1991; Lord et al., 1994). Consistent with this argument, Potter et al. (2006) found that fear ads created more tension arousal in a negative program context than in a positive program context. A contextual manipulation of fear tension arousal—using the same stimulus ad—would also be consistent with researchers who have argued that fear response to a given threat stimulus can depend on contextual factors (Rotfeld, 1988; LaTour and Rotfeld, 1997). We addressed this issue in study 2 by contextually manipulating fear tension arousal. A second objective of study 2 was to further probe the process underlying humor in fear advertising. The results of study 1 indicate that the effect of humor in fear advertising is mediated by positive brand-related thoughts, that is, message elaboration. As discussed earlier, another way in which humor could work in fear advertising is by increasing perceived vulnerability to threat. Past research indicates that denying one’s personal vulnerability is a common avoidance response to threat (Keller, 1999; Duhachek, 2005). A reduced sense of vulnerability, in turn, can reduce adoption of the recommended solution. Humor could break this chain of reduced vulnerability leading to reduced persuasion. If humor provides a safe context for individuals to think about the message in the ad, then such individuals should be better able to acknowledge their vulnerability to the threat. As perceived vulnerability rises, likelihood of adopting the proposed solution should also increase. We tested this mediating process in study 2 by measuring respondents’ perceived vulnerability to skin cancer. STUDY 2 Design and participants This study was designed as a 2 (fear tension arousal: moderate vs. high)
2 (humor: absent vs. present) betweensubjects factorial. 132 undergraduate students (33 per group) at a large university in Canada participated in the study for a $10 incentive. The sample included 68 females and 64 males, with an age range of 18—24 years. Procedure Participants were told that the experiment consisted of two studies. The first study was described as a pretest to select a spokesperson for a skin cancer awareness campaign. The spokesperson was said to be a real cancer patient; information about this patient manipulated fear tension in a contextual manner. Next, participants started the second study said to be J. Consumer Behav. 11: 147–161 (2012) DOI: 10.1002/cb 152 A. Mukherjee and L. Dubé an advertising test for a new brand of sunscreen lotion. The ad in this study manipulated the presence/absence of humor. Subsequently, participants responded to questions measuring persuasion, perceived vulnerability, and their thoughts and feelings while reading the ad. Finally, participants completed the manipulation checks and were asked to guess the real purpose of the studies. Independent variables Fear tension arousal was contextually manipulated in the patient information provided to participants before they saw the target ad. The high fear tension version showed the patient suffering from grievous consequences of skin cancer, including disfiguring scars, while the moderate fear tension version showed relatively minor consequences of skin cancer without scarring (Appendices G and H show male versions; female versions are available from the authors). To ensure comparability with the fear tension manipulation of study 1, the same scarred (unscarred) picture was used in high (moderate) fear tension condition of study 2. The name of the patient was redacted to increase face validity in terms of patient privacy. After reading the patient information, participants answered a filler question asking them to judge similarity in age between themselves and the patient. Next, humor was manipulated in the ad for sunscreen lotion shown to participants in the second part of the study (see Appendices I and J). The humor-present and humor-absent versions of the ad were the same as in study 1, but without the scarred/unscarred picture, which was now being used in the contextual manipulation of fear tension. The humor and fear tension arousal manipulations were checked with the same scales as in study 1. Dependent and mediating variables The dependent variable of persuasion was measured by the three-item scale for Abr used in study 1. Inasmuch as the target ad could also be persuasive by increasing the intent to use sunscreen lotion in general, as opposed to a specific brand of sunscreen lotion, sunscreen usage intent was measured with a three-item scale validated by Keller (2006): “I intend to use sunscreen,” “I intent to buy a bottle of sunscreen if I don’t have any,” and “I intend to check whether I have sunscreen on before I go out.” The first mediating variable of message elaboration was measured, as in study 1, by coding net positive brand thoughts. The second mediating variable of perceived vulnerability was measured with a three-item scale adopted from Luce and Kahn (1999); this assessed the likelihood of getting skin cancer at some point in an individual’s life, the possibility of being diagnosed with skin cancer at some point in an individual’s life, and an individual’s personal vulnerability to getting skin cancer (a = 0.78). Finally, we measured mood at the end of the study using a two-item scale (Mackie and Worth, 1989): “At the present moment, I feel very sad/very happy, not at all pleased/very pleased” (a = 0.90). The only mood effect was a positive main effect of humor on mood, with no other main or interaction effects relevant to our hypotheses; thus, mood is not discussed further. All items were measured with nine-point scales. Results As in study 1, the fear tension arousal manipulation was checked by comparing the moderate and high fear tension conditions in the absence of humor. The fear tension manipulation was successful with a significant difference in fear tension arousal between moderate- and high-fear conditions (M = 5.85 vs. 6.95, t(64) = 3.84, p < 0.001). Moreover, fear tension in the moderate-fear condition was significantly greater than the neutral midpoint of the scale (M = 5.85 vs. 5.0, t(32) = 4.16, p < 0.001). The humor manipulation was also successful, with the mean humor rating significantly greater than the neutral midpoint of the scale (M = 6.03 vs. 5.0, t(65) = 7.04, p < 0.001). Participants’ statements about the purpose of the study indicated that they did not guess its true purpose, which was to investigate the interactive effect of humor and fear in advertising. The means and standard deviations of the outcome variables are summarized in Table 2. Hypothesis 1 was tested by the interaction of humor and fear tension arousal, with Abr and sunscreen usage intent as dependent variables. With Abr as a dependent variable, there was a significant interaction of humor and fear tension (F(1, 128) = 6.99, p < 0.01), along with a main effect of humor (F(1,128) = 16.21, p < 0.001) and no main effect of fear tension (F(1,128) = 0.32, p < 0.56). Consistent with H1, increasing the level of fear tension arousal decreased Abr marginally when humor was absent (M = 5.44 vs. 5.17, t(64) = 1.72, p < 0.08) but increased Abr when humor was present (M = 5.62 vs. 6.05, t(64) = 2.01, p < 0.04). The results were similar with sunscreen usage intent as the dependent variable. There was a significant interaction of humor and fear tension (F(1, 128) = 7.60, p < 0.007), a main effect of Table 2. Humor in fear advertising: Study 2 Humor absent Humor present Outcome variable Moderate fear High fear Moderate fear High fear Attitude toward the brand Usage intent for sunscreen Positive brand thoughts Perceived vulnerability Fear tension arousal Humor 5.44 (0.75) 5.33 (0.70) 1.81 (0.98) 2.44 (0.86) 5.85 (1.18) 5.17 (0.50) 5.04 (0.55) 1.57 (0.93) 2.10 (0.66) 6.95 (1.14) 5.62 (0.85) 5.53 (0.93) 1.96 (1.15) 2.54 (0.79) 5.64 (1.12) 6.11 (1.03) 6.05 (0.86) 6.03 (1.00) 2.42 (1.37) 3.03 (1.04) 6.48 (1.08) 5.95 (1.34) Note: Values in cells are means (standard deviations). Copyright © 2012 John Wiley & Sons, Ltd. J. Consumer Behav. 11: 147–161 (2012) DOI: 10.1002/cb Copyright © 2012 John Wiley & Sons, Ltd. Usage intent Abr Perceived vulnerability Usage intent Humor
fear ! Abr, p < 0.01 Humor
fear ! positive brand thoughts, p < 0.07 Humor
fear ! usage intent, Humor
fear ! positive p < 0.007 brand thoughts, p < 0.07 Humor
fear ! Abr, p < 0.01 Humor
fear ! perceived vulnerability, p < 0.006 Humor
fear ! usage intent, Humor
fear ! perceived p < 0.007 vulnerability, p < 0.006 Abr Mediating variable Outcome variable Table 3. Mediation analyses: Study 2 Fear is used for advertising many products, services, and causes such as safe driving, antismoking, and sunscreen and condom usage. Past research indicates that high levels of fear tension arousal prompt defensive responses in the audience, which, in turn, reduce the persuasive effect of the ad. In the present research, we show that humor can reduce these defensive responses and hence increase the persuasive effect of fear advertising. The results of two studies using sunscreen lotion ads showed that increasing fear tension arousal from moderate to high increased persuasion when humor was included in the ad but decreased persuasion when there was no humor in the ad. This result was robust across two manipulations of fear tension arousal—within ad and contextual—as well as across attitude and intent measures of persuasion. Further, results of both studies showed that the interaction of humor and fear tension was mediated by reduction in defensive responses to the ad, measured by positive thoughts about the brand and perceived vulnerability to the threat. These results make theoretical and practical contributions to the literature on fear appeals. Our results make a theoretical contribution by identifying a new moderator of the persuasiveness of fear advertising, namely, the absence or presence of humor in the ad. Step 1 DISCUSSION AND IMPLICATIONS Positive brand thoughts Step 2 Step 3 Step 4 humor (F(1, 128) = 17.41, p < 0.001), and no main effect of fear tension (F(1, 128) = 0.50, p < 0.48). Consistent with H1, increasing fear tension arousal marginally reduced sunscreen usage intent when humor was absent (M = 5.33 vs. 5.04, t(64) = 1.87, p < 0.06) but increased sunscreen usage intent when humor was present (M = 5.53 vs. 6.03, t(64) = 2.06, p < 0.04). H2 posited that the interactive effect of humor and fear tension on persuasion is mediated by reduction in defensive responses to the ad. In this study, we measured defensive responses by the number of positive brand thoughts and perceived vulnerability to skin cancer. The results of fourstep mediation analyses were consistent with H2 for both mediating variables, with significant or marginally significant results at all steps. Table 3 summarizes the results of the mediation analyses (see also Table 2 for means and standard deviations). As in study 1, we tested an alternative fear-reduction mechanism in which humor transforms moderate/high fear tension into low/moderate fear tension. However, the data did not support this mechanism (see Table 1). Similar to the results in study 1, fear tension arousal means in the presence of humor (M = 5.64 vs. 6.48, t(64) = 3.08, p < 0.003) were similar to corresponding means in the absence of humor (M = 5.85 vs. 6.95, t(64) = 3.84, p < 0.001). Furthermore, mediation analysis did not support a fear-reduction mechanism, with a nonsignificant interactive effect of humor and fear tension on the fear tension arousal scale (F(1,128) = 0.44, p < 0.50). Taken together, these results do not support a fear-reduction mechanism. Instead, as shown in Table 3, mediation analysis indicates that reduction in defensive response underlies the effect of humor in fear advertising. 153 Positive brand thoughts ! Abr, p < 0.001 Humor
fear ! Abr (covariate: positive brand thoughts), p < 0.04, Sobel p < 0.08 Positive brand thoughts ! usage intent, Humor
fear ! usage intent (covariate: positive p < 0.001 brand thoughts), p < 0.01, Sobel p < 0.10 Perceived vulnerability ! Abr, p < 0.001 Humor
fear ! Abr (covariate: perceived vulnerability), p < 0.03, Sobel p < 0.02 Perceived vulnerability ! usage intent, Humor
fear ! usage intent (covariate: perceived p < 0.001 vulnerability), p < 0.07, Sobel p < 0.01 Fear and humor J. Consumer Behav. 11: 147–161 (2012) DOI: 10.1002/cb 154 A. Mukherjee and L. Dubé Consistent with past research, we found that ads generating high (compared to moderate) fear tension were less effective in terms of persuasion. In contrast, and extending past research, we found that ads generating high (compared to moderate) fear tension were more effective when humor was mixed into the ad. This result is important because marketers often use advertising materials that can create high fear tension, such as graphic pictures in antismoking, drinking and driving, and health-related ads. Furthermore, past research indicates that a given fear ad can prompt high fear tension in particular audiences and contextual conditions (Rotfeld, 1988; LaTour and Rotfeld, 1997). We show that humor can provide a buffer against the negative effects of high fear tension and hence improve the effectiveness of fear advertising. Our results add to past research on methods to increase the effectiveness of fear advertising, such as self versus other reference (Keller and Block, 1996) and relief in the ad (Rossiter and Thornton, 2004). For example, as discussed earlier, Keller and Block (1996) reported greater effectiveness of a high-fear antismoking advertisement when the ad used a statement of the form “Cigarette smoking is dangerous to the health of those close to you” rather than “Cigarette smoking is dangerous to your health.” Similarly, Rossiter and Thornton (2004) showed that a high-fear safe driving ad worked better with relief at the end of the ad, namely, the positive outcome of avoiding accidents at lower speeds. In addition to methods such as self-reference and relief, we show that humor can be used as a tool to increase the effectiveness of fear advertising. Our results also extend past research indicating that humor in fear advertising improves persuasion for individuals high in masculinity (Conway and Dubé, 2002). In addition to personality factors such as masculinity, we show that stimulus factors such as the level of fear tension arousal can also influence the effectiveness of humor in fear advertising. Our studies provide insight into the mechanism underlying the effect of humor in fear advertising. We found that the interactive effect of humor and fear tension arousal on persuasion is mediated by defensive responses to the ad. This mediation was replicated for two measures of defensive responses—message elaboration and perceived vulnerability to threat—as well as two measures of persuasion, namely, Abr and usage intent. These results support the idea that playfulness of humor provides a safe context for the audience, within which they can elaborate on the threatening message and acknowledge their personal vulnerability to the threat. These audience responses, in turn, increase the persuasive effect of the ad. Notably, data did not support two affective mechanisms—one based on mood and the other based on fear reduction—that could potentially underlie humor in fear advertising. Regarding mood, although humor improved mood as a main effect, the interaction of humor and fear tension arousal was not mediated by mood. Regarding fear reduction, as may be expected by affect transfer, the addition of humor to the high- and medium-fear ads reduced the level of experienced fear to some extent. However, this reduction of fear was not significant. Consequently, humor did not transform high fear tension to a moderate level, or moderate Copyright © 2012 John Wiley & Sons, Ltd. fear tension to a low level, thus ruling out a fear-reduction explanation for our results. A possible reason for the sustained levels of fear tension in the presence of humor is that humor does not change the reality of the threat shown in the ad. For example, the humorous cartoon used in our studies did not change the fact that skin cancer can happen if people stay in the sun too long. Thus, it was not surprising that the presence of humor did not significantly dilute the level of fear tension arousal. At the broadest level, our results indicate that fear increases motivation to process, while humor increases ability to process the message in the ad. Consistent with this view, we found that the combination of humor and high fear tension arousal was most effective in terms of persuasion, outperforming high or moderate fear tension arousal alone. Our studies have practical implications for advertisers who use fear in their ads. Many advertising campaigns on topics such as sunscreen, drunk driving, insurance, and antismoking have relied exclusively on fear to motivate desired outcomes (e.g., Appendix B). The present research suggests that such fear-only messages can backfire if they generate high levels of fear tension arousal, or if the context magnifies fear tension arousal evoked by the ad. Our results also suggest that adding an element of humor to fear advertising can mitigate this problem by reducing defensive responses in the audience, hence increasing advertising effectiveness. Notably, recent campaigns have started using humor and fear together in the ad (e.g., Appendix A), and our research provides conceptual understanding of the effectiveness of such mixed emotions in advertising. FUTURE RESEARCH The studies herein suggest several avenues for future research. First, as discussed earlier, fear is a psychophysiological construct, with tension arousal and energy arousal as its underlying components (Henthorn et al., 1993; LaTour and Rotfeld, 1997). In the present research, we focused on tension arousal and its interaction with humor. Future research could investigate the interaction of energy arousal with humor. Arousal theory of humor posits that humor involves an initial increase of energy arousal, followed by a decrease of energy arousal at the end of the joke (Speck, 1991; Woltman-Elpers et al., 2004). Hence, it is possible that adding humor to a fear ad can reduce the net level of energy arousal at the end of the ad. Inasmuch as energy arousal is said to be related to Aad (LaTour and Rotfeld, 1997), humor could influence attitude toward fear advertising through the energy arousal route. Future research could investigate this hypothesis through physiological measures of energy arousal in the presence or absence of humor (LaTour et al., 1990; Potter et al. 2006). In the present studies, we manipulated high fear tension arousal using a graphic picture of a scarred skin cancer patient, who was similar to the participant in terms of age and gender. Future research could replicate the effect of fear tension arousal on persuasion using verbal (i.e., words) versus visual (i.e., pictures) stimuli to manipulate fear arousal. This would clarify whether visual stimuli J. Consumer Behav. 11: 147–161 (2012) DOI: 10.1002/cb Fear and humor are necessary to generate the levels of fear tension arousal at which the audience exhibits defensive responses to the ad. A second avenue for future research pertains to the ethicality of using humor in fear advertising. Past research has examined the ethics of fear in advertising (Hackley and Kitchen, 1999; Hastings et al., 2004). According to this research, fear advertising could be unethical when it has negative effects on vulnerable sections such as children, elderly, infirm, and addicts. For example, fear advertising could create chronic anxiety in vulnerable audience members when they see grievous harm being visited on others. In our research, we focus on the use of humor in fear advertising. Audiences could potentially find humor in our fear ad to be unethical if they view humor as making light of the serious problem of skin cancer or making fun of the misfortune of a skin cancer patient. As reported earlier, we empirically investigated this issue in study 1 by analyzing thought listings of respondents. Only 6 per cent of respondents stated in their thought listings that humor was an inappropriate element in the ad, and our results were unchanged when these respondents were excluded from the analysis. This thought listing analysis indicated that ethicality may not be a major issue with our stimuli. However, it should be noted that thought listing is an unprompted measure because respondents were not directly asked about the inappropriateness of humor in the ad. Thus, it is possible that respondents thought that humor was inappropriate but did not write down such thoughts because they were not asked directly about it. To address this issue, we conducted a follow-up study in which 31 student respondents were shown the high fear and humor ad from study 1 (i.e., Appendix D) and asked to rate this ad on two items: “In your opinion, the use of humor in this ad is not at all appropriate/very appropriate (1–9), not at all proper/ very proper (1–9).” The two items were reliable (a = 0.89) and were combined into an index of humor ethicality in fear advertising (e.g., Hastings et al., 2004). The mean response on this index (6.01) was significantly greater than the neutral midpoint represented by “somewhat appropriate/proper” (6.01 vs. 5.0, t(30) = 3.76, p < 0.001). This suggests that respondents felt the use of humor in our fear ad was relatively ethical. Notably, past research has also used humor in advertising with a health threat, for example, humorous cartoons showing condom use against AIDS (Conway and Dubé, 2002). It is worth speculating why respondents felt that humor was relatively ethical in our fear ad. One reason could be an implicit assumption on the part of the respondents that the skin cancer patient had given permission for using his or her picture in the ad. If permission was willingly granted by the patient, it would be difficult to conclude that the ad was exploiting the patient’s suffering. Another reason for respondents’ relative acceptance of humor in our fear ad could be that the end justifies the means. In other words, the objective of preventing skin cancer could be important enough to permit the use of humor in the ad. Future research could investigate these reasons for ethicality of humor in fear advertising. For example, studies could manipulate whether permission was granted by the spokesperson versus not granted, and also the salience of health benefit to the audience versus profits to the marketer. Another moderator of Copyright © 2012 John Wiley & Sons, Ltd. 155 ethicality and hence effectiveness of humor in fear advertising could be the type of humor in the ad. Two types of humor often used in social and commercial contexts are satire and slapstick (Speck, 1991). While satire pokes fun at people’s weaknesses, slapstick capitalizes on amusing situations and events. It is possible that slapstick—which was used in our humorous cartoon—is more acceptable in fear advertising than satirical humor. Inasmuch as satire is based on criticism of others’ weaknesses, it may be construed as making fun of a vulnerable victim in the ad. Hence, future research could investigate the relative effectiveness of slapstick versus satire in fear advertising. Similarly, future research could investigate the use of ironic humor to express threats in a thought-provoking manner. For example, consider a popular antismoking ad that shows a father and son playing together. When the father drops to the base of a tree to light up, the son does too. The voiceover then says “Like father, like son.” This ironic approach, using a wellknown aphorism to illustrate negative family influence, may be more effective than a direct statement of the dangers of smoking in front of your children. Another potential moderator of humor in fear advertising is the relevance of humor. Past research indicates that humor is more effective when it is relevant to product usage or product benefits (Weinberger et al., 1995; Cline and Kellaris, 2007). The humor in our studies was designed to show product usage, with the cartoon illustrating amusing uses of sunscreen lotion on the beach. Future research could vary the relevance of humor in terms of product usage as well as benefits. In addition to relevance, the usage frequency of humor could also influence its effectiveness in fear advertising. Currently, humor is a relatively novel element in fear advertising. If humor is used more frequently in fear advertising, its value as a buffer against threat may go down, which in turn could reduce its persuasive effect. This question can be investigated using a repeated-measures design, where participants are shown several humor and fear ads in succession such that the novelty of this combination is progressively lessened over time. Finally, future research could assess the behavioral effects of humor in fear advertising. Persuasion was measured in our studies through Abr and sunscreen usage intent. Conceptual frameworks such as the theory of reasoned action (Ajzen and Fishbein, 1980) posit that attitudes and intent predict behavior. Consistent with this view, meta-analysis has found a correlation of 0.44 between intention and behavior in the case of condom use (Sheeran and Orbell, 2000). Research also indicates that intentions have a stronger link to behavior when—as in our studies—the communication focuses on encouraging rather than discouraging behaviors (Floyd et al., 2000). A possible moderator of the behavioral effect of humor is complexity of the behavior promoted in the ad. Past research indicates that humor can sometimes distract the audience from learning the message in the ad (Chattopadhyay and Basu, 1990). The distraction effect of humor is likely to be a bigger issue when the behavior promoted in the ad is relatively complex, such as safe driving, compared to relatively simple behaviors such as applying sunscreen lotion. Hence, future research could compare the effectiveness of humor in fear advertising that J. Consumer Behav. 11: 147–161 (2012) DOI: 10.1002/cb 156 A. Mukherjee and L. Dubé promotes simple versus complex behaviors. In conclusion, we can summarize the present research as follows: humor provides a buffer against the negative effects of high fear tension arousal and hence maximizes the persuasive effect of fear advertising. APPENDIX A Anti-Fur Ad ACKNOWLEDGEMENT This research was supported by a grant from the Social Sciences and Humanities Research Council of Canada. APPENDIX C High Fear and Humor Absent: Male APPENDIX B Sun Safety Ad Copyright © 2012 John Wiley & Sons, Ltd. J. Consumer Behav. 11: 147–161 (2012) DOI: 10.1002/cb Fear and humor 157 APPENDIX D High Fear and Humor Present: Male APPENDIX E Medium Fear and Humor Absent: Male Copyright © 2012 John Wiley & Sons, Ltd. J. Consumer Behav. 11: 147–161 (2012) DOI: 10.1002/cb 158 A. Mukherjee and L. Dubé APPENDIX F Medium Fear and Humor Present: Male APPENDIX G APPENDIX H High Fear Patient: Male Medium Fear Patient: Male Patient ID#: 45932MTQ Patient ID#: 45932MTQ Name: John Smith Name: John Smith Age: 25 Years Age: 25 Years Gender: Male Diagnosis: Malignant skin cancer lesions, left side of face Treatment: Major surgical excision (with stitches) Copyright © 2012 John Wiley & Sons, Ltd. Gender: Male Diagnosis: Benign melanoma spot, lower back Treatment: Minor procedure (removal) J. Consumer Behav. 11: 147–161 (2012) DOI: 10.1002/cb Fear and humor 159 APPENDIX I Humor Absent APPENDIX J Humor Present Copyright © 2012 John Wiley & Sons, Ltd. J. Consumer Behav. 11: 147–161 (2012) DOI: 10.1002/cb 160 A. Mukherjee and L. Dubé REFERENCES Agrawal N, Maheshwaran M. 2005. Motivated reasoning in outcome bias effects. Journal of Consumer Research 31(March): 798–805. Ajzen I, Fishbein M. 1980. Understanding attitudes and predicting social behavior. Prentice Hall: Englewood Cliffs, NJ. Alden DL, Hoyer WD, Lee C. 1993. Identifying global and culturespecific dimensions of humor in advertising: A multinational analysis. Journal of Marketing 57(April): 64–75. Alden DL, Mukherjee A, Hoyer WD. 2000. The effects of incongruity, surprise and positive moderators on perceived humor in advertising. Journal of Advertising 29(Summer): 1–16. Baron RM, Kenny DA. 1986. The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology 51(6): 1173–1182. Burnett JJ, Oliver RJ. 1979. Fear appeal effects in the field: A segmentation approach. Journal of Marketing Research 16(May): 181–190. Chapman AJ. 1996. Social aspects of humorous laughter. In Chapman AJ, Foot HC (eds). Humor and Laughter: Theory, Research and Applications. Transaction Publishers: New Brunswick, NJ; 155–186. Chattopadhyay A, Basu K. 1990. Prior brand evaluation as a moderator of the effects of humor in advertising. Journal of Marketing Research 26(4): 466–476. Christenson LJ. 2005. Incidence of Basal Cell and Squamous Cell Carcinomas in a Population Younger than 40 Years. Journal of the American Medical Association 294(6): 681–690. Cline TW, Kellaris JJ. 2007. The influence of humor strength and humor-message relatedness of ad memorability. Journal of Advertising 36(1): 55–67. Conway M, Dubé L. 2002. Humor in persuasion on threatening topics: Effectiveness is a function of audience sex role orientation. Personality and Social Psychology Bulletin 28(July): 863–873. Dixon NF. 1980. Humor: A cognitive alternative to stress? In Saranson IG, Spielberger CD (eds). Stress and anxiety. Hemisphere: Washington; 281–289. Duhachek A. 2005. Coping: A multidimensional, hierarchical framework of responses to stressful consumption episodes. Journal of Consumer Research 32(June): 41–53. Floyd DL, Prentice-Dunn S, Rogers RW. 2000. A meta-analysis of research on protection motivation theory. Journal of Applied Social Psychology 30(2): 407–429. Folkman S, Lazarus RS. 1984. Stress, Appraisal and Coping. Springer Publishing Company: New York. Hastings G, Stead M, Webb J. 2004. Fear appeals in social marketing: Strategic and ethical reasons for concern. Psychology and Marketing 21(11): 961–986. Hackley CE, Kitchen PJ. 1999. Ethical perspectives on postmodern communications. Journal of Business Ethics 20: 252–254. Henthorn TL, LaTour MS, Nataraajan R. 1993. Fear appeals in print advertising: An analysis of arousal and elaboration. Journal of Advertising 22(2): 56–69. Horowitz M. 1983. Psychological responses to serious life events. In Breznitz S (ed). The Denial of Stress. International Universities Press: New York; 129–161. Janis IL. 1967. Effects of fear arousal on attitude change: Recent developments in theory and experimental research. In Berkowitz L (ed). Advances in Experimental Social Psychology 3. Academic Press: New York; 166–224. Kellaris JJ, Cline TW. 2007. Humor and ad memorability: On the contributions of humor expectancy, relevancy, and need for humor. Psychology and Marketing 24(6): 497–509. Keller PA. 1999. Converting the unconverted: The effect of inclination and opportunity to discount health-related fear appeals. Journal of Applied Psychology 84(3): 403–415. Keller PA. 2006. Regulatory focus and the efficacy of health messages. Journal of Consumer Research 33(June): 109–114. Keller PA, Block L. 1996. Increasing the persuasiveness of fear appeals: The effect of arousal and elaboration. Journal of Consumer Research 22(March): 448–460. Copyright © 2012 John Wiley & Sons, Ltd. Kuiper NA, Martin RA, Olinger LJ. 1993. Coping, humour, stress, and cognitive appraisals. Canadian Journal of Behavioral Science 25: 81–96. LaTour MS, Pitts RE. 1989. Using fear appeals in advertising for AIDS prevention in college-age population. Journal of Health Care Marketing 9(September): 5–14. LaTour MS, Pitts RE, Snook-Luther DC. 1990. Female nudity, arousal, and ad response: An experimental investigation. Journal of Advertising 36(March/April): 59–67. LaTour MS, Rotfeld HJ. 1997. There are threats and (maybe) fearcaused arousal: Theory and confusions of appeals to fear and fear arousal itself. Journal of Advertising 26(3): 45–59. LaTour MS, Tanner JF. 2003. Radon: Appealing to our fears. Psychology and Marketing 20(5): 377–394. Lazarus RS. 1983. The costs and benefits of denial. In Breznitz S (ed). Denial of Stress. International Universities Press: New York; 1–30. Leventhal H. 1970. Findings and theory in the study of fear communications. In Berkowitz L (ed). Advances in Experimental Social Psychology. Academic Press: New York; 119–186. Lord KR, Lee MS, Sauer PL. 1994. Program context antecedents to attitudes towards radio commercials. Journal of the Academy of Marketing Science 22(1): 3–16. Luce MF, Kahn BE. 1999. Avoidance or vigilance? The psychology of false-positive test results. Journal of Consumer Research 26(3): 242–259. Mackie DM, Worth LT. 1989. Processing Deficits and the Mediation of Positive Affect in Persuasion. Journal of Personality and Social Psychology 57(1): 27–40. Martin RA, Dobbin JP. 1988. Sense of humor, hassles, and immunoglobulin: Evidence for a stress-moderating effect of humor. Journal of Psychiatry and Medicine 18(2): 93–105. McMath BF, Prentice-Dunn S. 2005. Protection motivation theory and skin cancer risk: The role of individual differences in response to persuasive appeals. Journal of Applied Social Psychology 35: 621–635. Miniard PW, Bhatla S, Lord KR, Dickson PR, Unnva HR. 1991. Picture-based persuasion processes and the moderating role of involvement. Journal of Consumer Research 18: 92–107. Passyn K, Sujan M. 2006. Self accountability emotions and fear appeals: Motivating behavior. Journal of Consumer Research 32(March): 583–589. Potter RF, LaTour MS, Braun-LaTour KA, Reichert T. 2006. The impact of program context on motivational system activation and subsequent effects on processing a fear appeal. Journal of Advertising 35(3): 67–80. Raskin V. 1985. Semantic Mechanisms of Humor. D. Reidel: Boston. Rippetoe PA, Roger RW. 1987. Effects of components of protection-motivation theory on adaptive and maladaptive coping with a health threat. Journal of Personality and Social Psychology 52(March): 595–604. Rogers RW. 1985. Attitude change and information on integration in fear appeals. Psychological Reports 56(1): 179–182. Rotfeld H. 1988. Fear appeals and persuasion: Assumptions and errors in advertising research. Current Issues and Research in Marketing 11(1): 21–40. Rossiter JR, Thornton J. 2004. Fear pattern analysis supports the fear-drive model for antispeeding road-safety TV ads. Psychology and Marketing 21(11): 945–960. Schwarz N. 1990. Feelings as information: Implications for affective influences on information processing. In Martin LL, Clore GL (eds). Theories of Mood and Cognition. Lawrence Erlbaum Associates: Mahwah, NJ; 159–173. Sheeran P, Orbell S. 2000. Using implementation intentions to increase attendance for cervical cancer screening. Health Psychology 19(3): 283–289. Speck PS. 1991. The humorous message taxonomy: a framework for the study of humorous ads. In Leigh J, Martin C (eds). Current Issues and Research in Advertising. University of Michigan: Ann Harbor; 1–44. Steenkamp JE, Baumgartner MH, Wulp EVD. 1996. The relationships among arousal potential, arousal and stimulus J. Consumer Behav. 11: 147–161 (2012) DOI: 10.1002/cb Fear and humor evaluation, and the moderating role of need for stimulation. International Journal of Research in Marketing 13: 319–329. Sternthal B, Craig SC. 1974. Fear appeals: Revisited and revised. Journal of Consumer Research 1(3): 22–34. Strong JT, Dubas KM. 1993. The optimal level of fear arousal in advertising: An empirical study. Journal of Current Issues and Research in Advertising 15(2): 93–99. Tanner JFJ, Hunt JB, Eppright DR. 1991. The protection motivation model: A normative model of fear appeals. Journal of Marketing 55(3): 36–45. Time. 2009. Teen tanning. Time 174(14): 53–55. Ventis WL, Higbee G, Murdock SA. 2001. Using humor in systematic desensitization to reduce fear. The Journal of General Psychology 128(2): 241–253. Voss KE. 2009. Using humor in conjunction with negative consequences in advertising. Journal of Current Issues and Research in Advertising 31(2): 25–39. Copyright © 2012 John Wiley & Sons, Ltd. 161 Weinberger MG, Gulas CS. 1992. The impact of humor in advertising: A review. Journal of Advertising Research 31(December/January): 44–52. Weinberger MG, Spotts H, Campbell L, Parsons AL. 1995. The use and effect of humor in different advertising media. Journal of Advertising Research 35(3): 44–56. Witte K, Allen M. 2000. A meta analysis of fear appeals: Implications for effective public health campaigns. Health Education & Behavior 27(5): 608–632. Woltman-Elpers JLCM, Mukherjee A, Hoyer WD. 2004. Humor in television advertising: A moment-to-moment analysis. Journal of Consumer Research 31(December): 592–598. Wyer RS, Collins JE. 1992. A Theory of Humor Elicitation. Psychological Review 99: 663–688 Zhang Y, Zinkhan GM. 1991. Humor in television advertising. In Holman RH, Solomon MR (eds). Advances in Consumer Research 18. Association for Consumer Research: Provo, UT; 813–818. J. Consumer Behav. 11: 147–161 (2012) DOI: 10.1002/cb