Getting Consumers to Recycle NOW ! When andWhyCuteness Appeals Influence Prosocial andSustainable Behavior

Tingting Wang, Anirban Mukhopadhyay, andVanessa M. Patrick

Policy makers and social agencies need novel insights to manage the major global challenge of encouragingprosocial and sustainable behaviors. This research identifies a positive aesthetic cue, “Kindchenschemacuteness,” that can reliably induce some people—specifically, those who exhibit high approachmotivational orientation (per the behavioral approach system [BAS]; Carver and White 1994)—to engagein prosocial and conservation behaviors. Studies 1 and 2 show that consumers high (vs. low) in BAS(measured) react more favorably to conservation appeals featuring cuteness, an effect mediated byexperienced feelings of tenderness. Study 3 replicates the effect using a prime of approachmotivation (BAS)to assess donation intentions. Study 4, a large-scale field experiment conducted over eight weeks atmultiplelocations, shows that people recycle more at bins featuring cute visuals with active (high-BAS)messages compared with bins featuring cute visuals with passive (low-BAS) messages. The authorsconclude with a discussion of practical implications for policy makers.

Keywords: Kindchenschema cuteness, aesthetics, sustainability, prosocial behavior, motivation

Modern economic development has led to great improve-ments in living standards, but there is growing rec-ognition that unprecedented rates of overconsumption

and waste, promoted by economic development, are inflictingincreasing harmon the environment (Geller 2002). The prognosisis grim, and action is needed on multiple fronts. In recognition ofthe scale and scope of the crisis facing our planet, the UnitedNations has set 17 Sustainable Development Goals for 2030,

which aim to tackle environmental and social issues such aseradication of poverty, sustainable agriculture, sustainable watersanitation, and sustainable consumption and production patterns(United Nations Economic and Social Council 2016). The mes-sage is clear: all stakeholders need to take actions to steer theworldonto “a sustainable and resilient path, leaving no one behind”(United Nations Economic and Social Council 2016, p. 1).

In efforts to communicate prosocial messages and instillbehavior change, public policy makers and social agenciesfrequently rely on mass media communications such as ad-vertisements and promotional campaigns. Researchers havenoted that extant policies and promotion tactics are pre-dominantly coercive in nature, aiming either to frighten peopleabout the plight of the environment (e.g., “Every 60 seconds aspecies dies out. Each minute counts”) or to induce them usingmonetary rewards (Griskevicius, Cantu, and Van Vugt 2012;Moller, Ryan, and Deci 2006). Some researchers have claimedthat the coercive strategies tend to prompt defiance and re-sentment and are therefore ineffective, particularly in terms oflong-term behavior change (Geller 2002;Moller, Ryan, andDeci2006). Clearly, prosocial and sustainability initiatives need to beimplemented and communicatedmore effectively. Summarizingthe situation, Prothero et al. (2011) called for fresh perspectivesto help create and implement policies and strategies.

In this spirit, the current research investigates the interplaybetween a noncoercive and frequently used but understudiedelement in communications—namely, cute visuals—and anindividual difference in approach tendency (behavioral ap-proach system [BAS]) with the goal of identifying their jointeffectiveness in driving prosocial and sustainable behaviors.

TingtingWang is Assistant Professor of Marketing, Lingnan College, SunYat-sen University (email: wangtt43@mail.sysu.edu.cn). AnirbanMukhopadhyay is Professor of Marketing and Associate Dean, Hong KongUniversity of Science and Technology (email: anirban.mukhopadhyay@ust.hk). Vanessa M. Patrick is Bauer Professor of Marketing, University ofHouston (email: vpatrick@bauer.uh.edu). This article is based on thefirst author’s doctoral dissertation, which was supervised by thesecond author, with the third author as a member of the dissertationcommittee. The authors thank the other committee members, MelodyChao, Amy Dalton, and Rongrong Zhou, for their helpful comments.The recycling field study was jointly conducted with and funded by theSustainability Unit and the School of Business and Management,Hong Kong University of Science and Technology. The authors wouldlike to express sincere thanks to David Bookhart and all the staffmembers and student helpers at the HKUST Sustainability Unit fortheir continuous help and assistance during the field study. Additionalsupport from the Fundamental Research Funds for the CentralUniversities of China to the first author (16WKPY15) and supportfrom the Hong Kong Research Grants Council to the second author(16502816) is gratefully acknowledged.

© 2017, American Marketing Association Journal of Public Policy & MarketingISSN: 0743-9156 (print) Vol. 36 (2) Fall 2017, 269–283

1547-7207 (electronic) DOI: 10.1509/jppm.16.089269

Cuteness is a popular tactic inmany product categories (Nenkovand Scott 2014) and is not uncommon in prosocial appeals. Forexample, organizations such as the World Wide Fund forNature and the American Society for the Prevention of Crueltyto Animals sometimes employ images of cute animals in theiradvertising, in addition to the more frequently used fearstrategies. Given the enduring success of cuteness in themarket(e.g., Hello Kitty reaps $5 billion annually in profits for Sanrio),it is possible that cuteness may well “sell” people on prosocialand sustainable behaviors (Shrum, Lowrey, and McCarty1994). However, there is no academic research on this im-portant phenomenon, and it is unclear whether, when, and howcuteness can enhance people’s prosocial and sustainability-enhancing behaviors.

Moreover, not every consumer responds to cute products inthe market in the same way. In fact, it is likely that the per-suasiveness of cute prosocial advertising varies across con-sumers (Kollmuss andAgyeman 2002). Therefore, this researchaims to investigate the personal characteristics that shapeconsumer responses to cute prosocial appeals. Building on priorresearch on motivation (Gray 1970, 1981) as well as researchon the evolutionary significance of cuteness (Kringelbach et al.2016; Lorenz 1943; Shaver, Mikulincer, and Shemesh-Iron2010), we propose and demonstrate that consumers’ behavioralapproach tendencies (i.e., BAS; Carver and White 1994) interactwith prosocial cuteness appeals to influence their behaviors, suchthat consumers who are high in BAS respond more positivelyto cuteness in prosocial communications. Moreover, we iden-tify that it is the emotional response to cuteness—specifically,feelings of tenderness—that drive this effect. Our results holdtrue using both self-reported intentions in experiments and ac-tual behaviors in a large-scale field experiment, and our findingscorroborate previous insights into the advantages of noncoercivepolicies (Moller, Ryan, andDeci 2006;Ryan andDeci 2000) andinform policymakers and social entrepreneurs about an effectivenew approach to stimulate behavior change.

Theoretical Development

BAS and Consumer Responses to RewardsIt is widely acknowledged that human beings tend to approachrewards (Gray 1970), a fact that explains the success of sus-tainability programs that provide monetary rewards to engageconsumers (Griskevicius, Cantu, and Van Vugt 2012). How-ever, people differ in their sensitivity to and their general ap-proach tendencies toward rewards. These differential responsesto reward are attributable to an individual difference variable:baseline levels of activation of the BAS (Carver and White1994; Gray 1970, 1981).

TheBAS is a primary biologically basedmotivational systemthat regulates approach motivations. It facilitates the expendi-ture of energy, prompts actions to pursue rewards, and guidesresponses to rewards (Gray 1970, 1981). The strength of theBAS varies across individuals, and these variations in sensi-tivity to rewards are often measured using Carver and White’s(1994) behavioral inhibition system/behavioral approach sys-tem (BIS/BAS) scale. In other words, people with strong BASpossess strong general approach tendencies; are highly responsiveto rewarding cues in the environment; and tend to show intenseemotional, cognitive, and behavioral reactions to rewards. In

contrast, people with weak BAS are relatively insensitive torewarding cues and tend to exhibit indifferent responses to re-warding stimuli and neutral stimuli. According to Henriques andDavidson (2000), people with clinical depression typically haveweak BAS and exhibit reduced sensitivity to rewards. In theconsumption domain, Wadhwa, Shiv, and Nowlis (2008) foundthat participants who scored low on BAS did not approachrewarding consumption cues whether in the presence of a re-warding consumption cue or no cue, while participants whoscored high on BAS exhibited approach behaviors to obtainother rewarding consumption cues when in the rewarding-cue(vs. no-cue) condition. Corroborating these behavioral studies,recent functional magnetic resonance imaging (fMRI) researchprovides direct evidence that when in receipt of monetary re-wards, low-BAS (vs. high-BAS) people showed reduced (vs.heightened) activity in the ventral striatum and the medialorbitofrontal cortex, the neural substrates of reward processing(Simon et al. 2010). These findings all suggest that a rewardingstimulus elicits heightened approach-related responses fromhigh-BAS consumers but fails to elicit such responses from low-BAS consumers.

Cuteness is an inherently rewarding aesthetic cue for humanbeings, who are born with instinctive predispositions to ap-proach cute entities and treat them as rewards (Kringelbach et al.2016; Lorenz 1943). Next, we provide the definition of “cuteness”as investigated in this research and develop propositions forthe interplay between prosocial cuteness appeals and BAS—specifically, howpeoplewith varying general approach tendenciesrespond to prosocial cuteness appeals.

Kindchenschema CutenessColloquial usages of “cute” have various connotations, rangingfrom “someone that is sweet andnice” to “behaviors that are funnyand humorous” (per 2014 entries for “cute” on UrbanDictionary.com). Fortunately, the academic representation of this construct islimited to two distinct facets: Kindchenschema and whimsicality.Kindchenschema is the classical definition of cuteness, whichassesses the degree to which a given stimulus is baby-like inappearance (Lorenz 1943). A prototype would be the humaninfants photographed byAnneGeddes,with their big eyes, roundfaces, and chubby bodies. By definition, Kindchenschemacuteness is linked to infantility and the helplessness of theyoung. In contrast, whimsical cuteness is a more recent additionto the literature; this emphasizes representations of whimsicalfun, playfulness, and capricious humor, as created by marketers(Nenkov and Scott 2014, p. 327)—for example, “an ice-creamscoop shaped like a miniature person or a dress with tropicalcolors and pink flamingos.” In the current research, we focus onKindchenschema cuteness. This concept is not limited to humaninfants but generalizes to other infant-like stimuli, such as humanadults, animals, and inanimate objects (Kringelbach et al. 2016;Lorenz 1943). An entity that possesses a sufficient proportion ofinfantile features, including a relatively big head, round andprotruding cheeks, large eyes, and plump body with soft-elasticsurface texture simulates the physical attractiveness of infantsand thereby causes people to perceive it as being cute (Alley1981, 1983; Lorenz 1943).

Germane to the current research, Kindchenschema cutenesshas the capacity to ignite nurturing instincts and elicits innatelyunique human behaviors, to the extent that Kringelbach et al.

270 Getting Consumers to Recycle NOW!

(2016, p. 545) referred to it as “one of the most basic andpowerful forces shaping our behavior.” Specifically, helplessinfants (human and animal) are highly vulnerable and incapableof caring for themselves and, thus, depend heavily on adults’protection and nurturing to survive. The cute appearance ofthese infant creatures serves the evolutionary purpose ofstimulating instinctive nurturing behaviors in adults, motivatingcaregiving and protection, which thereby ensures the survival ofthe young and increases the evolutionary fitness of the species(Glocker et al. 2009; Lorenz 1943). As a result,Kindchenschemacuteness represents a rewarding stimulus for people. Cutenessprovides hedonic pleasure in and of itself, spurs approachingactions from people, and even generates rich responses frompeople (Hildebrandt and Fitzgerald 1978; Kringelbach et al.2016). It is specifically due to these response patterns that thisresearch focuses on Kindchenschema cuteness to explorewhether, when, and why it is effective in motivating prosocialand sustainable behaviors.

BAS Regulates Responses to Cuteness AppealsBecause a person’s level of BAS influences his or her sensitivityand reactions to rewards, we predict that individual responses toKindchenschematically cute entities are subject to the influenceof BAS. Indeed, Desjardins, Zelenski, and Coplan (2008) foundthat mothers who scored high on BAS reported being morenurturing and caring toward their children than mothers whoscored low on BAS. This suggests that even the basic instinctiveresponses toward cuteness are regulated by people’s BAScharacteristics. Low-BAS mothers have generally weak ap-proach tendencies toward rewards and thus are not as stronglyspurred by cuteness as high-BAS mothers, resulting in reducednurturing behaviors toward their children. Therefore, we predictthat when exposed to a cute (vs. noncute) stimulus, people withstrong motivational approach tendencies (high BAS) are morelikely to express the unique responses elicited by the cutestimulus. In contrast, people with weak approach motivationaltendencies (low BAS) have reduced sensitivity to cuteness andtherefore are less responsive to cute stimuli. Following this logic,we predict that employing cute (vs. noncute) visuals in prosocialcommunications should influence prosocial behaviors amonghigh-BAS consumers but not among low-BAS consumers.

Cuteness and Prosocial and SustainableBehaviorsSustainable behaviors are “deliberative and efficient actions thatresult in the conservation of the socio-physical environment forpresent and future generations” (Corral-Verdugo et al. 2011,p. 95). Sustainable behaviors are, by definition, prosociallyoriented, similar to other prosocial behaviors such as helpfulinterventions and charitable donation, which look out for otherpeople’s well-being and serve collective benefits (Batson 2009).Extant research has stated that prosociality has both evolu-tionary and psychological antecedents. Evolutionary theories ofprosociality build on Darwin’s (1982 [1871]) groundbreakingwork on natural selection to study the existence of prosocialitythroughout evolution and explore why people have evolved tosometimes benefit others at a cost to themselves.

Despite some differences, researchers largely agree thatprosocial behaviors can enhance the evolutionary fitness of the

human species (De Waal 1996, 2008; Sober and Wilson 1998).The argument here is that prosociality is an extension of thenurturing instinct, which is the most fundamental altruisticbehavior that enhances the reproduction and survival of thehuman species (Batson et al. 2005; McDougall 1908; Shaver,Mikulincer, and Shemesh-Iron 2010). Relatedly, a key featureof Kindchenschema cuteness is its capacity to induce nurturingresponses. This implies a possible evolution-based link betweenprosocial behaviors and Kindchenschema cuteness, because thecuteness of an infant spurs adults to care for and protect it.

Importantly, because of the social nature of humans, thesenurturing responses are not restricted to one’s own offspring butrather generalize to “other human beings’ needs for comfort,protection, support, and encouragement” (Shaver, Mikulincer,and Shemesh-Iron 2010, p. 73), which are prosocial behaviors.Thus, human prosociality represents a direct extension of thenurturing instinct that can be triggered by cuteness. Corrobo-rating this, Glocker et al. (2009) find that exposure to cute infantfaces can activate the ventral striatum, a brain region associatedwith general altruistic behaviors, signaling that responses tocuteness and prosocial behaviors may share a common neuralbasis. This reasoning suggests that the use of cuteness inprosocial appeals may induce heightened general prosocialintentions. However, as we reasoned previously, people withstrong BAS are responsive to cuteness, whereas people withweakBAS tend to exhibit indifferent reactions toward cute andneural stimuli. Therefore, we argue that the potential positiverelationship between cuteness and prosociality shouldmanifestitself among people high in BAS but not among people low inBAS. Formally,

H1: People’s approach tendencies (BAS) moderate their responsestoKindchenschema cute (vs. noncute) appeals, such that peoplehigh in BAS show enhanced prosocial and sustainable moti-vations in response to the same Kindchenschema cute (vs.noncute) appeal, whereas people low in BAS do not.

The Mediating Effect of TendernessProsocial behaviors are not automatic but, rather, depend onpsychological and situational triggers (Batson et al. 2005; DeWaal 2008). Existing research has agreed that empathic con-cern, defined as “an other-oriented emotional response elicitedby and congruent with the perceived welfare of someone inneed” (Batson 2009, p. 8), is the motivating impetus for pro-social behaviors (Batson 2009; DeWaal 2008; Lishner, Batson,and Huss 2011). Moreover, empathy is composed of two dis-tinct emotions: tenderness and sympathy. Tenderness is a warm,positive emotional response triggered by perceiving someone’svulnerability (Niezink et al. 2012), whereas sympathy representsa state of compassion toward someone’s plight and is elicited byan appraisal of someone’s current need (Lishner, Batson, andHuss 2011).

Critically, Kindchenschema cuteness has been argued totrigger empathy (Kringelbach et al. 2016), and specifically, itproduces the emotion of tenderness, not sympathy (Lishner,Batson, andHuss 2011). This suggests that the positive effect ofcuteness on prosociality may be caused by its produced tenderfeelings. This theorizing is consistent with the evolutionary linkbetween cuteness and prosociality, as we have articulated.Many researchers have contended that feelings of tendernessoriginate from the nurturing instinct; people’s experienced

Journal of Public Policy & Marketing 271

tender feelings elicited by cuteness help release caregivingbehaviors toward their cute offspring (Batson et al. 2005; DeWaal 2008; McDougall 1908; Sober andWilson 1998). AsFrijda (1986, p. 83) states, “Tenderness can be regarded as theimpulse toward tender—that is, caregiving—behavior.” Fur-thermore, people with strong BAS who have enhanced re-sponsiveness to cuteness should experience stronger feelings oftenderness toward a cute versus noncute stimulus, but in con-trast, people with weak BAS who are insensitive to cuteness donot experience stronger feelings of tenderness toward a cuteversus noncute object. Because tenderness prompts prosocialmotivation, the stronger the tender feelings experienced by high-BAS (but not low-BAS) people, themoremotivated they shouldbe to behave prosocially. Formally,

H2: The positive effect of Kindchenschema cuteness (vs. non-cuteness) on prosociality among people high (vs. low) in BASis mediated by tender feelings produced by cuteness, such thatpeople with high BAS (vs. low BAS) experience greatertenderness in response to a cute appeal than to a noncute appeal,causing them to exhibit stronger prosocial intentions.

Our conceptual framework (see Figure 1) diagrammaticallyrepresents the theoretical model that guides this research.

Overview of the Empirical InvestigationWe report four studies testing our proposition that high-BAS(vs. low-BAS) people exhibit heightened prosocial and sus-tainable intentions toward cute prosocial appeals. Across thesestudies, we (1) expose people to cute appeals that promoteprosociality, (2) measure (Studies 1 and 2) or prime (Studies 3and 4) BAS orientation, and (3) measure proenvironmental in-tentions (Studies 1 and 2), charitable donation intention (Study3), and actual recycling behavior (Study 4) to test the hypoth-esized effects. Study 1 examines participants’ intentions to re-cycle and to consume environmentally friendly products afterexposure to conservation advertisements that feature eitherKindchenschema cuteness or noncuteness; it reveals heightenedprosocial intentions among high-BASparticipants after exposureto the cute ads. Study 2 demonstrates that feelings of tendernessproduced by cuteness mediate this effect. Study 3 primesparticipants’ BAS by using active versus passive appeals andgeneralizes the effect to charitable donation intentions. Study 4tests our propositions in a large-scale field study conducted atmultiple locations on a university campus over eight weeks,

demonstrating the effect of an intervention combining cutenessand active high-BAS appeals on the amount of materialsrecycled.

Across studies, we use a self-reported measure of participants’motivational approach tendencies with Carver and White’s(1994) scale (Studies 1 and 2) or prime participants’ approachmotivation using messages that emphasize (vs. do not emphasize)action (Studies 3 and 4). This method of manipulating BAS isnew to the literature and stems from its core function of regulatingresponses to rewards. The defining feature of high-BAS people istheir heightened responsiveness to and active approach-relatedbehaviors toward rewards, compared with low-BAS people’slack of these actions toward rewards (Gray 1970, 1981;Wadhwa,Shiv, andNowlis 2008). As such, emphasizing action orientationmight prime strong BAS and thus elicit heightened prosocialintentions toward cute prosocial ads. We implement this logic inStudies 3 and 4 by creating active prosocial slogans that stressaction orientation. Specifically, to serve the purpose of enhancingBAS, these slogans (1) call for immediate prosocial actions withspecific words adopted from the Carver and White (1994) scale;(2) are written in italic font, which heightens behavioralmovement (Cian,Krishna, andElder 2015); and (3) are expressedin a relatively assertive tone, which is associated withextraversion, a personality feature linked with strong BAS(Barrick andMount 1991;Muris et al. 2005; i.e., “DonateNOW!”[Study 3] and “RecycleNOW!” and “Recycling isFUN!” [Study4]). We contrasted these active slogans with passive slogans innormal font and in a nonassertive tone (i.e., “Donate please”[Study 3] and “Recycle please” and “Recycle for me” [Study 4]),which functioned as the low-BAS messages. Importantly, boththe scale measure and the prime of BAS yielded similar resultsthat are consistent with our theorizing.

Study 1: Cuteness EnhancesSustainability-Related Intentions Among

High-BAS (vs. Low-BAS) PeopleThe purpose of Study 1 was to test the proposed interactiveeffect of cuteness and participants’ BAS on sustainability-related intentions. For this purpose,wedesignedposters promotingrecycling, which featured images of either Kindchenschemacute animals or noncute animals, and measured respondents’BAS and their intentions to engage in sustainability-relatedbehaviors. We predicted that for high-BAS (but not low-BAS) participants, there would be a positive effect of the cute(but not the noncute) images.

Method

Design and StimuliThis experiment was in a 2 (appeal: cute vs. noncute) × 2 (BAS:high vs. low) between-subjects design, with BAS as ameasuredvariable. Participants (N = 202; 125men;Mage = 33 years) wererecruited for US$.30 on Amazon Mechanical Turk (MTurk).Using the cover story of examining the effectiveness of posterdesign, we showed participants a poster that featured either fourcute animals or four noncute animals with the slogan “Recy-cling saves animals. Please recycle” (see Appendix A).

We selected images of four cute animals (to appear in the cuteposter) that possess multiple prototypical Kindchenschema

Figure 1. Conceptual Framework

BAS• Measured (Studies 1–2)• Primed (Studies 3–4)

Appeal(Cute vs.Noncute)

Feelings ofTenderness

Prosocial andSustainableBehaviors

H1

H2

272 Getting Consumers to Recycle NOW!

features as identified in the literature (Alley 1981, 1983) and,correspondingly, four equivalent noncute animals that pos-sessed few Kindchenschema features to appear in the noncuteposter. We then pretested the selected images. Participants (N =27) were shown a set of four animals, either all cute or allnoncute, and rated the cuteness of each animal (1 = “not cuteat all,” and 100 = “extremely cute”). A one-way analysis ofvariance (ANOVA) showed that the cute animals were indeedperceived as being cuter than the noncute ones (all ps < .05).

ProcedureParticipants were randomly assigned to view either the cuterecycling poster or the noncute one, depending on condition.Afterward, we assessed participants’ sustainability-related in-tentions by assessing their intentions to recycle (“Towhat extentare you willing to recycle after seeing this poster?,” “How likelyare you to recycle after seeing this poster?,” and “Towhat extentdoes this poster motivate you to recycle?”) and their willingnessto try environmentally friendly products (“Howmuch do youwant to try products that are made of recycled materials/biodegradable trash bags/phosphate-free detergents?”). Allquestions were administered on seven-point scales (1 = “not atall,” and 7 = “very much”) and later combined to form an indexof sustainability-related intentions (a = .91). Then, as a ma-nipulation check, we asked participants to rate the cuteness ofthe animals on the poster (1 = “not cute at all,” and 7 = “ex-tremely cute”). Finally, participants responded to the BIS/BASquestionnaire (Carver andWhite 1994), which was presented asan ostensibly different consumer survey.

Results and Discussion

Manipulation CheckA one-way ANOVA showed that the animals in the cute posterwere rated as being much cuter than the equivalent animals inthe noncute condition (Mcute = 6.24 vs. Mnoncute = 5.32; p <.0001).

Sustainability-Related IntentionsParticipants’ scores on the index of sustainability-related in-tentions were regressed on type of appeal (1 = cute, _1 =noncute), BAS (standardized), and the appeal × BAS in-teraction. In support of H1, there was a significant interactioneffect (b = .29, t = 2.93, p < .01), and no other effects weresignificant (p > .10). Follow-up spotlight analyses revealed thathigh-BAS participants (i.e., +1 SD) reported higher intentionsto behave sustainably after having viewed the cute recyclingposter compared with the noncute one (Mcute = 5.38 vs.Mnoncute = 4.74; b = .32, t = 2.31, p < .03), suggesting a positiveeffect of cuteness on sustainability-related intentions amonghigh-BAS consumers. In contrast, among low-BAS participants(i.e., –1 SD) there was a directionally significant difference inthe opposite direction (Mcute = 4.47 vs.Mnoncute = 4.99;b =_.26,t = _1.86, p = .07; see Figure 2). We did not expect thismarginally significant negative effect of cuteness among low-BAS participants. We test this further in the following studiesand return to it in the “General Discussion” section.

We also analyzed the slopes of BAS for the two posters andfound that BAS had a significantly positive effect for the cuteposter (b = .45, t = 3.53, p < .01) but no effect for the noncute

poster (b = _.13, t = _.84, p > .40). That is, among those whosaw the cute poster, higher BAS led to increasingly positiveintentions. In contrast, all participants regardless of BASresponded similarly to the noncute poster.

These results provided initial evidence for our hypothesizedpositive effect of cuteness in enhancing sustainability-relatedmotivation among people who have strong general approachtendencies (H1). Consistent with our predictions, exposure toa poster that presented images of cute (vs. noncute) animalsenhanced high-BASparticipants’ intentions to recycle and to tryenvironmentally friendly products; this effect was not apparentfor participants with low BAS.

Study 2: The Mediating Role of TendernessThe goal of Study 2 was to test the mechanism underlying theeffect observed in Study 1 (H2). We expected that high-BASpeople would be more predisposed to respond with tendernessto a cute (vs. noncute) appeal and that these feelings of ten-derness would drive their prosocial intentions. In this study, wemeasured two different empathic emotions—tenderness andsympathy (Lishner, Batson, and Huss 2011; Niezink et al.2012)—to test whether it is indeed tenderness, rather thansympathy, that drives the proposed cuteness byBAS interaction.

Another purpose of Study 2 was to examine an alternativeexplanation for the observed effect. Prior research has arguedthat cuteness is generally liked and produces general positiveaffect among its viewers (Hildebrandt and Fitzgerald 1978),which raises the possibility that high-BAS people’s positiveresponses to cute prosocial appeals might be caused not bytenderness but by enhanced general positive feelings towardcuteness. Drawing on prior research, we theorize that thegeneral positive affect account is less likely. The extant liter-ature has revealed no conclusive findings regarding the re-lationship between general positive affect and prosocialbehaviors (e.g., Fisher, Vandenbosch, and Antia 2008; Smalland Verrochi 2009). Rather, it has been suggested that prosocial

Figure 2. Study 1: Cuteness Interacts with BAS to EnhanceSustainability-Related Intentions

3

4

5

6

Low BAS (–1 SD)

Sust

aina

bilit

y-R

elat

ed In

tent

ions

High BAS (+1 SD)

Noncute poster Cute poster

Journal of Public Policy & Marketing 273

appeals that produce positive feelings may not be more ef-fective in eliciting prosocial behaviors (Small and Verrochi2009) and could even discourage prosocial motivations (Fisher,Vandenbosch, and Antia 2008). To empirically disentangle theunderlying process, in Study 2 we measured participants’ af-fective responses, including tenderness, sympathy, and generalpositive affect, and conducted mediation analyses to test whichconstruct mediates the observed effect.

MethodStudy 2 employed a 2 (appeal: cute vs. noncute) × 2 (BAS: highvs. low; measured factor) between-subjects design. Peoplerecruited from MTurk (N = 240; 145 men; Mage = 35 years)participated in return for US$.50. We adopted a similar pro-cedure to that of Study 1, except that we also measured par-ticipants’ affective responses. As a cover story, we claimed wewere examining the effectiveness of recycling poster designs.Participants were randomly assigned to view one of the twoposters used in Study 1 and then indicated their willingness torecycle on a seven-point scale (1 = “not at all,” and 7 = “verymuch”). Afterward, participants responded to multiple itemsthat describe how they might have felt when viewing the posteron seven-point scales (1 = “strongly disagree,” and 7 = “stronglyagree”). These items were displayed in randomized order andmeasured feelings of tenderness (three items: “tender,” “warm,”and “softhearted”; Niezink et al. 2012), sympathy (three items:“compassionate,” “moved,” and “sympathetic”; Niezink et al.2012), and general positive feelings (five items: “happy,”“alert,” “inspired,” “determined,” and “attentive”; Positiveand Negative Affect Schedule [PANAS short version];Thompson 2007). Following this, as manipulation check,participants rated the cuteness of the animals in the poster(1 = “not cute at all,” and 7 = “extremely cute”). Finally,they completed the BIS/BAS questionnaire, which wasadministered as an ostensibly different consumer survey.

Results

Manipulation CheckA one-way ANOVA of the averaged cuteness ratings showedthat participants perceived the animal images in the cute posteras being cuter than those used in the noncute poster (Mcute =6.09 vs. Mnoncute = 5.50; F(1, 238) = 18.06, p < .001).

Willingness to RecycleWe regressed participants’ willingness to recycle on appeal(1 = cute, _1 = noncute), BAS (standardized), and theappeal × BAS interaction. As we predicted, there was asignificant interaction effect (b = .21, t = 1.96, p = .05). Noother effects were significant (ps > .10). Follow-up spotlightanalyses revealed that high-BAS participants (i.e., +1 SD)were directionally more willing to recycle after viewing thecute poster than after viewing the noncute one (Mcute = 5.50 vs.Mnoncute = 4.99; b = .25, t = 1.66, p < .05 [one-tailed]). Incontrast, low-BAS participants (i.e., –1 SD) did not differ intheir stated intention to recycle (Mcute = 4.71 vs.Mnoncute = 5.05;b = _.17, t = _1.11, p = .27). This null effect is different fromwhat we observed in Study 1, and we return to this point in the“General Discussion” section.

In addition, we conducted simple slopes analyses and foundthat in the cute poster condition, higher BAS led to significantlyhigherwillingness to recycle (b = .39, t = 2.64,p< .01).However,in the noncute poster condition, there was no effect of BAS onrecycling intentions (b = _.03, t = _.20, p = .84).

Tenderness as the Underlying MechanismWe averaged participants’ responses to each of the three sets ofitems measuring tenderness (a = .96), sympathy (a = .95), andgeneral positive feelings (a = .86), to create indices of each.Wethen regressed the tenderness index on appeal (1 = cute, _1 =noncute), BAS (standardized), and the appeal × BAS in-teraction. There was a marginally significant interaction effect(p = .09), such that high-BAS consumers reported significantlygreater feelings of tenderness in response to the cute poster thanto the noncute poster (Mcute = 5.60 vs. Mnoncute = 4.86; b = .37,t = 2.38, p < .02). In contrast, low-BAS consumers experiencedsimilar levels of tenderness (Mcute = 4.84 vs.Mnoncute = 4.85;b =.0, t = _.02, p = .98). Separate similar analyses with sympathyand positive affect as dependent variables revealed no signif-icant interaction effect on sympathy (p < .19, n.s.) or positiveaffect (p < .17, n.s.), suggesting that neither sympathy nor pos-itive affect mediates the observed effect.

To directly test whether tenderness, and not the other twoaffective states, mediates the observed effect, we conductedthree separate mediated moderation analyses with tenderness,sympathy, and general positive feelings as the possible medi-ators (Model 7, Hayes 2013). Most importantly, introducingtenderness to the regression model on participants’ willingnessto recycle yielded a significant indirect effect of the appeal ×BAS interaction through tenderness. Among high-BAS par-ticipants, as we predicted in H2, tenderness mediated the effecton their willingness to recycle (b = .21, bootstrapped 95%confidence interval [CI]: [.04, .44]; 5,000 samples). In contrast,tenderness did not mediate the effect among low-BAS partic-ipants (b = 0, bootstrapped 95%CI: [_.17, .17]; 5,000 samples),and the direct effectwas no longer significant (b = _.08, p= .40).

Additional analyses ruled out possible mediating roles ofsympathy and positive affect. When sympathy (instead of ten-derness)was used as themediator, therewas no significant indirecteffect of the appeal × BAS interaction through sympathy amongeither high-BASparticipants (b = .16, bootstrapped 95%CI: [_.01,.37]; 5,000 samples) or low-BAS participants (b = 0, bootstrapped95% CI: [_.17, .17]; 5,000 samples). Similarly, using generalpositive affect (instead of tenderness) revealed no significantindirect effect of the appeal × BAS interaction through positiveaffect among high-BAS participants (b = .17, bootstrapped95% CI: [_.001, .38]; 5,000 samples) or low-BAS participants(b = 0, bootstrapped 95% CI: [_.15, .16]; 5,000 samples).

DiscussionStudy 2 builds on Study 1 to support our proposition thatparticipants’ general approach tendencies (BAS) determine theirresponses to prosocial cuteness appeals (H1). Importantly, itprovides direct support for our proposed underlying mechanismimplicating feelings of tenderness (H2). As the mediation anal-yses results show, high-BAS participants’ enhanced prosocialmotivation was mediated by their experienced feelings of ten-derness. Notably, similar tests of mediation revealed that neitherparticipants’ experienced sympathy nor their generalized positive

274 Getting Consumers to Recycle NOW!

affect mediated the effect, ruling these out as alternativemechanisms. These results also help rule out the possibility ofa demand effect, because the different items measuring partic-ipants’ affective feelings were randomized and intermingled,which should have led to reduced awareness of the study’spurpose. Furthermore, had participants responded to these af-fective items as they thought they should (e.g., participants in thecute condition intentionally reporting greater affective feelingsthan they really experienced), we would have observed similarpatterns with regard to sympathy and generalized positive affect.In addition, the null effect of general positive affect is consistentwith previous research showing that positive affect may not ef-fectively elicit prosocial behaviors (Fisher, Vandenbosch, andAntia 2008; Small andVerrochi 2009). Importantly, in this studywe were able to tease apart two distinct empathic emotions,tenderness and sympathy. The mediation analyses revealed thathigh-BAS participants’ enhanced prosocial motivation is drivenby experienced tenderness rather than sympathy. This findingcorroborates previous research stating that cuteness elicitsfeelings of tenderness, which are emotional antecedents of nur-turing tendencies (Frijda 1986; Sober and Wilson 1998).

Study 3: Cuteness and Primed ApproachMotivation Enhance Charitable IntentionsStudy 3 aims to lend additional support to the conceptualframework and extend it in two key ways. First, it examines amore general prosocial intention—people’s intention to donateto charities—to generalize the scope of the effect. In addition, itcaptures the role of approachmotivation by priming (instead ofmeasuring) BAS by displaying either an action-oriented activemessage versus a passive one. To recap our logic, cuteness, as areward, spurs approach motivation, and BAS regulates ap-proach tendencies toward rewards. People with strong generalapproach tendencies actively approach rewards, whereas peoplewith weak approach tendencies show reduced sensitivity torewards (Gray 1970, 1981). As such, messages that stressaction orientation may prime strong BAS and subsequentlyelicit enhanced responses toward cuteness. Following this logic,we expect that prosocial advertisements that combine high-BASmessages with cute visuals should be more likely to induceheightened donation intentions, akin to the effect of cuteness forhigh-BAS people who display chronically strong approachtendencies.

Method

Design and StimuliThis study followed a 2 (appeal: cute vs. noncute) × 2 (BASpriming: active vs. passive) between-subjects design and ex-amined participants’ responses toward charitable advertise-ments that request donations to help animals. We created fourposters featuring images of a dog with a caption (see Appen-dix B). We manipulated cuteness using either a cute puppy or anoncute adult dog in the images. These images were pretestedon participants (N = 30) who were randomly shown one of thetwo images and rated its cuteness (1 = “not cute at all,” and 7 =“very cute”). A two-sample t-test confirmed that the cute dogwas perceived as significantly cuter than the noncute one(Mcute = 6.03 vs. Mnoncute = 4.23; t(29) = 6.60, p < .001).

As manipulation of BAS, the advertisements featured eitheran action-oriented message that called for a donation written initalics with an exclamation mark (“Your donation can helpanimals. Donate NOW!”) in the high-BAS condition or apassive message written in normal font (“Your donation canhelp animals. Please Donate”) in the low-BAS condition. Weexpected that the BAS priming would affect participants’prosocial intentions only in the cute image condition, and thatthe priming alone would not directly influence prosocial in-tentions. To confirm this, we conducted a separate study withparticipants (N=64), inwhich theywere randomly shown eitherthe poster featuring the high-BAS message or the poster fea-turing the low-BAS message (no visual images appeared ineither poster). There were no differences in prosocial intentionsacross the two BAS-priming conditions (Mactive = 4.02 vs.Mpassive = 4.39; F(1, 62) = 1.17, p = .28).

Participants and ProcedureParticipants (N = 144; 81 men; Mage = 36 years) were recruitedfor US$.50 onMTurk. Participants were told that a local animalprotection agency had designed a poster to generate donations tohelp save animals and improve their living conditions, and wewere interested in their reactions to the poster. Depending oncondition, participants were shown one of the four posters. Afterthey saw the poster, participants indicated the extent to whichthey were willing to donate their payment for participating inthis research to help the animals (1 = “not at all,” and 7 = “verymuch”). As amanipulation check, they also rated the cuteness ofthe animal appearing in the poster (1 = “not at all,” and 7 =“extremely cute”).

Results

Manipulation CheckA one-way ANOVA showed that the animal image used in thecute condition was rated as significantly cuter than the one usedin the noncute condition (Mcute = 6.19 vs. Mnoncute = 5.16;F(1, 142) = 9.55, p < .001).

Willingness to Donate Research PaymentA two-way cuteness × prime ANOVA on participants’ will-ingness to donate their payment revealed a significant in-teraction (F(1, 140) = 3.87, p = .05). No other effects reachedsignificance (ps > .30). As predicted, there was a significantsimple effect of BAS priming in the cute condition, such thatparticipants who saw the cute appeal that displayed the active(high-BAS)message weremore willing to donate their paymentthan those who saw the cute appeal with the passive (low-BAS)message (Mactive = 3.21 vs. Mpassive = 2.22; F(1, 140) = 4.23, p <.05). In contrast, participants had similar intentions to donate afterviewing the noncute active versus the noncute passive poster(Mactive = 2.73 vs. Mpassive = 3.05; F(1, 140) = .49, p > .45).

DiscussionStudy 3 replicates the interactive effect of cuteness and BAS(primed) on charitable intentions to provide further supportfor our conceptual framework. Participants who saw the pro-social appeal featuring the combination of a cute visual anda high-BAS message reported greater intentions to donate

Journal of Public Policy & Marketing 275

than others. The generalization of our proposed effect fromproenvironmental intentions (Studies 1 and 2) to charitable do-nation intentions lends additional external validity to our findings.

Taken together, the results of Studies 1–3 are supportive ofour conceptual framework, but each of them measures in-tentions and not actual behaviors. It is well known that there issome inconsistency between consumers’ articulated attitudesand their actual behaviors (Fazio, Powell, and Williams 1989).This attitude–behavior gap is particularly relevant to the currentcontext of prosocial and sustainable behaviors because manypeople may articulate what they believe is the “right” thing to dobut do not follow through with their actions (Kollmuss andAgyeman 2002).We explore this notion in Study 4, a large-scalefield study conducted at a university campus, in which wedemonstrate the positive effect of prosocial cuteness appealswith BAS orientation on actual recycling behaviors over anextended period of time.

Study 4: Recycling Field ExperimentThis field experiment had two purposes. First, we aimed to testwhether our proposed interactive effect of cuteness andBAS, asobserved on intentions in the preceding studies, influencedpeople’s actual recycling behaviors. Second, we wanted toobserve whether recycling behaviors promoted by high-BAScuteness appeals had a relatively long-lasting effect or wore outquickly. To serve these purposes, we collaborated with a uni-versity’s sustainability unit on a field intervention, in which weused appropriately designed, visually appealing cute images andmessages on recycle bins and measured the amount of recycledmaterials over eight weeks.

Our study design was subject to some key considerations andconstraints and, thus, has some limitations. Most notably, thetotal cost of one recycling bin, including sourcing the bin anddesigning and producing the images, was over US$500. Thisimposed a limit on the total number of binswe could install, withimplications for the study design. Given that our key propositionwas that approach motivation regulates responses to cuteness(H1), we chose to design bins that were all cute, but featuredeither active (high-BAS) or passive (low-BAS) messages.

Method

Quasi-Experimental DesignsGiven the amount of traffic, estimated amount of use of re-cyclable materials, and ease of comparisons, we initially se-lected four undergraduate dormitories (UG Halls 1, 2, 8, and 9)and the new business school building (SBM hereinafter) for thisfield study. Due to unforeseen data collection issues, we had toexclude UG Halls 8 and 9 from the study. Consequently, ourfinal data are fromUGHalls 1 and 2 (separate buildings), whichare similar in architectural structure and could be closelymatched to each other, and from SBM, in which the comparisonwas within the building across floors. UG Halls 1 and 2 ac-commodate approximately 1,000 residential undergraduatestudents (mixed genders, from different majors and years), andSBM hosts approximately 4,300 students and 250 faculty andstaff. The study lasted eight weeks, fromMarch until earlyMay,in two phases: a control phase lasting three weeks, and anexperimental phase that lasted five weeks.

UGHalls 1 and2werepaired in a time-trenddesign. Prior to thestudy, the dorms had only one recycling bin placed outside the hallentrance (each bin had three compartments for paper, plastic, andmetal). During the study period, we installed one recycling bin inthe common roomoneach floor of the dorms. In the control phase,normal three-compartment recycling bins were introduced to bothhalls. Then, in the experimental phase, all bins in UGHall 1 wereswitched to reveal a passive (low-BAS) message coupled with animage of a passive cute animal (described next). At the same time,all bins inUGHall 2were switched to reveal an active (high-BAS)message coupled with an image of an active cute animal.

The SBM design was more complicated because differentpopulations use this building and there is a varied amount oftraffic on each floor. The first three floors consist of classroomswhile the top three floors house faculty offices (offices from thesame department are located together on the same floor). In thecontrol phase, normal recycling bins were introduced on eachfloor. Then, in the experimental phase, all bins were switched toreveal the cute visuals and messages. On student floors, threesets of recycling bins were systematically placed at three dif-ferent types of locations: high-traffic areas (i.e., near a lift), high-visibility areas (i.e., at the end of a hallway), or common-studyseating areas. On faculty floors, two sets of recycling bins wereplaced near each of the twodepartments’general offices.Alternatefloors displayed the active and passive executions.

Pretest: Stimulus DevelopmentOur objective was to create four different images to be used asfrontage for the recycling bins. The images all needed to be cute,but had to vary in the extent of the approach motivation theyelicited. Following the same logic as in Study 3, we createddifferent executions, which combined either images of cute babyanimals that were running toward the viewer, with active slogansin italic font and exclamation marks (i.e., “Recycle NOW!” and“Recycling is FUN!”; high-BAS condition), or images of pas-sive cute baby animals with passive slogans in normal font(i.e., “Recycle please” and “Recycle forme”; low-BAScondition).In line with Study 3, we expected that the executions in the high-BAS condition would prime people with approach motivations.

Using the created slogans and selected images of cute ani-mals, we initially designed six different executions, either cuteactive (high-BAS) or cute passive (low-BAS). To avoid ex-posing the stimuli to the focal subjects, we pretested these on 89college students (37 men) at a different local university. Par-ticipants were randomly presented with one of the six execu-tions and asked to answer two questions: (1) “Does this postermake youwant to recycle?” and (2) “Howmuch do you like thisposter?” (1 = “not at all,” and 7 = “very much”). One-wayANOVAs revealed that there was no difference in participants’statedmotivation to recycle (p > .14), but there was a significantdifference in liking (F(5, 83) = 2.42, p < .05). Thus, we deletedthe two posters that participants liked the most and the least,leaving four executions (two active and two passive) rated asequally likable (F(3, 39) = 2.05, p> .12). To increase confidencein any effects we might observe, we used both executions foreach condition. A posttest in which MTurk participants (N =374) were shown either the active or the passive posters,between-subjects, and were then administered the BIS/BASscale (Carver and White 1994), revealed that the active postersenhanced approach motivation (BAS score) directionally morethan the passive posters did (Mactive = 39.20 vs.Mpassive = 38.20;

276 Getting Consumers to Recycle NOW!

F(1, 373) = 3.03, p = .08). The four selected executions (seeAppendix C) were printed on high-quality foam boards of thesame dimensions as the recycling bins (900 mm × 760 mm),which were stuck on the bins.

Data Tracking: Custom-Made Bins and BagsFifty-two recycling bins were custom-made, each containingthree separate built-in compartments for paper, metal, andplastic. Each bin was assigned a three-digit code, such that thefirst digit represented the location (e.g., UG Hall 1 = 1), thesecond digit represented the floor the bin was on, and the lastdigit represented the type of location (e.g., residence hallcommon room = 0). For example, the bin assigned the code“120” was placed in the common room on the second floor inUG Hall 1. We then ordered thousands of transparent plasticcollection bags and marked each bag with the code number ofthe bin inwhich itwould be placed and letters representing paper(PP), metal (M), and plastic (PL). For example, bags markedwith “120PP”were put under the paper-collecting compartmentof the bin numbered “120.”

Data CollectionAcross the control and experimental periods, the amount ofrecycling in all locations was regularly weighed and recorded.We coordinated 7 student helpers, 29 janitorial staff, 34 UG hallattendants, and 12 managerial staff in the collection of data.Specifically, the janitorial staff collected recyclables daily(twice per week for SBM) from our new recycling bins,replaced them with appropriately labeled empty plastic bags,and transferred the marked filled bags to designated weighingand storage areas. The student volunteers weighed the recycledmaterials and recorded the data for each bin daily (twice perweek for SBM). They then transferred the recycled materialsto a central collection center for municipal pickup.

Results and DiscussionDuring the study period, over 3,087 pounds of materials wererecycled at the installed bins at these five locations. Approxi-mately 1,103 pounds came from SBM, and the remaining 1,984pounds was from the undergraduate dormitories.

SBM BuildingIn the business school building, we computed a measure ofrelative increase (or decrease) in recycling for each individualbin during the experimental versus control period by dividingthe amount of material recycled at that bin during the exper-imental phase by the average amount of goods recycled thereduring the control phase. Thus, each bin functioned as itsown control. We regressed this relative amount of recyclingon condition (high-BAS active bins = 1, low-BAS passivebins = _1), while controlling for time (coded from 1 to 16,representing the first to the last data point in the study) and binnumber. The regression results revealed significant effects ofcondition (Mactive = 1.55 vs.Mpassive = 1.02;b = .20, t = 2.32, p<.03) and time (b = .19, t = 2.20, p = .03). As we predicted, thecute active posters enhanced the amount of recycling by up to55% compared with the regular recycling bins, whereas thepassive recycling posters did not have any incremental effect.Moreover, the time trend was positive, which suggests that the

positive effect of the manipulated bins did not wear off but,rather, was sustained over the study period.

Additional one-way ANOVAs investigating different typesof users (student floors = 1, faculty floors = 2) revealed nodifference (Ms > 1.10; F(1, 126) = .35, p > .80). One-wayANOVAs on different types of locations also yielded no dif-ferences among social study areas, high-traffic areas, and high-visibility areas on the student floors (F(2, 77) = 1.46, p = .24)and no differences between the two wings on the faculty floors(F(1, 46) = .29, p = .60). The patterns of total quantities recycleddid not vary across replicate (i.e., the two active executionsweresimilar to each other, as were the two passive executions).Moreover, further investigation revealed that the observed ef-fects were driven by quantities of plastic and paper recycledrather than metal (metal constituted less than 15% of therecycled materials by weight, far less than the others).

UG HallsFor each recycling bin, we again divided the amount of re-cycling in the experimental period by the average amount ofrecycling across the control period across all recycling bins. Wethen regressed this measure of relative increase or decrease inrecycled amounts on bin design condition, location, and time.The regression analysis revealed significant effects of condition(Mactive = 1.27 vs. Mpassive = .93; b = .24, t = 2.04, p < .05),location (b = _.28, t = _2.39, p < .05), and a marginally sig-nificant effect of time (b = .08, t = 1.70, p < .09). In particular,the active cute recycling bins increased the amount of recyclingby 27% compared with the regular recycling bins, while thepassive cute recycling bins had no effect (if anything, they had aslightly negative effect), replicating the result from SBM. Dormoccupants are assigned by gender to floor, and, using floor as proxyfor gender,we foundnovariationbygender (p> .30) (seeFigure 3).

Time TrendThe positive effect of time in the regression analyses suggeststhat the cute active recycling bins induced a sustained effectover time. As an additional test, we conducted a t-test of theweights (in pounds) collected in each recycling bin in UGHalls1 and 2 on the first versus the last day of the experimental phase.The results revealed amarginal increase (Mfirst = 1.41 vs.Mlast =3.40; t(17) = _.18, p < .09). For SBM, a similar analysisrevealed a significant difference (Mfirst = 5.45 vs. Mlast = 14.48;t(19) = _3.84, p < .001). More recyclables were collected in thelast day than the first day of the experimental phase. Pooling thedata of UG Halls and SBM yielded a significant difference(Mfirst = 3.51 vs. Mlast = 9.24; t(37) = _3.93, p < .001), showingthat people recycled more materials at the end of the experi-mental phase than when they first saw the cute bins.

In addition, we counted the number of recycling bins thatcollected equal or greater amounts of materials on the last day/half-week of the experimental phase than on the first day/half-week of the experimental phase. In UGHalls 1 and 2, out of thetotal 18 recycling bins, 10 bins (56%) collected more recyclablematerials on the last day than on the first day, and 2 bins (11%)collected equal amounts on both days, accounting for 67% ofthe recycling bins. Similarly, we found that in SBM, out of the20 recycling bins, 18 bins (90%) collectedmorematerials on thelast half-week than on the first half-week of the experimentalphase, and one recycling bin collected equal amounts on bothhalf-weeks (5%), constituting 95% of the bins. Consistently, we

Journal of Public Policy & Marketing 277

observed that the semiweekly absolute quantity of recycledgoods averaged across all active bins (and all passive bins) inSBM first increased but then decreased in the control phase(i.e., the impact of introducing new bins wore off quickly), andthen exhibited a slow upward trend over the experimental phase(we observed a similar trend for the passive bins; i.e., the effectof the cute interventions tended to sustain). These results sug-gest that the effect of the cute high-BAS interventions did notdwindle over time.

DiscussionThe results of this field study replicate Study 3 and demonstratethat all cute appeals are not uniformly effective—those com-bined with approach messages have a positive impact onpeople’s actual recycling behaviors, which is consistent with ourproposition (H1). The Study 4 results provide cogent and strongevidence for the generalizability of the cuteness × BAS effectover different populations: we observed the same pattern ofresults across the business school building and undergraduateresidence halls, where different manipulations of BAS wereutilized with different target populations.

Moreover, the observed time trends showed that people’srecycling behaviors lasted the duration of the experimentalportion of the study (i.e., eight weeks), which suggests that theobserved positive effect does not wear off quickly but ratherinduces a sustained increase in recycling over several weeks.This is probably driven by the intrinsically rewarding nature

of the evolutionarily functional Kindchenschema cuteness forpeople, as previous research has shown that communicationsthat use intrinsic incentives tend to elicit sustained behaviorchange over time (Moller, Ryan, andDeci 2006; Ryan andDeci2000). Another possibility is that during the eight-week period,people may have developed enhanced familiarity with therecycling bins and even developed habits (thus reducing theperceived effortfulness of recycling, which previous researchhas shown to encourage recycling behaviors; Ludwig, Gray,and Rowell 1998; Pieters 1991). It is worth noting that in thecontrol phase after the introduction of the new bins, the recyclingactivity first increased but then decreased, which suggests that asthe recycling bins became more familiar, people’s motivation torecycle may have decreased. However, the upward trend ofrecycling in the experimental phase seems to suggest that thepositive effect of our interventions may have overridden thedownward trend and sustained its effect over several weeks.

We note several limitations in the design of field study. First,we did not measure the amount of trash, as opposed to re-cyclables, that was disposed of at the chosen locations. Thisallows for the alternative explanation that the cute high-BAS binsenhanced people’s tendency to dispose of materials in general,leading to increases not only in recycling but also in the amount ofregular trash.1 Our data do not allow us to address this possi-bility. However, it has been suggested that Kindchenschemacuteness enhances consumers’ motivation to retain (vs. discard)objects (Jia, Pol, and Park 2017), a finding that contradicts theaforementioned alternative explanation. Another limitation of thisstudy is that we did not measure potential recyclers’ chronic BAScharacteristics and thus are not able to investigate the relativeinfluence of our interventions on people’s chronic BAS. In ad-dition, we did not measure specific demographic informationand were unable to generate more insights regarding the role ofindividual differences (Iyer and Kashyap 2007). Furthermore, therecycling bins were positioned in public locations, and thus, therecycling activity was measured at group level rather than atindividual level. Finally, we discontinued measuring and re-cording of recycled materials immediately after the study pe-riod, but the bins with the designated posters remained in thelocations for some time after the study ended; consequently, wewere unable to examine whether the influence of our interventionssustained even after the manipulated posters were removed.

General Discussion

Theoretical ContributionsConvergent findings from the lab and the field provide strongsupport for our propositions. Visual cues of cuteness enhanceprosocial and sustainable motivations in people who havestrong (vs. weak) approach motivation. This proposition holdstrue for appeals that encourage recycling behaviors (Studies 1,3, and 4) and those that request for donations to help animals(Study 3). Moreover, the mediation analyses in Study 2 indicatesthat feelings of tenderness (vs. sympathy or general positiveaffect) underlie the observed cuteness × BAS interaction effect.

Most importantly, our findings contribute to the literature onprosociality and sustainability by demonstrating that unlike thefrequently used coercive strategies to elicit behavior change,

Figure 3. Study 4: Cute, Active, High-BAS Bins EnhanceRecycling at SBM and UG Dorms

A: SBM

200%

0%

50%

100%

150%

Passive Bins Active BinsRel

ativ

e A

mou

nt o

f Rec

yclin

g in

the

Exp

erim

enta

l (vs

. Con

trol

) Per

iod

B: UG Dorms

0%

20%

40%

60%

80%

100%

120%

140%

UG Hall 1 (Passive) UG Hall 2 (Active)Rel

ativ

e A

mou

nt o

f Rec

yclin

g in

the

Exp

erim

enta

l (vs

. Con

trol

) Per

iod

1We thank a reviewer for bringing this to our attention.

278 Getting Consumers to Recycle NOW!

Kindchenschematically cute visuals, which are positivelyvalenced and aesthetically appealing stimuli, can enhanceprosocial and sustainable behaviors among high-BAS people.The extant literature has investigated a range of persuasionstrategies for prosociality but has somewhat neglected the role ofvisual aesthetics.Our research adds to this literature by revealingthat aesthetically cute messaging enhances high-BAS (but notlow-BAS) people’s motivation to engage in prosocial andsustainable causes. Moreover, this finding provides a betterunderstanding of how consumers with different personalitycharacteristics respond to cuteness. Extant research has pre-dominantly focused on stimulus-based determinants of cuteness(Alley 1981, 1983), and little research has investigated indi-vidual differences in responses to cuteness. Gender has been amain exception, and research has found no reliable gender dif-ferences in responses to cuteness (Parsons et al. 2011). Thisresearchgoesbeyondgender and shows that individual differencesin motivational approach tendencies moderate people’s responsesto cuteness. Approach tendencies, whether measured (Studies1 and 2) or manipulated (Studies 3 and 4), regulate responses tocuteness. Gender, in contrast, has no effect in our data.

In addition, our field study showed that people’s enhancedrecycling activity in response to communications featuring thecombination of cute visuals and high-BAS active messages canlast over a period of several weeks. This finding is congruentwith previous research showing that noncoercive tactics thatsupport people’s autonomy rather than tactics that pressure orscare people into changing are relatively more effective instimulating behavioral compliance (Moller, Ryan, andDeci 2006;Ryan and Deci 2000).

Furthermore, our priming of strong BAS using action-oriented messages adds to related prior research. The high-BAS messages used an italic font, which has been shown toheighten behavioral movement (Cian, Krishna, and Elder2015). Moreover, research has shown that assertive (vs. non-assertive) communication may decrease consumer compliancebut can be more effective in persuasion when the concernedissues are perceived as important (Kronrod, Grinstein, andWathieu 2012). Adding to these findings, in Study 3 and thefield study we employed assertive (vs. nonassertive) languagecombined with other methods to induce strong (vs. weak) BAS,because assertiveness is common among extraverts, which isexplained by their strong BAS. We found that the active (high-BAS) slogans, together with the display of cute (vs. noncute)visuals, can motivate heightened prosocial intentions and be-haviors. This suggests that assertive language combinedwith cutevisuals might be more persuasive than nonassertive language andcute visuals, a question that future research can further investigate.

The observed null effect of cuteness among low-BAS par-ticipants is consistent with previous findings that people withweak general approach tendencies have lower sensitivity torewards and lack approach-related behaviors toward rewardingstimuli (Gray 1970, 1981; Henriques and Davidson 2000). Thefinding that a motivational factor rather than a cognitive factorregulates the effect of cuteness is important because it is deriveddirectly from the theory of cuteness as a construct and itsevolutionary basis. This effect also derives directly from evo-lutionary theories of prosociality (De Waal 1996, 2008; Soberand Wilson 1998) and corroborates the proposition thathumans’ nurturing instinct is a basis of general prosocial be-haviors (Batson et al. 2005; McDougall 1908).

In line with previous research, we found that cues of cutenesselicit feelings of tenderness, which is a basic type of empathicemotion (Lishner, Batson, and Huss 2011; Niezink et al. 2012),and these tender feelings lead to heightened motivation tobehave prosocially. This finding agrees with the consensus inthe prosociality literature that empathic concern plays a crucialrole in motivating prosocial behaviors (Batson 2009; De Waal2008; Lishner, Batson, and Huss 2011). Moreover, the medi-ation analyses revealed that neither sympathy nor generalpositive feelings mediate the observed effect. This rules outpotential alternative explanations and is consistent with pre-vious findings that positive affect may not always motivateprosocial behaviors (Fisher, Vandenbosch, and Antia 2008;Small and Verrochi 2009). Altogether, the findings suggestthat the observed effect is not due to reliance on feelingsversus cognition but is driven by the specific emotion oftenderness.

Limitations and Future ResearchOne may wonder whether the construct of anthropomorphism,which has been shown to elicit heightened intentions to protectthe environment (Tam, Lee, and Chao 2013), is relevant to ourfindings. For example, Tam, Lee, and Chao (2013) showed thatan advertisement with a talking tree can increase environ-mentally friendly attitudes resulting from increased feelings ofconnectedness with nature. However, the literature has clearlyshown that cuteness and anthropomorphism are two differentconstructs: human-like features may lead to humanization(Tam, Lee, and Chao 2013), but only Kindchenschema featuresconstitute the perception of cuteness (Lorenz 1943). Essentially,not all cute objects are humanized, and not all humanizedobjects are cute. In our investigation, we used nonhuman an-imals in all conditions. Our cute and noncute stimuli possessequal amounts of human-like features and thus do not differ inanthropomorphization. Therefore, unless one argues that non-cute animals are somehow less human than cute baby animals,it is not possible to attribute the observed differences acrossconditions to anthropomorphization. In addition, themoderatingeffect of participants’ motivational approach tendencies origi-nates from the unique evolutionary advantage of cuteness as amajor elicitor of parental instinct, which has nothing to do withanthropomorphization. Our finding that feelings of tendernessproduced by cutenessmediate the observed effect help triangulateon the role of cuteness rather than humanization.

Another research direction worth exploring relates to theinfluence of Kindchenschema versus whimsical cuteness onprosocial motivation. The playful nature of whimsical cutenesssuggests that it may have no effect, or even a negative effect, onprosocial motivation, and this may be a notable point of contrastfor future research. Moreover, recent work has suggested thatthe concept of Kindchenschema cuteness can be extended toinclude babyish features of senses other than vision, such aspositive infantile sound and smell, and that cutenessmay affecthuman behaviors through these senses (Kringelbach et al.2016). Such an expansion of Kindchenschema suggests thatadopting cute infantile sounds in prosocial advertisingmay havesimilar positive effects on encouraging prosocial behaviors, anidea worth future examination.

Finally, apart from the established positive effect of cutenessfor people with strong BAS across studies, we believe that the

Journal of Public Policy & Marketing 279

question of how to motivate low-BAS people’s prosociality istheoretically interesting and practically important. Our datarevealmixed results regarding the effect of cuteness among low-BAS people. In Study 1, we find that low-BAS people respondless favorably to cute appeals than to noncute ones (p = .07),suggesting a potentially negative influence of cuteness amonglow-BAS consumers. This effect is not replicated in Study 2 (p=.84, self-reported BAS). In Study 3, when participants’ BASwas manipulated, we observe a marginally significant negativeeffect of cuteness in the weak BAS priming condition (p = .08).Given the mixed findings from our data, we believe additionalinvestigation is needed to establish the reliability of and thepossible reasons for this negative effect among low-BASpeople.It is possible that low-BAS people might react more strongly topersuasion attempts that aim to shape their behaviors and thusrespond negatively to prosocial advertisements that employcute visuals (Friestad and Wright 1994). Future research caninvestigate this proposition.

Practical Implications for Policy Implementationand CommunicationThis research shows that using aesthetic stimuli characteristicof cuteness can enhance prosocial and conservation behav-iors among consumers who have strong approachmotivation.In terms of implementation, policy makers have prioritizedreducing transaction costs related to sustainable behaviors,such as increasing the provision of recycling bins to enhancethe convenience and ease of recycling (Pieters 1991). Con-sistently, our field study showed that the mere introductionof recycling bins can increase the amount of recycling. Thisstudy revealed that the appropriate use of cuteness can furtherstimulate recycling behavior; specifically, the recycling binsthat feature the combination of cute visuals and strong BASmessages engendered more recycling activity than the cutelow-BAS bins. Moreover, people’s enhanced recycling ac-tivity toward the cute high-BAS recycling bins lasted over aperiod of several weeks. Across studies, we found that thecombination of cute visuals and strong BAS heightenedpeople’s intentions to behave in prosocial and sustainablemanners as well. Therefore, drawing on these findings, werecommend that governments, environmental practitioners,and policy makers consider adopting the noncoercive aesthet-ically cute appeals to subtly motivate prosocial behaviors. Ourfindings suggest two distinct approaches to implement thecuteness tactic: (1) segment the market on the basis of con-sumers’ BAS characteristics and target high-BAS consumerswith cuteness appeals (Studies 1 and 2) or (2) prime strong BASusing action-oriented messages in ads that feature cute visualsand target these ads to the public (Studies 3 and 4).

The segmenting approach corroborates a basic marketingtenet that it is more efficient to divide people into differentsegments in line with meaningful criteria and target persuasionefforts to each segment (Shrum, Lowrey, and McCarty 1994).This approach points to the need for policy makers to accountfor differences across individuals when selecting target markets,positioning products, and designing communication strategies(Sciandra, Lamberton, and Reczek 2017). Relatedly, ourfinding provides a valuable indicator of high-yield consumerprofiles—consumers who possess strong BAS. However,from a practical standpoint, it is indeed challenging to segment

the general population and identify target consumers accordingto their BAS characteristics. Sciandra, Lamberton, and Reczek(2017) offer a practical solution to address this issue: usesurveys to identify observable and easy-to-measure variables(demographic, psychographic, or behavioral variables) that arehighly correlated with people’s approach tendencies, and thenuse these variables as proxies to segment and spot targetconsumers. Given the limitation of our data, which contain fewdemographic variables (gender, age, nationality, and nativelanguage), we recommend that policy makers and social en-trepreneurs pinpoint specific observable correlates ofBAS in thegeneral population (e.g., demographic profiles, common life-styles, shopping patterns) to use for segmentation. Previousresearch has revealed several potential correlates of strong BAS,including personality traits such as extroversion and impulsivity(Muris et al. 2005), behavioral predispositions such as risky be-haviors (Cooper, Agocha, and Sheldon 2000), and demographicfactors such as relatively strong BAS among adolescents(Doremus-Fitzwater, Varlinskaya, and Spear 2010). Future en-deavors can further investigate these variables to guide policymakers.After identifying consumerswhohave strongBAS,policymakers could employ cute visuals in the design of prosocial andsustainability-related messages to induce them to act accordingly.

The priming approach provides a valid alternative methodof employing the cuteness tactic—apart from using ob-servable correlates of BAS, policy makers can prime strongBAS through action-oriented messages. Across both lab andfield experiments, we found that priming BAS using action-oriented slogans combined with cute visuals enhancedpeople’s prosocial and sustainability-related behaviors be-yond the influence of their chronic BAS characteristics. Thispriming approach can also be effective in influencing rel-atively large and diverse populations. As such, we suggestthat public policy makers combine BAS-related primes withcute visuals to elicit desired prosocial behaviors. Policymakersmight consider designing prosocial and sustainability-relatedappeals that feature a combination of cute images and action-oriented messages (e.g., active slogans that call for immediateactions in italic font and in assertive tone, which function toprime strong BAS), and target these ads to the general pop-ulation to elicit enhanced prosocial and sustainable behaviorsthat leverage the influence of the cuteness aesthetic for thegreater good. Like our recycling field study, environmentprotection agencies may consider decorating recycling binswith posters that display the combination of action-orientedhigh-BAS messages and cute images to motivate participationin recycling.

Finally, our results suggest a potentially effective placementstrategy, particularly for prosocial and sustainability-relatedcommunications that target young peoplewho are heavy usersof the Internet and social media platforms. Images of cutebabies and animals are abundant online, and people activelyshare, tweet, and forward them. Because cuteness can en-hance high-BAS people’s prosociality, we advise policymakers to place prosocial messages featuring images andvideos of cute entities on social media platforms to attracthigh-BAS people and stimulate their engagement. Suchpractical applications of our findings, with some thought givento localization and creativity in implementation, may helpfurther efforts to attain the United Nations’ 2030 SustainableDevelopment Goals.

280 Getting Consumers to Recycle NOW!

ReferencesAlley, Thomas R. (1981), “Head Shape and the Perception of

Cuteness,” Developmental Psychology, 17 (5), 650–54.

Alley, Thomas R. (1983), “Age-Related Changes in Body Pro-portions, Body Size, and Perceived Cuteness,” Perceptual andMotor Skills, 56 (2), 615–22.

Barrick, Murray R., and Michael K. Mount (1991), “The Big FivePersonality Dimensions and Job Performance: AMeta-Analysis,”Personnel Psychology, 44 (1), 1–26.

Batson, C. Daniel (2009), “Empathy-Induced Altruistic Motiva-tion,” in Prosocial Motives, Emotions, and Behavior: The BetterAngels of Our Nature, M. Mikulincer and P.R. Shaver, eds.Washington, DC: American Psychological Association, 15–34.

Batson, C. Daniel, David A. Lishner, Jennifer Cook, and StaceySawyer (2005), “Similarity and Nurturance: Two PossibleSources of Empathy for Strangers,” Basic and Applied SocialPsychology, 27 (1), 15–25.

Carver, Charles S., and Teri L. White (1994), “Behavioral In-hibition, Behavioral Activation, and Affective Responses toImpending Reward and Punishment: The BIS/BAS Scales,”Journal of Personality and Social Psychology, 67 (2), 319–33.

Cian, Luca, Aradhna Krishna, and Ryan S. Elder (2015), “A Sign ofThings to Come: Behavioral Change Through Dynamic Ico-nography,” Journal of Consumer Research, 41 (6), 1426–46.

Cooper, M. Lynne, V. Bede Agocha, and Melanie S. Sheldon(2000), “A Motivational Perspective on Risky Behaviors: TheRole of Personality and Affect Regulatory Processes,” Journal ofPersonality, 68 (6), 1059–88.

Corral-Verdugo, Victor, Jose Mireles-Acosta, Cesar Tapia-Fonllem,and Blanca Fraijo-Sing (2011), “Happiness as Correlate ofSustainable Behavior: A Study of Pro-Ecological, Frugal, Eq-uitable and Altruistic Actions that Promote Subjective Well-being,” Human Ecology Review, 18 (2), 95–104.

Appendix A: Cute Versus Noncute Recycling Posters (Study 1)

A: Cute Recycling Poster

B: Noncute Recycling Poster

Recycling Saves Animals. Please Recycle.

Recycling Saves Animals. Please Recycle.

Appendix B: Charitable Donation Posters (Study 3)

A: Cute, High-BAS Poster

B: Cute, Low-BAS Poster

C: Noncute, High-BAS Poster

D: Noncute, Low-BAS Poster

Your donation can help animalsDonate NOW!

Your donation can help animalsPlease Donate

Your donation can help animalsPlease Donate

Your donation can help animalsDonate NOW!

Appendix C: Recycling Bins Used in Field Study 4

A: High-BAS, Active Executions

B: Low-BAS, Passive Executions

recycling is FUN!

recycle for me recycle please

recycle NOW!

Journal of Public Policy & Marketing 281

Darwin, Charles (1982 [1871]), The Descent of Man, and Selectionin Relation to Sex. Princeton, NJ: Princeton University Press.

Desjardins, Julie, John M. Zelenski, and Robert J. Coplan (2008),“An Investigation of Maternal Personality, Parenting Styles, andSubjective Well-Being,” Personality and Individual Differences,44 (3), 587–97.

De Waal, Frans B.M. (1996), Good Natured: The Origins of Rightand Wrong in Humans and Other Animals. Cambridge, MA:Harvard University Press.

De Waal, Frans B.M. (2008), “Putting the Altruism Back intoAltruism: The Evolution of Empathy,” Annual Review ofPsychology, 59, 279–300.

Doremus-Fitzwater, Tamara L., Elena I. Varlinskaya, and Linda P.Spear (2010), “Motivational Systems in Adolescence: PossibleImplications for Age Differences in Substance Abuse andOther Risk-Taking Behaviors,” Brain and Cognition, 72 (1),114–23.

Fazio, Russell H., Martha C. Powell, and Carol J. Williams (1989),“The Role of Attitude Accessibility in the Attitude-to-BehaviorProcess,” Journal of Consumer Research, 16 (3), 280–88.

Fisher, Robert J., Mark Vandenbosch, and Kersi D. Antia (2008),“An Empathy-Helping Perspective on Consumers’ Responses toFund-Raising Appeals,” Journal of Consumer Research, 35 (3),519–31.

Friestad, Marian, and Peter Wright (1994), “The PersuasionKnowledge Model: How People Cope with Persuasion Attempts,”Journal of Consumer Research, 21 (1), 1–31.

Frijda, Nico H. (1986), The Emotions. New York: CambridgeUniversity Press.

Geller, E. Scott (2002), “The Challenge of Increasing Pro-Environment Behavior,” in Handbook of Environmental Psy-chology, Robert B. Bechtel and Arza Churchman, eds. NewYork:John Wiley & Sons, 525–40.

Glocker, Melanie L., Daniel D. Langleben, Kosha Ruparela, JamesW. Lougheada, Jeffrey N. Valdeza, Mark D. Griffina, et al.(2009), “Baby Schema in Infant Faces Induces Cuteness Per-ception and Motivation for Caretaking in Adults,” Ethology,115 (3), 257–63.

Gray, Jeffrey A. (1970), “The Psychophysiological Basis of In-troversion and Extraversion,” Behaviour Research and Therapy,8 (3), 249–66.

Gray, Jeffrey A. (1981), “A Critique of Eysenck’s Theory of Per-sonality,” in A Model for Personality, J. Hans, ed. Berlin:Springer-Verlag, 246–76.

Griskevicius, Vladas, Stephanie M. Cantu, and Mark van Vugt(2012), “The Evolutionary Bases for Sustainable Behavior: Im-plications for Marketing, Policy, and Social Entrepreneurship,”Journal of Public Policy & Marketing, 31 (Spring), 115–28.

Hayes, Andrew F. (2013), Introduction to Mediation, Moderation,and Conditional Process Analysis. New York: Guilford Press.

Henriques, Jeffrey B., and Richard J. Davidson (2000), “DecreasedResponsiveness to Reward in Depression,” Cognition and Emo-tion, 14 (5), 711–24.

Hildebrandt, Katherine A., and Hiram E. Fitzgerald (1978), “Adults’Responses to Infants Varying in Perceived Cuteness,” Behav-ioural Processes, 3 (2), 159–72.

Iyer, Easwar S., and Rajiv K. Kashyap (2007), “Consumer Recy-cling: Role of Incentives, Information, and Social Class,” Journalof Consumer Behaviour, 6 (1), 32–47.

Jia, He (Michael), Gratiana Pol, and C.Whan Park (2017), “I’ll Keepthe Cuddly One: Effects of Visual Cuteness Versus Elegance onProduct Retention,” working paper.

Kollmuss, Anja, and Julian Agyeman (2002), “Mind the Gap: WhyDo People Act Environmentally andWhat Are the Barriers to Pro-Environmental Behavior?” Environmental Education Research,8 (3), 239–60.

Kringelbach,Morten L., Eloise A. Stark, Catherine Alexander, MarcH. Bornstein, and Alan Stein (2016), “On Cuteness: Unlockingthe Parental Brain and Beyond,” Trends in Cognitive Sciences,20 (7), 545–58.

Kronrod, Ann, Amir Grinstein, and Luc Wathieu (2012), “GoGreen! Should Environmental Messages Be So Assertive?”Journal of Marketing, 76 (1), 95–102.

Lishner, David A., C. Daniel Batson, and Elizabeth Huss (2011),“Tenderness and Sympathy: Distinct Empathic Emotions Elicitedby Different Forms of Need,” Personality and Social PsychologyBulletin, 37 (5), 614–25.

Lorenz, Konrad (1943), “Die Angeborenen Formen MoglicherErfahrung [The Innate Forms of Potential Experience],” Zeitschriftfur Tierpsychologie, 5, 233–519.

Ludwig, Timothy D., TimothyW. Gray, and Allison Rowell (1998),“Increasing Recycling in Academic Buildings: A SystematicReplication,” Journal of Applied Behavior Analysis, 31 (4),683–86.

McDougall, William (1908), An Introduction to Social Psychology.London: Methuen.

Moller, Arlen C., Richard M. Ryan, and Edward L. Deci(2006), “Self-Determination Theory and Public Policy: Im-proving the Quality of Consumer Decisions Without UsingCoercion,” Journal of Public Policy & Marketing, 25 (1),104–16.

Muris, Peter, CorMeesters, Elske de Kanter, and Petra E. Timmerman(2005), “Behavioural Inhibition and Behavioural Activation Sys-tem Scales for Children: Relationships with Eysenck’s Person-ality Traits and Psychopathological Symptoms,” Personalityand Individual Differences, 38 (4), 831–41.

Nenkov, Gergana Y., and Maura L. Scott (2014), “So Cute I CouldEat It Up: Priming Effects of Cute Products on Indulgent Con-sumption,” Journal of Consumer Research, 41 (2), 326–41.

Niezink, Lidewij W., Frans W. Siero, Pieternel Dijkstra, AbrahamP. Buunk, and Dick P. H. Barelds (2012), “Empathic Concern:Distinguishing Between Tenderness and Sympathy,” Motivationand Emotion, 36 (4), 544–49.

Parsons, Christine E., Katherine S. Young, NinaKumari, Alan Stein,andMorten L. Kringelbach (2011), “TheMotivational Salience ofInfant Faces Is Similar for Men and Women,” PLoS One, 6 (5):e20632.

Pieters, Rik G.M. (1991), “Changing Garbage Disposal Patternsof Consumers: Motivation, Ability, and Performance,” Journal ofPublic Policy & Marketing, 10 (Fall), 59–76.

Prothero, Andrea, Susan Dobscha, Jim Freund, William E.Kilbourne, Michael G. Luchs, Lucie K. Ozanne, et al. (2011),“Sustainable Consumption: Opportunities for Consumer Re-search and Public Policy,” Journal of Public Policy &Marketing,30 (Spring), 31–38.

Ryan, Richard M., and Edward L. Deci (2000), “Self-DeterminationTheory and the Facilitation of Intrinsic Motivation, Social De-velopment, and Well-Being,” American Psychologist, 55 (1),68–78.

282 Getting Consumers to Recycle NOW!

Sciandra, Michael R., Cait Lamberton, and Rebecca Walker Reczek(2017), “The Wisdom of Some: Do We Always Need HighConsensus to Shape Consumer Behavior?” Journal of PublicPolicy & Marketing, 36 (Spring), 15–35.

Shaver, Phillip R., Mario Mikulincer, and Moran Shemesh-Iron(2010), “A Behavioral Systems Perspective on Prosocial Behavior,”in Prosocial Motives, Emotions, and Behavior: The Better Angelsof Our Nature, Mario Mikulincer and Phillip R. Shave, eds.Washington, DC: American Psychological Association, 73–92.

Shrum, L.J., Tina M. Lowrey, and John A. McCarty (1994),“Recycling as a Marketing Problem: A Framework for StrategyDevelopment,” Psychology and Marketing, 11 (4), 393–416.

Simon, Joe J., Stephan Walther, Christian J. Fiebach, Hans-Christoph Friederich, Christoph Stippich, MatthiasWeisbrod, et al.(2010), “Neural Reward Processing Is Modulated by Approach-and Avoidance-Related Personality Traits,” NeuroImage, 49 (2),1868–74.

Small, Deborah A., and Nicole M. Verrochi (2009), “The Face ofNeed: Facial Emotion Expression on Charity Advertisements,”Journal of Marketing Research, 46 (6), 777–87.

Sober, Elliott, and David Sloan Wilson (1998), Unto Others: TheEvolution and Psychology of Unselfish Behavior. Cambridge,MA: Harvard University Press.

Tam, Kim-Pong, Sau-Lai Lee, and Melody M. Chao (2013),“Saving Mr. Nature: Anthropomorphism Enhances Connected-ness to and Protectiveness Toward Nature,” Journal of Experi-mental Social Psychology, 49 (3), 514–21.

Thompson, Edmund R. (2007), “Development and Validation ofan Internationally Reliable Short-Form of the Positive and Nega-tive Affect Schedule (PANAS),” Journal of Cross-Cultural Psy-chology, 38 (2), 227–42.

United Nations Economic and Social Council (2016), “ProgressTowards the Sustainable Development Goals: Report of theSecretary-General,” report (June 3), https://unstats.un.org/sdgs/files/report/2016/secretary-general-sdg-report-2016–EN.pdf.

Wadhwa, Monica, Baba Shiv, and Stephen M. Nowlis (2008), “ABite to Whet the Reward Appetite: The Influence of Samplingon Reward-Seeking Behaviors,” Journal of Marketing Research,45 (4), 403–13.

Journal of Public Policy & Marketing 283