- Email: [email protected]
Smoker's recall of fear appeal imagery: Examining the effect of fear intensity and fear type
ARTICLE IN PRESS Australasian Marketing Journal ■■ (2014) ■■–■■
Contents lists available at ScienceDirect
Australasian Marketing Journal j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / a m j
Smoker’s recall of fear appeal imagery: Examining the effect of fear intensity and fear type Ethan Rayner a, Stacey M. Baxter a,*, Jasmina Ilicic b a b
Newcastle Business School, University of Newcastle, Callaghan, NSW 2308, Australia Business School, University of Adelaide, North Terrace Campus, 10 Pulteney Street, SA 5000, Australia
A R T I C L E
I N F O
Article history: Received 14 February 2014 Revised 4 November 2014 Accepted 9 November 2014 Available online Keywords: Fear Intensity Physical harm Social harm Recall
A B S T R A C T
This study examines the effect of fear intensity and type of fear on smokers’ recall of fear-based graphic stimuli. A 2 × 2 factorial design manipulates fear intensity (high vs. low) and fear type (physical vs. social). Results show high intensity messages promote superior recall, with recall heightening when the message also depicts physical harm. This study also shows that viewing time moderates the interaction effect of fear intensity and fear type on recall. Findings demonstrate that smokers report greater recall when exposed to highly intense messages that depict physical harm, with the effect only significant when viewing time is low. We argue that high intensity physical harm messages encourage greater message recall when viewed for only a short period due to shock arousal. © 2014 Australian and New Zealand Marketing Academy. Published by Elsevier Ltd. All rights reserved.
1. Introduction Over the past forty years, the Australian government has promoted anti-smoking messages within the community. In 1972, anti-smoking messages came in the form of production and distribution of signs requesting smokers to not smoke within the nearby area. Later, in the 1980s and 1990s when the true effects of smoking became known, intervention campaigns were introduced (Tobacco in Australia, 2013). Ever since, Australian government bodies have aimed to raise consumers’ awareness of the harms of tobacco use, motivate quitting and discourage the adoption of smoking (Miller et al., 2011). Although some research suggests initiatives, such as plain packaging, reduce the attractiveness and promotional appeal of packaging (Moddie et al., 2011), other researchers find that the removal of point of sale displays that aim to reduce tobacco use will not eliminate all the retail triggers that cue smoking and tobacco purchases (Burton et al., 2013). Tobacco in Australia’s (2013) review of Australian-based messages designed to inhibit the adoption of smoking and/or promote the cessation of smoking behaviour indicates that two main approaches have been incorporated by Australian state and federal governments. The first approach is the presentation of simulated health effects associated with smoking (i.e., symbolic/abstract imagery). For example, the ‘Bubblewrap’ (2005) campaign burned a hole in bubblewrap with a cigarette as a way to depict human lungs and thus the effect that smoking has on the respiratory system and the onset on emphysema. The second approach typically used is the presentation of graphic imagery that depicts either
* Corresponding author. Tel.: +61 2 49216279; fax: +61 2 49216911. E-mail address: [email protected] (S. Baxter).
(1) the current harm to the body that smoking causes (i.e. smokers cough, loss of breath); or (2) the harm that smoking can cause in the future (i.e. mouth cancer, emphysema); or lastly, (3) the effects of smoking on others (i.e. the impact smoking has on children). Whilst it is evident that smoking behaviour in Australia is decreasing (Cancer Council NSW, 2013), a large number of Australians still continue to smoke. According to the Cancer Council NSW (2013), Australians with the highest daily smoking rates are men aged between 30 and 49 years (20.2%) and women aged 40 to 49 years (18.8%). With over 15,000 smoking-related deaths each year in Australia, as well as rising social costs associated with smoking (estimated $31.5 billion in 2004–2005; Cancer Council NSW, 2013), research designed to understand how to best promote the government’s quit smoking message is important for both individuals, the economy, and society at large. Combining approaches used in Australian campaigns, this research examines the effect of the type of fear and the intensity of harm depicted in smoking messages on smokers’ recall of simulated graphic health messages. To date, research has examined the effect of fear type and/or intensity on behavioural outcomes (e.g., intention to cease behaviour, Janis and Feshbach, 1953; Powell, 1965; Brooker, 1981; Schoenbachler and Whittler, 1996; Reardon and Miller, 2008). However, no research has examined the effect of fear type and intensity on consumer message recall. With recall of advertising messages remaining a key measure of advertising effectiveness, which is shown to be a critical driver of consumer behaviour (Mehta and Purvis, 2006), the results of this study have significant practical implications regarding advertising effectiveness. Our aim here is to investigate consumer recall of graphic fearbased messages. We begin with a brief overview of fear appeal
http://dx.doi.org/10.1016/j.ausmj.2014.11.003 1441-3582/© 2014 Australian and New Zealand Marketing Academy. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Ethan Rayner, Stacey M. Baxter, Jasmina Ilicic, Smoker’s recall of fear appeal imagery: Examining the effect of fear intensity and fear type, Australasian Marketing Journal (2014), doi: 10.1016/j.ausmj.2014.11.003
ARTICLE IN PRESS E. Rayner et al./Australasian Marketing Journal ■■ (2014) ■■–■■
2
research in advertising. We then turn to specific research on fear intensity and fear type. From this theoretical foundation, we conduct an experiment examining the role of fear type and fear intensity on consumer recall. We conclude with a discussion of the implications of our findings and offer directions for future research. 2. Fear appeals Fear is “an unpleasant emotional state characterised by anticipation of pain or great distress…that evolved as a mechanism to protect humans from life threatening situations” (Merriam-Webster, 2002, n.p.). Fear, as an emotional response based on high arousal, has been found to have a significant effect on message recall (Snipes et al., 1999) and message persuasion (Peters et al., 2014), influencing a viewer’s affective and behavioural responses (Higbee, 1969; Hyman and Tansey, 1990; King and Reid, 1990; LaTour and Rotfield, 1997). Messages employing fear remain relevant and effective within intervention-targeting campaigns as “fear appeals are one of the most frequently used motivators to get people to help themselves” (Bagozzi and Moore, 1994, p. 56). Interestingly, individuals exposed to various types of appeals have rated negatively valanced fear appeals to be more effective than both rational appeals and other emotional appeals, such as humour (Kim and Lee, 2012; Santa and Cochran, 2008; Siegel, 2002). For example, Santa and Cochran (2008) demonstrated that a fear appeal led to greater negative affect towards driving under the influence (DUI) than an information-based appeal. Further, Kim and Lee (2012) identified that exposure to fear-eliciting stimuli enhanced individuals’ attitudes towards a health issue. A review of the literature reveals two dominant categorisations of fear appeals: (1) intensity of the fear depicted (low and high) and (2) type of fear depicted (physical harm and social fear). 2.1. Fear intensity A large number of studies have focused on the effect of fear intensity on message effectiveness (namely in terms of affective and behavioural responses). Mixed results are found when examining the effect of message intensity, with some researchers arguing that high intensity messages result in more positive attitudes and behaviour (LaTour and Pitts 1989; Snipes, LaTour and Bliss 1996, 1999) whilst others suggest that low intensity messages are more persuasive and effective (Duke et al., 1993; LaTour and Zahra, 1989). When considering message recall, research has demonstrated that high arousal advertisements are most effective (Biener et al., 2004; Lang et al., 1995). Drawing from the notion of ‘shock’, which is synonymous with highly intense emotional messages (e.g., sexual, violent; Hyllegard et al., 2009), researchers have demonstrated that a message (or story) which has strong emotional arousal (or impact) leads to an increase in memory and thus recall (Biener et al., 2004; Lang et al., 1995, 1996; Sutherland and Mather, 2012). We, therefore, expect a high intensity fear message to promote greater arousal, facilitating superior message recall. Formally, we hypothesise: Hypothesis 1. Exposure to imagery depicting a high level of harm will result in higher message recall than imagery depicting a lower level of harm. 2.2. Fear type Research also examines fear appeals in the context of the type of fear depicted in the advertisement. Typically, researchers distinguish between physical fear (harm to oneself) and social fear (harm inflicted by one’s actions on another individual) (e.g. Schoenbachler and Whittler, 1996; Laroche et al., 2001; Brennan and Binney, 2010). Powell (1965) presented the first known study focusing on the impact of fear type. Powell (1965) observed that fear appeals which depicted harm to the listener’s family (social harm) were more effective than fear
appeals that depicted harm to the listener themselves. Since this original study, results regarding the effectiveness of various types of fear appeals are inconsistent in the literature. For example, whilst Schoenbachler and Whittler (1996) demonstrated that messages depicting physical harm were less persuasive than those depicting social harm, Reardon and Miller (2008) found that messages depicting physical harm were more effective than those depicting social harm, in terms of attitude towards the advertisement and behavioural intentions. As such, we posit that the relative effectiveness of fear type is dependent on the outcome investigated. In the context of recall, research in evolutionary psychology has demonstrated that cognitive processes are shaped by adaption (Tooby and Cosmides, 2005). Founded on the notion that an individual’s ultimate goal is to stay alive, psychologists argue that individuals have evolved to attend to situations where their survival is made salient (e.g., Challis et al., 1996), with information retention and message compliance found to be superior in situations of self-relevance (Maille and Hoffmann, 2013; Nairne et al., 2007). Drawing from evolutionary-based explanations of memory performance, we therefore argue that a message depicting harm to one’s self will enhance survival salience, promoting message recall. As a result, we hypothesise: Hypothesis 2. Exposure to imagery depicting harm to the viewer (physical harm) will result in higher message recall than imagery depicting harm to others (social harm).
2.3. Interaction of fear intensity and fear type Despite numerous studies examining the effects of both fear intensity and fear type (e.g., Brooker, 1981; King and Reid, 1990; Reardon and Miller, 2008), only one study, Powell (1965), has demonstrated an interaction between these fear variables. Specifically, Powell’s (1965) study of anxiety arousal identified that fear appeals which depicted social harm (harm to family) were more effective when presented utilising a high intensity appeal, than low intensity appeal. Drawing from prior research focusing on message intensity (e.g., LaTour and Pitts 1989; LaTour, Snipes and Bliss 1996, 1999) and evolutionary-based memory performance, however, we posit the occurrence of an alternate interaction. Specifically, we suggest that a message that enhances survival salience (harm to self) will promote superior recall, with these effects heightened when the survival threat is of a high intensity. We therefore hypothesise: Hypothesis 3. Exposure to imagery depicting a high level of physical harm will result in higher message recall than when imagery depicts a high level of social harm. Exposure to imagery depicting a low level of harm will result in equal levels of message recall, whether the imagery depicts physical or social harm. Recent research suggests that high severity health-based messages foster selective exposure across cultures (Hastall and KnoblochWesterwick, 2013). Specifically, negatively valanced fear-based messages are often used to initially ‘shock’ and arouse a viewer, with the shock effect shown to diminish after extended exposure to the stimuli (Moore and Harris, 1996). As a result, we expect the interaction between fear intensity and fear type to be dependent upon to the time spent viewing the fear-based message. Specifically, we argue that the interaction will have a significant effect on message recall when message viewing time is low (shock/arousal maintained); however, the effect of the interaction will not affect message recall when viewing time is longer, with shock subsiding. As such, the following is proposed: Hypothesis 4. Viewing time moderates the interaction between fear type and fear intensity on message recall (three-way interaction), with message recall greater when viewing time is low.
Please cite this article in press as: Ethan Rayner, Stacey M. Baxter, Jasmina Ilicic, Smoker’s recall of fear appeal imagery: Examining the effect of fear intensity and fear type, Australasian Marketing Journal (2014), doi: 10.1016/j.ausmj.2014.11.003
ARTICLE IN PRESS E. Rayner et al./Australasian Marketing Journal ■■ (2014) ■■–■■
Table 1 Experimental stimuli selection – summary of results.
Condition Low intensity Physical harm Social harm High intensity Physical harm Social harm
Type of fear categorisation (%)
Fear intensity rating
Physical
Social
Other
M (SD)
t
p
70.8 8.3
4.2 75.0
25.0 16.7
2.10 (2.60) 2.60 (1.80)
−3.860 −1.931
.002 .047
100 12.5
0 87.5
0 4.2
4.80 (1.90) 5.60 (1.80)
2.708 4.609
.017 .000
3
at least once a week (9.6%) or less than once a week (10%). Over half of the sample (50.6%) was male. The average age of participants was 45.95 (S.D. = 16.99) years. On average, participants had been smoking for 27.90 years (S.D. = 17.62), with the minimum age they began smoking being 10 years and maximum age they began smoking being 39 years of age (M = 18.05 years of age). 3.4. Manipulation checks
3. Method 3.1. Pre-test: stimuli selection With the aim of selecting four simulated-graphic images for an experiment (high/low fear intensity, physical/social harm), a pretest was undertaken with 24 current smokers. Pre-test participants were exposed to 20 images sourced from the internet. Images were initially selected to represent either physical or social harm at varying degrees of intensity. Pre-test participants were first asked to rate the intensity of the fear depicted in the image on a seven-point semantic differential scale (low–high). Participants were then asked to indicate the nature of harm/s depicted in the image (physical, social and/or other) and whether or not they had seen the image previously. Finally, participants reported their smoking status along with their age and gender. A total of 83.3% of participants indicated that they smoked every day, with the remaining participants indicating that they smoke at least once a week (8.3%), or less than once a week (8.3%). Over half of the sample (58.3%) was male, and the average age of participants was 43.20 years. Images were selected for inclusion in the experiment if the following three criterion were met: (1) the majority of participants reported the image to reflect a particular type of harm; (2) the average intensity rating for the image was either significantly above (high intensity) or significantly below (low intensity) the neutral point as determined by a one sample t-test; and (3) no participants reported having seen the image previously. Results for the four images selected for the study are detailed in Table 1. 3.2. Experimental procedure A total of 239 current smokers from the Australian general public participated in an online experiment. A 2 (intensity of fear; high vs. low) × 2 (type of fear; physical harm vs. social harm) between subjects factorial design was employed, with participants randomly allocated to one of the four experimental conditions. First, participants were shown one of the four experimental images (refer to Appendix A). Participants controlled the time that they viewed the image with viewing time recorded. After a five-minute distractor task, participants were asked to recall as many features of the smoking image as possible (measuring message recall). Participants were then shown the image again and asked to rate “the severity of the harm pictured in the image” using a 7-point semantic differential scale (low–high), identify the “nature of the harm depicted in the image” by selecting one of the following response options; “physical harm to self”, “harm to other individuals” and “other”, and indicate whether they had seen the image previously. Finally, simple demographic information was also collected. The online survey took no longer than 10 minutes to complete (including distractor task). 3.3. Participant profile A total of 192 (80.3%) of participants indicated that they smoked every day, with the remaining participants indicating that they smoke
Data regarding participant’s perception of the type of fear depicted in the image were analysed. One hundred and one participants who viewed imagery chosen to depict physical harm reported as intended, indicating that the image illustrated harm to self. Of those participants who viewed an image intended to depict social harm, 72.3% (n = 86) reported that the image depicted harm to other individuals. To ensure results were not impacted by manipulation errors a decision was made to remove all participants who did not perceive the image the way intended by the manipulation. As a result 54 participants were removed from the experiment (final sample, n = 185). Finally, an independent samples t-test was undertaken to evaluate the clarity of the fear intensity manipulation within the experiment. As intended, participants across all fear types rated the high intensity images as significantly more intense than the low intensity images (t = −7.73, p < .000, Mlow intensity = 4.24, SD = 1.88, Mhigh intensity = 6.01, SD = 1.17). 4. Results and discussion To examine the effect of fear type and fear intensity on recall of an anti-smoking message a 2 (type: physical versus social) × 2 (intensity: high versus moderate) ANOVA was conducted with the number of message items recalled as the dependent variable. A significant main effect is observed with respect to fear intensity (F (3, 181) = 24.03, p < .001), but not fear type (F (3, 181) = 1.71, p = .193), providing no support for Hypothesis 2. Results demonstrate that participants are able to accurately recall more aspects of an image depicting a high intensity of harm (Mrecall = 3.41, SD = 1.93) when compared to an image depicting a low intensity of harm (Mrecall = 2.30, SD = 1.21, t = 5.06, p < .001) providing support for Hypothesis 1. As hypothesised, a significant interaction effect is also observed between fear type and fear intensity (F (3, 181) = 5.05, p = .026; summarised in Table 2 and graphically represented in Fig. 1). Results demonstrate that whilst fear intensity has a significant affect on message recall when social fear is depicted (Mlow intensity = 2.41, Mhigh intensity = 3.02, t = −2.08, p = .041), the effect of fear intensity is heightened when the image displays physical harm (Mlow intensity = 2.19, Mhigh intensity = 3.85, t = −4.77, p < .001), providing support for Hypothesis 3. We also sought to test our hypothesis that viewing time moderates the interaction effect between fear intensity and fear type on recall (three-way interaction). As fear-based marketing messages are developed to elicit ‘shock’, we argue that the effect of the interaction (fear type × fear intensity) on message recall is strengthened when viewing time is low. We suggest that once the image has been viewed for a longer time, recall of a fear-based message will be achieved, irrespective of fear intensity and fear type. To test our hypothesis, the Preacher et al. (2007) PROCESS macro bootstrapping procedure (n = 5000, Model 3) was employed. Results show a significant three-way interaction between fear type, fear intensity, and viewing time (β = −.17, p = .036, R2Δ = .021). As
Table 2 Fear intensity × fear type – message recall. Condition
M
SD
Low intensity/physical harm Low intensity/social harm High intensity/physical harm High intensity/social harm
2.19 2.41 3.85 3.02
1.28 1.13 2.04 1.58
Please cite this article in press as: Ethan Rayner, Stacey M. Baxter, Jasmina Ilicic, Smoker’s recall of fear appeal imagery: Examining the effect of fear intensity and fear type, Australasian Marketing Journal (2014), doi: 10.1016/j.ausmj.2014.11.003
ARTICLE IN PRESS E. Rayner et al./Australasian Marketing Journal ■■ (2014) ■■–■■
4
Panel A: Point 1 (View time = 7.26sec)
Panel B: Point 2 (View time = 14.15sec)
Fig. 1. Fear intensity × fear type – message recall.
shown in Table 3, results demonstrate that the effect of the interaction between fear type and message intensity on message recall is significant only when there are low levels of viewing time, providing support for Hypothesis 4. Specifically, at low levels of viewing time, message recall is enhanced when participants are shown an image depicting a high intensity of harm, with recall strengthening when the image also displays physical harm (refer to Fig. 2, panel A). 5. Conclusion and future research The overall implementation and effectiveness of fear appeals has been a subject of great debate within the advertising and consumer behaviour literature for the best part of 55 years. Results of this study demonstrate that high intensity messages promote superior message recall. We suggest that the shocking nature of a high intensity message is rapidly implanted in the minds of individuals. This assertion is supported by a response provided by a participant who was exposed to a high intensity physical fear image, “The background, the way the man looked, how he was sitting and holding his cigarette, his “leg” and the ash from it on piling up on the ground. I can picture it vividly.” Unlike fear intensity literature, previous findings regarding the effectiveness of various type of fear have been much more inconsistent. For example, both Powell (1965) and Schoenbachler and Whittler (1996) identify fear appeals which depict harm to the listener’s family (social harm) are more effective than fear appeals that depict harm to the listener themselves (physical harm), whereas, Reardon and Miller (2008) found that messages depicting physical harm were more effective than those depicting social harm when considering attitude towards the ad and behavioural intentions. Findings from this study show that fear type does influence message recall. Consistent with evolutionary-based explanations of memory
Panel C: Point 3 (View time = 21.03sec)
Table 3 Conditional effects of interaction (Fear intensity × fear type) across viewing time. Moderator: Viewing time
Effect (95% CI)
t
p
Point 1 (View time = 7.26 s) Point 2 (View time = 14.15 s) Point 3 (View time = 21.03 s)
1.82 (.54, 3.09) .67 (−.26, 1.59) −.48 (−1.94, .98)
2.82 1.43 −.65
.005 .155 .520
Fig. 2. Conditional effects of interaction (Fear intensity × fear type) across viewing time.
Please cite this article in press as: Ethan Rayner, Stacey M. Baxter, Jasmina Ilicic, Smoker’s recall of fear appeal imagery: Examining the effect of fear intensity and fear type, Australasian Marketing Journal (2014), doi: 10.1016/j.ausmj.2014.11.003
ARTICLE IN PRESS E. Rayner et al./Australasian Marketing Journal ■■ (2014) ■■–■■
performance (e.g., Nairne et al., 2007), results of this study show that images depicting physical harm result in greater recall than images depicting social harm. Furthermore, viewing time was found to play a moderating role, whereby message recall was enhanced when consumers were shown an image depicting a high intensity physical harm; however, the interaction between fear intensity and type only occurs in situations of low viewing time. We suggest this effect is related to the nature of message, with negatively valanced messages more often used to initially ‘shock’ and arouse a viewer, with the shock effect shown to reduce after extended exposure to the stimuli (Moore and Harris, 1996). Since many Australian smoking intervention campaigns depict images of the physical harm associated with smoking (often at a high level of intensity), this study confirms that these images (in particular, when presented as a fleeting message) are effective in influencing recall, suggesting that this stimuli is more likely stored in consumer memory. Advertisers aim to enhance message recall as it is found to influence advertising effectiveness and consumer behaviour (Mehta and Purvis, 2006). However future research, which takes into consideration message viewing time, is needed to confirm this effect on attitudinal and behavioural outcomes. This study presents one of only a few studies that has utilised an adult sample of the general public. The majority of studies in the fear appeal literature have either adopted a student or youth sample for their research. This study has attempted to expand outside this confined demographic to focus on adult smokers over the age of 18. Students have been exposed to negative messages concerning
5
smoking for the complete duration of their lifetime to date, thus students commonly have disparaging attitudes towards the behaviour (Chambliss et al., 2006). Other research finds that adolescents exert high scepticism of social based advertising with reduced perceptions of the risks (Thakor and Goneau-Lessard, 2009). This study has broader and specific applicability by focusing on adult smokers within the general population, which provided useful insights into how anti-smoking fear appeal images affect this demographic’s ability to recall messages. Future research, however, should examine smokers’ versus non-smokers’ ability to recall information presented in smoking specific fear appeals to identify effective quit campaigns that specifically target smokers (through encouraging them to cease their behaviour) and also non-smokers (by encouraging them not to commence the behaviour). This study is limited in that all the images used within the experiment were artistic and unrealistic in nature, for example they did not depict the literal impacts of smoking but rather the symbolic impacts (i.e. smoking will not literally result in your leg burning and turning to ash, but it can result in loss of limbs and possible death). Although, the images utilised in this study may be considered unrealistic, there are numerous examples of these types of images in quit smoking campaigns conducted both here in Australia and around the world, providing significant ramifications for practice i.e. the development of effective fear-based stimuli. Future research, however, is needed to examine the effectiveness of both literal and symbolic stimuli on various outcomes including message recall and behaviour.
Appendix A: *
High Intensity/Physical Harm
Low Intensity/Physical Harm
High Intensity/Social Harm
Low Intensity/Social Harm
*Color on web only.
Please cite this article in press as: Ethan Rayner, Stacey M. Baxter, Jasmina Ilicic, Smoker’s recall of fear appeal imagery: Examining the effect of fear intensity and fear type, Australasian Marketing Journal (2014), doi: 10.1016/j.ausmj.2014.11.003
ARTICLE IN PRESS E. Rayner et al./Australasian Marketing Journal ■■ (2014) ■■–■■
6
References Bagozzi, R.P., Moore, D.J., 1994. Public service advertisements: emotions and empathy guide prosocial behavior. J. Mark. 58 (1), 56–70. Biener, L., Ji, M., Gilpin, E.A., Albers, A.B., 2004. The impact of emotional tone, message, and broadcast parameters in youth anti-smoking advertisements. J. Health Commun. 9, 259–274. Brennan, L., Binney, W., 2010. Fear, guilt and shame appeals in social marketing. J. Bus. Res. 63 (2), 140–146. Brooker, G., 1981. A comparison of the persuasive effects of mild humour and mild fear appeals. J. Advert. 10 (4), 29–40. Burton, S., Spanjaard, D., Hoek, J., 2013. An investigation of tobacco retail outlets as a cue for smoking. Australas. Mark. J. 21, 234–239. Cancer Council NSW, 2013. Cancer Council NSW: statistics on smoking in Australia. Retrieved from Cancer Council NSW Web Site.
LaTour, M.S., Zahra, S.A., 1989. Fear appeals as advertising strategy: should they be used? J. Consum. Mark. 6 (2), 61–70. Maille, V., Hoffmann, J., 2013. Compliance with veterinary prescriptions: the role of physical and social risk revisited. J. Bus. Res. 66 (1), 141–144. Mehta, A., Purvis, S.C., 2006. Reconsidering recall and emotion in advertising. J. Advert. Res. (March), 49–56. Merriam-Webster, 2002. Merriam-Webster’s Concise Dictionary of English Usage: The Essentials of Clear Expression. Merriam-Webster, Springfield, MA. Miller, C.L., Hill, D.J., Quester, P.G., Hiller, J.E., 2011. The impact of Australia’s new graphic cigarette packet warnings on smokers’ beliefs and attitudes. Australas. Mark. J. 19 (3), 181–188. Moddie, C., Mackintosh, A.M., Hastings, G., Ford, A., 2011. Young adult smokers’ perceptions of plain packaging: A pilot naturalistic study. Tob. Control 20 (5), 367–373. Moore, D.J., Harris, W.D., 1996. Affect intensity and the consumer’s attitude toward high impact emotional advertising appeals. J. Advert. 25 (2), 37–50. Nairne, J.S., Thompson, S.R., Pandeirada, J.N.S., 2007. Adaptive memory: survival processing enhances retention. J. Exp. Psychol. Learn. Mem. Cogn. 33, 263–273. Peters, G.J.Y., Ruiter, R.A.C., Kok, G., 2014. Threatening communication: a qualitative study of fear appeal effectiveness beliefs among intervention developers, policymakers, politicians, scientists, and advertising professionals. Int. J. Psychol. 49, 71–79. Powell, F.A., 1965. The effects of anxiety-arousing messages when related to personal, familiar, and interpersonal referents. Speech Monogr. 32, 95–101. Preacher, K.J., Rucker, D.D., Hayes, A.F., 2007. Assessing moderated mediation hypotheses: theory, methods and prescriptions. Multivariate Behav. Res. 42 (1), 185–227. Reardon, J., Miller, C., 2008. Smoking prevention messages for adolescents: how intensity, valence, and recipient of consequences affect attitude toward the ad and intent to smoke. J. Mark. Theory Pract. 16 (1), 67–77. Santa, A.F., Cochran, B.N., 2008. Does the impact of anti-drinking and driving public service announcements differ based on message type and viewer characteristics? J. Drug Educ. 38 (2), 109–129. Schoenbachler, D.D., Whittler, T.E., 1996. Adolescent processing of social and physical threat communications. J. Advert. 25 (4), 37–54. Siegel, M., 2002. Antismoking advertising: figuring out what works. J. Health Commun. 7, 157–162. Snipes, R.L., LaTour, M.S., Bliss, S.J., 1996. Don’t be afraid to use fear appeals: An experimental study. J. Advert. Res. 36 (2), 59–67. Snipes, R.L., LaTour, M.S., Bliss, S.J., 1999. A model of the effects of self-efficacy on the perceived ethicality and performance of fear appeals in advertising. J. Bus. Ethics 19 (3), 273–285. Sutherland, M.R., Mather, M., 2012. Negative arousal amplifies the effects of saliency in short-term memory. Emotion 12 (6), 1367–1372. Thakor, M.V., Goneau-Lessard, K., 2009. Development of a scale to measure skepticism of social advertising among adolescents. J. Bus. Res. 62, 1342–1349. Tobacco in Australia, 2013. Tobacco-control campaigns in Australia: experience. Retrieved from Tobacco in Australia Corporate Web site.. Tooby, J., Cosmides, L., 2005. Conceptual foundations of evolutionary psychology. In: Buss, D.M. (Ed.), The Handbook of Evolutionary Psychology. Wiley, Hoboken, NJ, pp. 5–67.
Please cite this article in press as: Ethan Rayner, Stacey M. Baxter, Jasmina Ilicic, Smoker’s recall of fear appeal imagery: Examining the effect of fear intensity and fear type, Australasian Marketing Journal (2014), doi: 10.1016/j.ausmj.2014.11.003
Contents lists available at ScienceDirect
Australasian Marketing Journal j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / a m j
Smoker’s recall of fear appeal imagery: Examining the effect of fear intensity and fear type Ethan Rayner a, Stacey M. Baxter a,*, Jasmina Ilicic b a b
Newcastle Business School, University of Newcastle, Callaghan, NSW 2308, Australia Business School, University of Adelaide, North Terrace Campus, 10 Pulteney Street, SA 5000, Australia
A R T I C L E
I N F O
Article history: Received 14 February 2014 Revised 4 November 2014 Accepted 9 November 2014 Available online Keywords: Fear Intensity Physical harm Social harm Recall
A B S T R A C T
This study examines the effect of fear intensity and type of fear on smokers’ recall of fear-based graphic stimuli. A 2 × 2 factorial design manipulates fear intensity (high vs. low) and fear type (physical vs. social). Results show high intensity messages promote superior recall, with recall heightening when the message also depicts physical harm. This study also shows that viewing time moderates the interaction effect of fear intensity and fear type on recall. Findings demonstrate that smokers report greater recall when exposed to highly intense messages that depict physical harm, with the effect only significant when viewing time is low. We argue that high intensity physical harm messages encourage greater message recall when viewed for only a short period due to shock arousal. © 2014 Australian and New Zealand Marketing Academy. Published by Elsevier Ltd. All rights reserved.
1. Introduction Over the past forty years, the Australian government has promoted anti-smoking messages within the community. In 1972, anti-smoking messages came in the form of production and distribution of signs requesting smokers to not smoke within the nearby area. Later, in the 1980s and 1990s when the true effects of smoking became known, intervention campaigns were introduced (Tobacco in Australia, 2013). Ever since, Australian government bodies have aimed to raise consumers’ awareness of the harms of tobacco use, motivate quitting and discourage the adoption of smoking (Miller et al., 2011). Although some research suggests initiatives, such as plain packaging, reduce the attractiveness and promotional appeal of packaging (Moddie et al., 2011), other researchers find that the removal of point of sale displays that aim to reduce tobacco use will not eliminate all the retail triggers that cue smoking and tobacco purchases (Burton et al., 2013). Tobacco in Australia’s (2013) review of Australian-based messages designed to inhibit the adoption of smoking and/or promote the cessation of smoking behaviour indicates that two main approaches have been incorporated by Australian state and federal governments. The first approach is the presentation of simulated health effects associated with smoking (i.e., symbolic/abstract imagery). For example, the ‘Bubblewrap’ (2005) campaign burned a hole in bubblewrap with a cigarette as a way to depict human lungs and thus the effect that smoking has on the respiratory system and the onset on emphysema. The second approach typically used is the presentation of graphic imagery that depicts either
* Corresponding author. Tel.: +61 2 49216279; fax: +61 2 49216911. E-mail address: [email protected] (S. Baxter).
(1) the current harm to the body that smoking causes (i.e. smokers cough, loss of breath); or (2) the harm that smoking can cause in the future (i.e. mouth cancer, emphysema); or lastly, (3) the effects of smoking on others (i.e. the impact smoking has on children). Whilst it is evident that smoking behaviour in Australia is decreasing (Cancer Council NSW, 2013), a large number of Australians still continue to smoke. According to the Cancer Council NSW (2013), Australians with the highest daily smoking rates are men aged between 30 and 49 years (20.2%) and women aged 40 to 49 years (18.8%). With over 15,000 smoking-related deaths each year in Australia, as well as rising social costs associated with smoking (estimated $31.5 billion in 2004–2005; Cancer Council NSW, 2013), research designed to understand how to best promote the government’s quit smoking message is important for both individuals, the economy, and society at large. Combining approaches used in Australian campaigns, this research examines the effect of the type of fear and the intensity of harm depicted in smoking messages on smokers’ recall of simulated graphic health messages. To date, research has examined the effect of fear type and/or intensity on behavioural outcomes (e.g., intention to cease behaviour, Janis and Feshbach, 1953; Powell, 1965; Brooker, 1981; Schoenbachler and Whittler, 1996; Reardon and Miller, 2008). However, no research has examined the effect of fear type and intensity on consumer message recall. With recall of advertising messages remaining a key measure of advertising effectiveness, which is shown to be a critical driver of consumer behaviour (Mehta and Purvis, 2006), the results of this study have significant practical implications regarding advertising effectiveness. Our aim here is to investigate consumer recall of graphic fearbased messages. We begin with a brief overview of fear appeal
http://dx.doi.org/10.1016/j.ausmj.2014.11.003 1441-3582/© 2014 Australian and New Zealand Marketing Academy. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Ethan Rayner, Stacey M. Baxter, Jasmina Ilicic, Smoker’s recall of fear appeal imagery: Examining the effect of fear intensity and fear type, Australasian Marketing Journal (2014), doi: 10.1016/j.ausmj.2014.11.003
ARTICLE IN PRESS E. Rayner et al./Australasian Marketing Journal ■■ (2014) ■■–■■
2
research in advertising. We then turn to specific research on fear intensity and fear type. From this theoretical foundation, we conduct an experiment examining the role of fear type and fear intensity on consumer recall. We conclude with a discussion of the implications of our findings and offer directions for future research. 2. Fear appeals Fear is “an unpleasant emotional state characterised by anticipation of pain or great distress…that evolved as a mechanism to protect humans from life threatening situations” (Merriam-Webster, 2002, n.p.). Fear, as an emotional response based on high arousal, has been found to have a significant effect on message recall (Snipes et al., 1999) and message persuasion (Peters et al., 2014), influencing a viewer’s affective and behavioural responses (Higbee, 1969; Hyman and Tansey, 1990; King and Reid, 1990; LaTour and Rotfield, 1997). Messages employing fear remain relevant and effective within intervention-targeting campaigns as “fear appeals are one of the most frequently used motivators to get people to help themselves” (Bagozzi and Moore, 1994, p. 56). Interestingly, individuals exposed to various types of appeals have rated negatively valanced fear appeals to be more effective than both rational appeals and other emotional appeals, such as humour (Kim and Lee, 2012; Santa and Cochran, 2008; Siegel, 2002). For example, Santa and Cochran (2008) demonstrated that a fear appeal led to greater negative affect towards driving under the influence (DUI) than an information-based appeal. Further, Kim and Lee (2012) identified that exposure to fear-eliciting stimuli enhanced individuals’ attitudes towards a health issue. A review of the literature reveals two dominant categorisations of fear appeals: (1) intensity of the fear depicted (low and high) and (2) type of fear depicted (physical harm and social fear). 2.1. Fear intensity A large number of studies have focused on the effect of fear intensity on message effectiveness (namely in terms of affective and behavioural responses). Mixed results are found when examining the effect of message intensity, with some researchers arguing that high intensity messages result in more positive attitudes and behaviour (LaTour and Pitts 1989; Snipes, LaTour and Bliss 1996, 1999) whilst others suggest that low intensity messages are more persuasive and effective (Duke et al., 1993; LaTour and Zahra, 1989). When considering message recall, research has demonstrated that high arousal advertisements are most effective (Biener et al., 2004; Lang et al., 1995). Drawing from the notion of ‘shock’, which is synonymous with highly intense emotional messages (e.g., sexual, violent; Hyllegard et al., 2009), researchers have demonstrated that a message (or story) which has strong emotional arousal (or impact) leads to an increase in memory and thus recall (Biener et al., 2004; Lang et al., 1995, 1996; Sutherland and Mather, 2012). We, therefore, expect a high intensity fear message to promote greater arousal, facilitating superior message recall. Formally, we hypothesise: Hypothesis 1. Exposure to imagery depicting a high level of harm will result in higher message recall than imagery depicting a lower level of harm. 2.2. Fear type Research also examines fear appeals in the context of the type of fear depicted in the advertisement. Typically, researchers distinguish between physical fear (harm to oneself) and social fear (harm inflicted by one’s actions on another individual) (e.g. Schoenbachler and Whittler, 1996; Laroche et al., 2001; Brennan and Binney, 2010). Powell (1965) presented the first known study focusing on the impact of fear type. Powell (1965) observed that fear appeals which depicted harm to the listener’s family (social harm) were more effective than fear
appeals that depicted harm to the listener themselves. Since this original study, results regarding the effectiveness of various types of fear appeals are inconsistent in the literature. For example, whilst Schoenbachler and Whittler (1996) demonstrated that messages depicting physical harm were less persuasive than those depicting social harm, Reardon and Miller (2008) found that messages depicting physical harm were more effective than those depicting social harm, in terms of attitude towards the advertisement and behavioural intentions. As such, we posit that the relative effectiveness of fear type is dependent on the outcome investigated. In the context of recall, research in evolutionary psychology has demonstrated that cognitive processes are shaped by adaption (Tooby and Cosmides, 2005). Founded on the notion that an individual’s ultimate goal is to stay alive, psychologists argue that individuals have evolved to attend to situations where their survival is made salient (e.g., Challis et al., 1996), with information retention and message compliance found to be superior in situations of self-relevance (Maille and Hoffmann, 2013; Nairne et al., 2007). Drawing from evolutionary-based explanations of memory performance, we therefore argue that a message depicting harm to one’s self will enhance survival salience, promoting message recall. As a result, we hypothesise: Hypothesis 2. Exposure to imagery depicting harm to the viewer (physical harm) will result in higher message recall than imagery depicting harm to others (social harm).
2.3. Interaction of fear intensity and fear type Despite numerous studies examining the effects of both fear intensity and fear type (e.g., Brooker, 1981; King and Reid, 1990; Reardon and Miller, 2008), only one study, Powell (1965), has demonstrated an interaction between these fear variables. Specifically, Powell’s (1965) study of anxiety arousal identified that fear appeals which depicted social harm (harm to family) were more effective when presented utilising a high intensity appeal, than low intensity appeal. Drawing from prior research focusing on message intensity (e.g., LaTour and Pitts 1989; LaTour, Snipes and Bliss 1996, 1999) and evolutionary-based memory performance, however, we posit the occurrence of an alternate interaction. Specifically, we suggest that a message that enhances survival salience (harm to self) will promote superior recall, with these effects heightened when the survival threat is of a high intensity. We therefore hypothesise: Hypothesis 3. Exposure to imagery depicting a high level of physical harm will result in higher message recall than when imagery depicts a high level of social harm. Exposure to imagery depicting a low level of harm will result in equal levels of message recall, whether the imagery depicts physical or social harm. Recent research suggests that high severity health-based messages foster selective exposure across cultures (Hastall and KnoblochWesterwick, 2013). Specifically, negatively valanced fear-based messages are often used to initially ‘shock’ and arouse a viewer, with the shock effect shown to diminish after extended exposure to the stimuli (Moore and Harris, 1996). As a result, we expect the interaction between fear intensity and fear type to be dependent upon to the time spent viewing the fear-based message. Specifically, we argue that the interaction will have a significant effect on message recall when message viewing time is low (shock/arousal maintained); however, the effect of the interaction will not affect message recall when viewing time is longer, with shock subsiding. As such, the following is proposed: Hypothesis 4. Viewing time moderates the interaction between fear type and fear intensity on message recall (three-way interaction), with message recall greater when viewing time is low.
Please cite this article in press as: Ethan Rayner, Stacey M. Baxter, Jasmina Ilicic, Smoker’s recall of fear appeal imagery: Examining the effect of fear intensity and fear type, Australasian Marketing Journal (2014), doi: 10.1016/j.ausmj.2014.11.003
ARTICLE IN PRESS E. Rayner et al./Australasian Marketing Journal ■■ (2014) ■■–■■
Table 1 Experimental stimuli selection – summary of results.
Condition Low intensity Physical harm Social harm High intensity Physical harm Social harm
Type of fear categorisation (%)
Fear intensity rating
Physical
Social
Other
M (SD)
t
p
70.8 8.3
4.2 75.0
25.0 16.7
2.10 (2.60) 2.60 (1.80)
−3.860 −1.931
.002 .047
100 12.5
0 87.5
0 4.2
4.80 (1.90) 5.60 (1.80)
2.708 4.609
.017 .000
3
at least once a week (9.6%) or less than once a week (10%). Over half of the sample (50.6%) was male. The average age of participants was 45.95 (S.D. = 16.99) years. On average, participants had been smoking for 27.90 years (S.D. = 17.62), with the minimum age they began smoking being 10 years and maximum age they began smoking being 39 years of age (M = 18.05 years of age). 3.4. Manipulation checks
3. Method 3.1. Pre-test: stimuli selection With the aim of selecting four simulated-graphic images for an experiment (high/low fear intensity, physical/social harm), a pretest was undertaken with 24 current smokers. Pre-test participants were exposed to 20 images sourced from the internet. Images were initially selected to represent either physical or social harm at varying degrees of intensity. Pre-test participants were first asked to rate the intensity of the fear depicted in the image on a seven-point semantic differential scale (low–high). Participants were then asked to indicate the nature of harm/s depicted in the image (physical, social and/or other) and whether or not they had seen the image previously. Finally, participants reported their smoking status along with their age and gender. A total of 83.3% of participants indicated that they smoked every day, with the remaining participants indicating that they smoke at least once a week (8.3%), or less than once a week (8.3%). Over half of the sample (58.3%) was male, and the average age of participants was 43.20 years. Images were selected for inclusion in the experiment if the following three criterion were met: (1) the majority of participants reported the image to reflect a particular type of harm; (2) the average intensity rating for the image was either significantly above (high intensity) or significantly below (low intensity) the neutral point as determined by a one sample t-test; and (3) no participants reported having seen the image previously. Results for the four images selected for the study are detailed in Table 1. 3.2. Experimental procedure A total of 239 current smokers from the Australian general public participated in an online experiment. A 2 (intensity of fear; high vs. low) × 2 (type of fear; physical harm vs. social harm) between subjects factorial design was employed, with participants randomly allocated to one of the four experimental conditions. First, participants were shown one of the four experimental images (refer to Appendix A). Participants controlled the time that they viewed the image with viewing time recorded. After a five-minute distractor task, participants were asked to recall as many features of the smoking image as possible (measuring message recall). Participants were then shown the image again and asked to rate “the severity of the harm pictured in the image” using a 7-point semantic differential scale (low–high), identify the “nature of the harm depicted in the image” by selecting one of the following response options; “physical harm to self”, “harm to other individuals” and “other”, and indicate whether they had seen the image previously. Finally, simple demographic information was also collected. The online survey took no longer than 10 minutes to complete (including distractor task). 3.3. Participant profile A total of 192 (80.3%) of participants indicated that they smoked every day, with the remaining participants indicating that they smoke
Data regarding participant’s perception of the type of fear depicted in the image were analysed. One hundred and one participants who viewed imagery chosen to depict physical harm reported as intended, indicating that the image illustrated harm to self. Of those participants who viewed an image intended to depict social harm, 72.3% (n = 86) reported that the image depicted harm to other individuals. To ensure results were not impacted by manipulation errors a decision was made to remove all participants who did not perceive the image the way intended by the manipulation. As a result 54 participants were removed from the experiment (final sample, n = 185). Finally, an independent samples t-test was undertaken to evaluate the clarity of the fear intensity manipulation within the experiment. As intended, participants across all fear types rated the high intensity images as significantly more intense than the low intensity images (t = −7.73, p < .000, Mlow intensity = 4.24, SD = 1.88, Mhigh intensity = 6.01, SD = 1.17). 4. Results and discussion To examine the effect of fear type and fear intensity on recall of an anti-smoking message a 2 (type: physical versus social) × 2 (intensity: high versus moderate) ANOVA was conducted with the number of message items recalled as the dependent variable. A significant main effect is observed with respect to fear intensity (F (3, 181) = 24.03, p < .001), but not fear type (F (3, 181) = 1.71, p = .193), providing no support for Hypothesis 2. Results demonstrate that participants are able to accurately recall more aspects of an image depicting a high intensity of harm (Mrecall = 3.41, SD = 1.93) when compared to an image depicting a low intensity of harm (Mrecall = 2.30, SD = 1.21, t = 5.06, p < .001) providing support for Hypothesis 1. As hypothesised, a significant interaction effect is also observed between fear type and fear intensity (F (3, 181) = 5.05, p = .026; summarised in Table 2 and graphically represented in Fig. 1). Results demonstrate that whilst fear intensity has a significant affect on message recall when social fear is depicted (Mlow intensity = 2.41, Mhigh intensity = 3.02, t = −2.08, p = .041), the effect of fear intensity is heightened when the image displays physical harm (Mlow intensity = 2.19, Mhigh intensity = 3.85, t = −4.77, p < .001), providing support for Hypothesis 3. We also sought to test our hypothesis that viewing time moderates the interaction effect between fear intensity and fear type on recall (three-way interaction). As fear-based marketing messages are developed to elicit ‘shock’, we argue that the effect of the interaction (fear type × fear intensity) on message recall is strengthened when viewing time is low. We suggest that once the image has been viewed for a longer time, recall of a fear-based message will be achieved, irrespective of fear intensity and fear type. To test our hypothesis, the Preacher et al. (2007) PROCESS macro bootstrapping procedure (n = 5000, Model 3) was employed. Results show a significant three-way interaction between fear type, fear intensity, and viewing time (β = −.17, p = .036, R2Δ = .021). As
Table 2 Fear intensity × fear type – message recall. Condition
M
SD
Low intensity/physical harm Low intensity/social harm High intensity/physical harm High intensity/social harm
2.19 2.41 3.85 3.02
1.28 1.13 2.04 1.58
Please cite this article in press as: Ethan Rayner, Stacey M. Baxter, Jasmina Ilicic, Smoker’s recall of fear appeal imagery: Examining the effect of fear intensity and fear type, Australasian Marketing Journal (2014), doi: 10.1016/j.ausmj.2014.11.003
ARTICLE IN PRESS E. Rayner et al./Australasian Marketing Journal ■■ (2014) ■■–■■
4
Panel A: Point 1 (View time = 7.26sec)
Panel B: Point 2 (View time = 14.15sec)
Fig. 1. Fear intensity × fear type – message recall.
shown in Table 3, results demonstrate that the effect of the interaction between fear type and message intensity on message recall is significant only when there are low levels of viewing time, providing support for Hypothesis 4. Specifically, at low levels of viewing time, message recall is enhanced when participants are shown an image depicting a high intensity of harm, with recall strengthening when the image also displays physical harm (refer to Fig. 2, panel A). 5. Conclusion and future research The overall implementation and effectiveness of fear appeals has been a subject of great debate within the advertising and consumer behaviour literature for the best part of 55 years. Results of this study demonstrate that high intensity messages promote superior message recall. We suggest that the shocking nature of a high intensity message is rapidly implanted in the minds of individuals. This assertion is supported by a response provided by a participant who was exposed to a high intensity physical fear image, “The background, the way the man looked, how he was sitting and holding his cigarette, his “leg” and the ash from it on piling up on the ground. I can picture it vividly.” Unlike fear intensity literature, previous findings regarding the effectiveness of various type of fear have been much more inconsistent. For example, both Powell (1965) and Schoenbachler and Whittler (1996) identify fear appeals which depict harm to the listener’s family (social harm) are more effective than fear appeals that depict harm to the listener themselves (physical harm), whereas, Reardon and Miller (2008) found that messages depicting physical harm were more effective than those depicting social harm when considering attitude towards the ad and behavioural intentions. Findings from this study show that fear type does influence message recall. Consistent with evolutionary-based explanations of memory
Panel C: Point 3 (View time = 21.03sec)
Table 3 Conditional effects of interaction (Fear intensity × fear type) across viewing time. Moderator: Viewing time
Effect (95% CI)
t
p
Point 1 (View time = 7.26 s) Point 2 (View time = 14.15 s) Point 3 (View time = 21.03 s)
1.82 (.54, 3.09) .67 (−.26, 1.59) −.48 (−1.94, .98)
2.82 1.43 −.65
.005 .155 .520
Fig. 2. Conditional effects of interaction (Fear intensity × fear type) across viewing time.
Please cite this article in press as: Ethan Rayner, Stacey M. Baxter, Jasmina Ilicic, Smoker’s recall of fear appeal imagery: Examining the effect of fear intensity and fear type, Australasian Marketing Journal (2014), doi: 10.1016/j.ausmj.2014.11.003
ARTICLE IN PRESS E. Rayner et al./Australasian Marketing Journal ■■ (2014) ■■–■■
performance (e.g., Nairne et al., 2007), results of this study show that images depicting physical harm result in greater recall than images depicting social harm. Furthermore, viewing time was found to play a moderating role, whereby message recall was enhanced when consumers were shown an image depicting a high intensity physical harm; however, the interaction between fear intensity and type only occurs in situations of low viewing time. We suggest this effect is related to the nature of message, with negatively valanced messages more often used to initially ‘shock’ and arouse a viewer, with the shock effect shown to reduce after extended exposure to the stimuli (Moore and Harris, 1996). Since many Australian smoking intervention campaigns depict images of the physical harm associated with smoking (often at a high level of intensity), this study confirms that these images (in particular, when presented as a fleeting message) are effective in influencing recall, suggesting that this stimuli is more likely stored in consumer memory. Advertisers aim to enhance message recall as it is found to influence advertising effectiveness and consumer behaviour (Mehta and Purvis, 2006). However future research, which takes into consideration message viewing time, is needed to confirm this effect on attitudinal and behavioural outcomes. This study presents one of only a few studies that has utilised an adult sample of the general public. The majority of studies in the fear appeal literature have either adopted a student or youth sample for their research. This study has attempted to expand outside this confined demographic to focus on adult smokers over the age of 18. Students have been exposed to negative messages concerning
5
smoking for the complete duration of their lifetime to date, thus students commonly have disparaging attitudes towards the behaviour (Chambliss et al., 2006). Other research finds that adolescents exert high scepticism of social based advertising with reduced perceptions of the risks (Thakor and Goneau-Lessard, 2009). This study has broader and specific applicability by focusing on adult smokers within the general population, which provided useful insights into how anti-smoking fear appeal images affect this demographic’s ability to recall messages. Future research, however, should examine smokers’ versus non-smokers’ ability to recall information presented in smoking specific fear appeals to identify effective quit campaigns that specifically target smokers (through encouraging them to cease their behaviour) and also non-smokers (by encouraging them not to commence the behaviour). This study is limited in that all the images used within the experiment were artistic and unrealistic in nature, for example they did not depict the literal impacts of smoking but rather the symbolic impacts (i.e. smoking will not literally result in your leg burning and turning to ash, but it can result in loss of limbs and possible death). Although, the images utilised in this study may be considered unrealistic, there are numerous examples of these types of images in quit smoking campaigns conducted both here in Australia and around the world, providing significant ramifications for practice i.e. the development of effective fear-based stimuli. Future research, however, is needed to examine the effectiveness of both literal and symbolic stimuli on various outcomes including message recall and behaviour.
Appendix A: *
High Intensity/Physical Harm
Low Intensity/Physical Harm
High Intensity/Social Harm
Low Intensity/Social Harm
*Color on web only.
Please cite this article in press as: Ethan Rayner, Stacey M. Baxter, Jasmina Ilicic, Smoker’s recall of fear appeal imagery: Examining the effect of fear intensity and fear type, Australasian Marketing Journal (2014), doi: 10.1016/j.ausmj.2014.11.003
ARTICLE IN PRESS E. Rayner et al./Australasian Marketing Journal ■■ (2014) ■■–■■
6
References Bagozzi, R.P., Moore, D.J., 1994. Public service advertisements: emotions and empathy guide prosocial behavior. J. Mark. 58 (1), 56–70. Biener, L., Ji, M., Gilpin, E.A., Albers, A.B., 2004. The impact of emotional tone, message, and broadcast parameters in youth anti-smoking advertisements. J. Health Commun. 9, 259–274. Brennan, L., Binney, W., 2010. Fear, guilt and shame appeals in social marketing. J. Bus. Res. 63 (2), 140–146. Brooker, G., 1981. A comparison of the persuasive effects of mild humour and mild fear appeals. J. Advert. 10 (4), 29–40. Burton, S., Spanjaard, D., Hoek, J., 2013. An investigation of tobacco retail outlets as a cue for smoking. Australas. Mark. J. 21, 234–239. Cancer Council NSW, 2013. Cancer Council NSW: statistics on smoking in Australia. Retrieved from Cancer Council NSW Web Site.
LaTour, M.S., Zahra, S.A., 1989. Fear appeals as advertising strategy: should they be used? J. Consum. Mark. 6 (2), 61–70. Maille, V., Hoffmann, J., 2013. Compliance with veterinary prescriptions: the role of physical and social risk revisited. J. Bus. Res. 66 (1), 141–144. Mehta, A., Purvis, S.C., 2006. Reconsidering recall and emotion in advertising. J. Advert. Res. (March), 49–56. Merriam-Webster, 2002. Merriam-Webster’s Concise Dictionary of English Usage: The Essentials of Clear Expression. Merriam-Webster, Springfield, MA. Miller, C.L., Hill, D.J., Quester, P.G., Hiller, J.E., 2011. The impact of Australia’s new graphic cigarette packet warnings on smokers’ beliefs and attitudes. Australas. Mark. J. 19 (3), 181–188. Moddie, C., Mackintosh, A.M., Hastings, G., Ford, A., 2011. Young adult smokers’ perceptions of plain packaging: A pilot naturalistic study. Tob. Control 20 (5), 367–373. Moore, D.J., Harris, W.D., 1996. Affect intensity and the consumer’s attitude toward high impact emotional advertising appeals. J. Advert. 25 (2), 37–50. Nairne, J.S., Thompson, S.R., Pandeirada, J.N.S., 2007. Adaptive memory: survival processing enhances retention. J. Exp. Psychol. Learn. Mem. Cogn. 33, 263–273. Peters, G.J.Y., Ruiter, R.A.C., Kok, G., 2014. Threatening communication: a qualitative study of fear appeal effectiveness beliefs among intervention developers, policymakers, politicians, scientists, and advertising professionals. Int. J. Psychol. 49, 71–79. Powell, F.A., 1965. The effects of anxiety-arousing messages when related to personal, familiar, and interpersonal referents. Speech Monogr. 32, 95–101. Preacher, K.J., Rucker, D.D., Hayes, A.F., 2007. Assessing moderated mediation hypotheses: theory, methods and prescriptions. Multivariate Behav. Res. 42 (1), 185–227. Reardon, J., Miller, C., 2008. Smoking prevention messages for adolescents: how intensity, valence, and recipient of consequences affect attitude toward the ad and intent to smoke. J. Mark. Theory Pract. 16 (1), 67–77. Santa, A.F., Cochran, B.N., 2008. Does the impact of anti-drinking and driving public service announcements differ based on message type and viewer characteristics? J. Drug Educ. 38 (2), 109–129. Schoenbachler, D.D., Whittler, T.E., 1996. Adolescent processing of social and physical threat communications. J. Advert. 25 (4), 37–54. Siegel, M., 2002. Antismoking advertising: figuring out what works. J. Health Commun. 7, 157–162. Snipes, R.L., LaTour, M.S., Bliss, S.J., 1996. Don’t be afraid to use fear appeals: An experimental study. J. Advert. Res. 36 (2), 59–67. Snipes, R.L., LaTour, M.S., Bliss, S.J., 1999. A model of the effects of self-efficacy on the perceived ethicality and performance of fear appeals in advertising. J. Bus. Ethics 19 (3), 273–285. Sutherland, M.R., Mather, M., 2012. Negative arousal amplifies the effects of saliency in short-term memory. Emotion 12 (6), 1367–1372. Thakor, M.V., Goneau-Lessard, K., 2009. Development of a scale to measure skepticism of social advertising among adolescents. J. Bus. Res. 62, 1342–1349. Tobacco in Australia, 2013. Tobacco-control campaigns in Australia: experience. Retrieved from Tobacco in Australia Corporate Web site.
Please cite this article in press as: Ethan Rayner, Stacey M. Baxter, Jasmina Ilicic, Smoker’s recall of fear appeal imagery: Examining the effect of fear intensity and fear type, Australasian Marketing Journal (2014), doi: 10.1016/j.ausmj.2014.11.003