LBS Research Online

C Cryder, S Botti and Y Simonyan
The Charity Beauty Premium: Satisfying Donors’ “Want” versus “Should” Desires
Article

This version is available in the LBS Research Online repository: http://lbsresearch.london.edu/
id/eprint/581/

Cryder, C, Botti, S and Simonyan, Y

(2017)

The Charity Beauty Premium: Satisfying Donors’ “Want” versus “Should” Desires.

Journal of Marketing Research, 54 (4). pp. 605-618. ISSN 0022-2437

DOI: https://doi.org/10.1509/jmr.14.0658

American Marketing Association
https://journals.sagepub.com/doi/10.1509/jmr.14.06...

Users may download and/or print one copy of any article(s) in LBS Research Online for purposes of
research and/or private study. Further distribution of the material, or use for any commercial gain, is
not permitted.

https://lbsresearch.london.edu/view/lbs_authors/926419.html
http://lbsresearch.london.edu/id/eprint/581/
http://lbsresearch.london.edu/id/eprint/581/
https://lbsresearch.london.edu/view/lbs_authors/926419.html
https://doi.org/10.1509/jmr.14.0658
https://journals.sagepub.com/doi/10.1509/jmr.14.0658


1 

The Charity Beauty Premium: Satisfying Donors' Want versus Should Desires 

 

CYNTHIA CRYDER 

SIMONA BOTTI 

YVETTA SIMONYAN* 

 

 

 

* Cynthia Cryder (cryder@wustl.edu) is an associate professor of marketing at Olin Business 

School, Washington University in St. Louis, Campus Box 1156, One Brookings Dr. St. Louis, 

MO 63130, 314-935-8114. Simona Botti (sbotti@london.edu) is an associate professor of 

marketing, London Business School, Regent’s Park, London NW1 4SA, United Kingdom, +44 

(0)20 7000 8646. Yvetta Simonyan (i.simonyan@bham.ac.uk) is a lecturer at Birmingham 

Business School, University of Birmingham, Edgbaston Park Road, Birmingham, B15 2TY, 

United Kingdom, +44 (0) 121 414 6701.  

 

We thank Charles Dorison, Rachel Gershon, Katherine Jaruzelski, Michael Lory, the Consumer 

Behavior lab at Washington University in St. Louis and the Behavioral Research Lab at London 

Business School for research assistance. We also thank Hengchen Dai and Theresa Kelly for 

useful insights and Stefano Puntoni and Claudia Townsend for providing valuable feedback on 

an earlier draft. Correspondence: Cynthia Cryder. 

 

  

tel:%2B44%20%280%2920%207000%208646
tel:%2B44%20%280%2920%207000%208646
tel:+44%20%20121%20414%206701


2 

The Charity Beauty Premium: Satisfying Donors’ Want versus Should Desires 

 

Despite widespread conviction that neediness is the most important criterion for charitable 

allocations, we observe a “charity beauty premium” in which donors often favor beautiful, but 

less needy charity recipients. We propose that donors hold simultaneous, yet incongruent 

preferences of wanting to support beautiful recipients (who tend to be judged as less needy) yet 

believing they should support needy recipients instead. We additionally posit that preferences for 

beautiful recipients are most likely to emerge when decisions are intuitive whereas preferences 

for needy recipients are most likely to emerge when decisions are deliberative. We test these 

propositions in several ways. First, when a beautiful recipient is introduced to basic choice sets, 

it becomes the most popular option and increases donor satisfaction. Second, heightening 

deliberation steers choices away from beautiful recipients and toward needier ones. Third, donors 

explicitly state that they “want” to give to beautiful recipients but “should” give to less beautiful, 

needier ones. Taken together, these findings reconcile and extend previous and sometimes 

conflicting results about beauty and generosity. 

 

Keywords: Prosocial Behavior, Altruism, Decision Making, Beauty Premium, Want versus 

Should Preferences, Intuitive versus Deliberative Decision Making 

 

  



3 

 

“Pandas are among a number of endangered animals that are sometimes classified, not 

flatteringly, as ‘charismatic megafauna,’ which attract money and attention out of proportion to 

their numbers. The British naturalist Christopher Packham once offered to ‘eat the last panda’ if 

doing so would free up funding for less photogenic species with better chances of survival 

outside zoos.” 

Owen 2013, 30 

 

Attractive donation targets, such as giant pandas, often receive the lion’s share of 

charitable contributions in the animal world, whereas needier but less photogenic creatures, such 

as the pygmy sloth, remain neglected (IUCN Red List 2014). This is one example of donation 

patterns that tend to overlook the world’s neediest recipients; for example, of the more than $350 

billion donated in the United States in 2014, only 4% went directly to international affairs, a 

fundraising category that encompasses the developing world where the most pressing human 

needs exist (Giving USA 2015; UNICEF 2014). 

In this research, we propose one reason why the neediest recipients may be overlooked: 

donors are swayed by recipient beauty. People naturally ascribe favorable qualities, and grant 

disproportionate benefits, to beautiful individuals in a phenomenon known as the “beauty 

premium” (Dion, Berscheid, and Walster 1972; Langlois et al. 2000). In the prosocial domain, 

however, a preference for beautiful recipients could lead to puzzling allocation choices. 

Perceptions of beauty and neediness are often negatively correlated (Fisher and Ma 2014; 

Langlois et al. 2000; Zaatari, Palestis, and Trivers 2009); therefore, a focus on beauty could 

result in neglect of the neediest individuals. 



4 

Previous research on recipient beauty and charitable giving shows seemingly 

contradictory patterns: some findings indicate that donors prefer beautiful recipients (Mims, 

Hartnett, and Nay 1975; West and Brown 1975), whereas other findings indicate that donors 

prefer less beautiful, needier recipients (Fisher and Ma 2014). In the current research, we attempt 

to understand the psychology behind both patterns. We posit that donors hold simultaneous, yet 

incongruent preferences including that they want to give to beautiful recipients but think they 

should give to needy recipients instead. In line with this notion, we find that when donors make 

decisions intuitively, their want preference for beautiful recipients emerges, whereas when they 

make decisions deliberatively, their should preference for needy recipients emerges. As we 

consider in a final discussion, the present findings allow insight into why the neediest recipients 

may often not receive priority in the real world. 

 

RECIPIENT NEEDINESS AS A CRITERION FOR GIVING 

 

Proponents of the nascent “effective altruism” movement, including the renowned 

philosopher Peter Singer, endorse the notion that donors should allocate their funds to the causes 

that prevent the most suffering (MacAskill 2015; Singer 2009, 2013). This often means donating 

to recipients in the developing world, where people’s standards of living are the lowest and the 

incidence of preventable death is the highest (GiveWell 2013; UNICEF 2014). 

Recipient neediness does not take precedence in the philosophical realm alone. A similar 

preference for the neediest recipients emerges from individuals’ explicitly stated donation 

criteria. We asked 228 Mechanical Turk participants (145 female, 83 male; MAge = 30) to 

generate an open-ended answer to the question “If multiple individuals needed help and you 

http://www.givewell.org/criteria


5 

were deciding which one to donate to, what would be the single most important thing you would 

consider when deciding which person to donate to?” Results, coded by research assistants blind 

to this paper’s hypotheses, showed that 56% of participants spontaneously answered that 

recipient neediness was the most important consideration. In comparison, the next most popular 

consideration was the potential impact of the donation, receiving 20% of responses; all other 

possible answers each received less than a 5% endorsement. 

Despite the common conviction that recipient neediness is the most important 

consideration for allocations, this criterion does not seem to receive priority in practice. A small 

fraction of donated dollars go to individuals in life-or-death situations (Giving USA 2015), 

though many such individuals exist (UNICEF 2014). 

Previous research sheds light on several important factors besides recipient neediness that 

influence donation decisions. These include donors’ sympathy for the recipient (Loewenstein and 

Small 2007; Wispé 1986), donors’ personal experiences (Ratner and Miller 2001; Small and 

Simonsohn 2008), donors’ need for emotional satisfaction (Andreoni 1990; Cialdini et al. 1987; 

Dunn, Aknin, and Norton 2008), and donors’ desire to maintain a positive image via both social 

and self-approval (Batson 1998; Cain, Dana, and Newman 2014). 

Yet another factor that we hypothesize sways donor’s choices away from the neediest 

recipients is recipients’ aesthetic appeal. This hypothesis stems from research about the “beauty 

premium,” a phenomenon in which beautiful individuals receive more favorable judgments and 

benefits than their less beautiful counterparts. 

 



6 

THE BEAUTY PREMIUM 

 

In a seminal study, Dion, Berscheid, and Walster (1972) showed that people attribute 

more positive qualities to attractive than to unattractive others, and more negative qualities to 

unattractive than to attractive others. This “beauty premium” is confirmed by numerous findings 

demonstrating that facial attractiveness positively influences judgments by both familiar and 

unfamiliar perceivers in multiple domains and across cultures (Langlois et al. 2000; Miller 

1970). More attractive children and adults are considered to be healthier and fitter, to have 

greater social appeal, to have higher academic competence, and to be more confident and better 

adjusted than their less attractive counterparts (Landy and Sigall 1974; Langlois et al. 2000; 

Mobius and Rosenblat 2006). More attractive individuals also are more successful in their 

attempts at persuasion (Landry et al. 2005) and receive sizable wage premiums compared to less 

attractive individuals (Hamermesh and Biddle 1994). 

There are two dominant explanations regarding why the beauty premium occurs. One 

explanation is rooted in an evolutionary perspective that centers on reproductive selection and 

fitness. This evolutionary account proposes that a preference for beauty arises because beauty is 

a signal of traits such as health, athleticism, and intelligence, which are important for fertility 

and/or survival (Buss and Schmitt 1993; Langlois et al. 2000; Mobius and Rosenblatt 2006). 

A second explanation proposes a more instinctive connection between beauty and 

favorable judgments. This account suggests that people simply have a taste, or predilection, for 

beauty even when fitness or beliefs about fitness are unrelated (Becker 1957; Hamermesh and 

Biddle 1994; Mulford et al. 1998; Newman and Bloom 2012). Such a taste-based preference for 

beauty has been posited to be automatic and affective at its core (Reimann et al. 2010) and to 



7 

arise as a by-product of information processing within the human brain (Rhodes 2006). Factors 

that are correlated with attractiveness, such as symmetry, are desirable not just for human faces 

but for many other stimuli, such as birds and wristwatches (Halberstadt and Rhodes 2000; 

Kubovy 2000). In the consumer domain, this innate draw toward beauty is manifested by 

favorable responses to aesthetic designs, including a desire to own an attractive product (Norman 

2004) and to pay more for it (Bloch, Brunel, and Arnold 2003), even in categories such as 

finance for which aesthetics are deemed to be irrelevant (Townsend and Shu 2010). 

An appreciation for beauty not only increases the likelihood that people choose more 

attractive options but also results in higher satisfaction among people who do so. Visually 

beautiful stimuli activate reward centers in the brain (Aharon et al. 2001), elicit intense positive 

emotional responses (Leder et al. 2004; Reimann et al. 2010), and generate pleasurable 

subjective experiences (Reber, Schwarz, and Winkielman 2004; Townsend and Sood 2012). 

 

WANT VERSUS SHOULD PREFERENCES IN CHARITABLE GIVING 

 

In the domain of charitable giving, an instinctive and gratifying preference for beauty 

could lead to puzzling allocation choices. Physically attractive people are perceived to be more 

competent, popular, and successful than their less attractive counterparts (Langlois et al. 2000) 

and, accordingly, should be perceived as less in need of help. Indeed, Zaatari, Palestis, and 

Trivers (2009) find that proposers in an ultimatum game rate the most attractive respondents as 

the least needy. Similarly, Fisher and Ma (2014) find that more attractive donation recipients are 

perceived to be less needy than less attractive recipients. Our own data reveal a similar pattern; 

in a correlational study, Mechanical Turk participants (N = 224) rated 1 of 20 charity recipient 



8 

photos from a fundraising website on dimensions of attractiveness and neediness. The set of 

photos included pictures of 13 women and 7 men across a variety of adult ages and ethnicities. 

We observe a significant negative correlation, r = −21, p < .01, between ratings of attractiveness 

and neediness. 

An instinctive draw to beautiful recipients might therefore be at odds with donors’ 

explicitly held belief that individuals in the greatest need should receive altruistic priority. In this 

research, we propose that this incongruence exists because donors simultaneously hold two 

distinct preferences: a want preference for beautiful recipients and a should preference for needy 

recipients (Bazerman, Tenbrunsel, and Wade-Benzoni 1998; Bitterly, Dai, and Milkman 2014; 

Milkman, Rogers, and Bazerman 2008, 2009). 

Previous research conceptualizes want preferences as those that are affect-rich and, 

sometimes, linked with desire (Bazerman, Tenbrunsel, and Wade-Benzoni 1998; Loewenstein 

1996) and should preferences as those that are reason-based, logical, and more easily justified 

(Bazerman, Tenbrunsel, and Wade-Benzoni 1998).
1
 For example, when choosing movies, 

consumers may hold a want preference to watch a comedy but a should preference to watch a 

documentary (Milkman, Rogers, and Bazerman 2009). In the charity domain, we hypothesize 

that if humans have an automatic and gratifying preference for beautiful stimuli, they may 

intuitively want to give to beautiful charity recipients, even if they deliberately believe that they 

should give to needy recipients. Beautiful recipients offer intuitive appeal and immediate 

satisfaction, whereas needy recipients fit with a reasoned priority to help the most desperate 

individuals. 

 We further propose that donors’ want preferences for beautiful recipients are most likely 

to emerge when they choose intuitively, whereas donors’ should preferences for needy recipients 



9 

are most likely to emerge when they choose deliberatively. Research from a breadth of 

psychology disciplines recognizes distinctions in cognitive functioning between “System 1” 

processing, or intuitive processing that depends on automatic associations, and “System 2” 

processing, or deliberative processing that depends on logical reasoning (Stanovich and West 

2000). Processes under System 1 operate quickly, rely on seemingly effortless associations to 

inform judgments, and tend to favor affect-rich options. Processes under System 2, by contrast, 

operate slowly and rely on effortful cognition to reach reasoned conclusions (Chaiken and Trope 

1999; Kahneman and Frederick 2002; Milkman, Chugh, and Bazerman 2009). 

Because preferences for beautiful recipients may arise from automatic, instinctive 

preferences, we predict that they will be most likely to emerge when donors process information 

intuitively via System 1. In contrast, because we posit that preferences for needy recipients arise 

on the basis of reason, we predict that they are most likely to emerge when donors process 

information deliberatively via System 2. These predictions are consistent with previous research 

implicating a dual-system model of information processing in want versus should preferences 

(Bitterly et al. 2014; Milkman, Rogers, and Bazerman 2008). The intuitive System 1 is reward 

seeking and often oriented toward what feels good, whereas the deliberative System 2 is rule 

based and oriented toward what is appropriate and sensible (Epstein 1994; Shiv and Fedorikhin 

1999). 

Understanding a distinction between donors’ intuitively versus deliberately derived 

preferences could shed light on discrepancies that exist in the altruism literature. Some past 

research finds, for example, that attractive people are more likely to receive help than 

unattractive people (Mims, Hartnett, and Nay 1975; West and Brown 1975). In contrast, other 

research finds that unattractive people are more likely to receive help than attractive people 



10 

(Fisher and Ma 2014). To reconcile these inconsistent findings, we propose that donors hold 

simultaneous yet contradictory preferences that are cued by different information-processing 

modes: when donors decide intuitively, they are more likely to select beautiful recipients in line 

with their want preferences; when donors decide deliberatively, they are more likely to select 

needy recipients in line with their should preferences. 

We test these propositions across eight studies. Studies 1A through 1C explore what 

happens to donors’ choices and satisfaction when beautiful options are included in basic choice 

sets, and document a preference for beautiful options. Studies 2 through 4 test the proposition 

that increasing deliberation shifts preferences away from beautiful recipients and toward needy 

recipients. Study 5 examines how very high levels of perceived need, and associated high 

empathy, weaken the charity beauty premium. Finally, Study 6 demonstrates that donors’ 

preferences for beautiful versus needy recipients specifically represent want versus should 

preferences. Throughout, we attempt to understand the psychology underlying donors’ 

preferences for beautiful versus needy recipients and, in the process, shed new light on previous 

findings in this domain. 

 

STUDY 1A 

 

Study 1A is a field study that solicited donations for an animal conservation center in the 

UK. The conservation center sells animal adoption packages that include receipt of a paper leaf 

inscribed with the donor’s name that is placed on the center’s adoption tree. While all species on 

the conservation center’s website need funds, each species’ conservation status differs because 



11 

some animals are more critically endangered than others. We test how judgments of beauty and 

endangerment predict donations to different animal species. 

Pretest 

One hundred seven university graduate and undergraduate students rated images of eight 

animals from the conservation center’s adoption webpage on beauty and perceived 

endangerment (1–10 scales; Please see Web Appendix A for photos); order of image and 

question were counterbalanced. Table 1 shows rating results as well as each animal’s actual 

endangerment status as determined by a prominent conservation organization (IUCN 2015). 

***** Insert Table 1 about here ***** 

Main Study 

Simulating the conservation center’s adoption packages, a poster featuring a tree was 

placed in the foyer of a busy university building in the UK. The poster included the message 

“Support [conservation center] today. Join our Support Tree! [conservation center] contributes to 

animal conservation.” Students, staff, and visitors passing by were asked if they would like to 

donate to support animal conservation. Those who were interested viewed a booklet with the 

animal images from the pretest and chose an animal to support. Donors could then write their 

name on a paper leaf and append the leaf to the poster. The suggested donation amount was 20 

pence (approximately $.30) per paper leaf, but donors were free to donate more both to the same 

animal and to different ones. Donations were collected for five consecutive days. 

 

Results 

Pretest. The giraffe and the zebra were rated the most beautiful and least endangered 

animals, whereas the penguin and orangutan were rated the least beautiful and most severely 



12 

endangered; these endangerment ratings were in line with actual endangerment status as 

determined by the IUCN Red List (2015; see Table 1). 

Main study. The giraffe and the zebra, rated the most beautiful animals in the pretest, 

received a significantly greater percentage of sponsorships than the two least beautiful animals, 

the penguin and orangutan (which also were correctly perceived as being more severely 

endangered), 32% versus 17% respectively, χ
2
 = 4.79, p < .03.  A similar test comparing 

sponsorships for the four most beautiful versus least beautiful animals showed that the former 

were selected significantly more often than the latter (64% versus 36% respectively, χ
2
 = 7.04, p 

< .01; see Figure 1). The same test conducted with the four most versus least endangered species, 

determined either by actual endangerment status or by participants’ ratings of perceived 

endangerment, showed no such difference (actual endangerment: High = 47% vs. Low = 53%, χ
2
 

= .38, p = .54, perceived endangerment: High = 48% vs. Low = 52%, χ
2
 =.17, p = .68). 

***** Insert Figure 1 about here ***** 

Discussion 

 Study 1A provides initial field evidence that beautiful recipients can be preferred even 

when they do not have the highest need. Study 1B explores this pattern further, incorporating an 

experimental test. 

 

STUDY 1B 

 

 Study 1B experimentally tested how adding a beautiful recipient to a basic choice set 

changes donors’ preferences and satisfaction. All participants viewed images of four children 



13 

who had received successful cleft palate surgery via the charity Smile Train, and chose which 

child they would sponsor. 

Pretest 

Four hundred eighty-one Mechanical Turk participants (229 male, 252 female; MAge = 

36) rated four children on attractiveness (“cuteness”) and neediness. One child (Angelica) was 

rated as significantly cuter than any of the other four children (MAngelica = 4.08, MVera = 3.03, 

MMaria = 3.00, MSilva = 2.86, MYaoLin = 2.83; ps < .0005). Angelica also was rated as significantly 

less needy than the other four (MAngelica = 2.57, MVera = 3.16, MMaria = 2.95, MSilva = 3.06, MYaoLin 

= 3.06; ps < .0005); one child, Vera, also consistently was rated as needier than the remaining 

three children (vs. Maria, p < .0005; vs. Silva, p < .05; vs. Yao Lin, p < .06).
2
 

Main Study 

Method. Three hundred sixty Mechanical Turk participants (173 female, 187 male, MAge 

= 32) participated in exchange for a small payment.
3
 Participants were presented, in random 

order, with the pretested pictures of four children who had been helped by Smile Train and read 

that these children needed financial support to move on with their lives after cleft palate surgery. 

The children were described as approximately the same age (1–2 years) and from the same 

region of the world (Central and South America). 

 The experiment included Control and Beauty conditions. In the Control condition, 

participants viewed the four children who, according to the pretest, were similar in attractiveness 

(Vera, Maria, Silva, and Yao Lin). In the Beauty condition, the child rated as significantly more 

beautiful and less needy than the others, Angelica, was substituted for Yao Lin; to maintain 

experimental control, the name “Angelica” was used in both conditions. 



14 

Participants then selected a child as if they were actually sponsoring her. Next, they 

answered follow-up questions about how happy, pleased, and content they felt as a result of their 

choice; these items were combined into a single “satisfaction” score (α = .89). 

Results. An omnibus chi-square test revealed a significant difference between conditions 

in sponsorship choices, χ
2
(4, N = 360) = 33.0, p < .001 (see Figure 2). Participants in the Control 

condition, who did not consider a clearly beautiful option, showed a preference for Vera, the 

neediest child, over all others (Vera, 45%; Maria, 22%, p = .001; Yao Lin, 20%; p < .001; Silva, 

14%, p < .001; all p-values relative to Vera’s choice share). 

***** Insert Figure 2 about here ***** 

 Results differed in the Beauty condition. Participants in the Beauty condition, who 

instead of Yao Lin considered Angelica, the most beautiful and least needy child according to the 

pre-test, chose Angelica more often than any other child (Angelica = 48%, Vera = 30%, p = .01; 

Maria = 12%, p < .001; Silva = 10%, p < .0001; all p-values relative to Angelica’s choice share). 

In this Beauty condition, compared to Control, participants also were less likely to choose Vera, 

the child perceived as the neediest (30% vs. 45%, z(359) = 2.92, p < .01). 

 In addition to selecting Angelica more often than any other option, participants in the 

Beauty condition reported greater satisfaction with their choice than did those in the Control 

condition, MBeauty = 3.60, MControl = 3.27, t(358) = 2.40, p < .02. 

Discussion 

Results from Study 1B are consistent with several past findings about beauty and 

generosity (Mims, Hartnett, and Nay 1975; West and Brown 1975): when a beautiful recipient is 

present, donors are more generous toward that beautiful recipient. Further, we observe that not 

only do donors choose a beautiful recipient when one is present, but they also are more satisfied 



15 

with their choice, supporting the notion that beauty is gratifying. In contrast, when no beautiful 

option is present, donors gravitate toward the neediest recipient, in line with their explicit 

prioritization of neediness. Study 1C continues to explore this pattern with novel stimuli and 

decision types.
4
 

 

STUDY 1C 

 

 Study 1C tests donors’ reactions to recipients who are more beautiful versus less 

beautiful using a different approach from Study 1B. Donors in Study 1C viewed a single 

recipient instead of several, considered donating to an adult instead of to a child, considered a 

photo of the same person that had been edited via photo software to be more versus less 

beautiful, and made real donations. The new study design also allowed follow-up questions 

measuring perceptions of facial expression so that we could control for this factor. 

Pretest 

Pretest participants recruited from Mechanical Turk (n = 285; 172 female, 113 male, one 

gender unknown; MAge = 33) rated a photo of either a man (“Evan”) or a woman (“Andrea”) 

from a fundraising website. In the Beauty condition, participants saw the original, attractive, 

photo of one of these two individuals. In the Control condition, the photos were edited, 

introducing asymmetries to make them less attractive. 

Participants were randomly assigned to view one of the four photos. Each participant 

rated the photo on the same dimensions (beauty/handsomeness and neediness) used in Study 1B, 

using 5-point scales. The two Beauty condition photos were rated more “beautiful or handsome” 

(Andrea: MBeauty = 3.85, MControl = 2.77, t(134) = 7.58, p < .0005; Evan: MBeauty = 3.29, MControl = 



16 

2.30, t(147) = 6.20, p < .0005) and less needy (Andrea: MBeauty = 1.82, MControl = 2.35, t(135) = 

3.75, p < .0005; Evan: MBeauty = 1.82, MControl = 2.15, t(147) = 2.30, p < .03) than the two Control 

condition photos. 

Main Study 

Method. Two hundred thirty-eight adults (134 female, 104 male; MAge = 33) participated 

online via Mechanical Turk. After reading initial instructions, participants learned that they 

would receive a 25-cent bonus that they could donate to charity as part of the study. 

The study included two between-subjects experimental conditions (Control, Beauty) with 

recipient gender randomized within each condition. All participants read that they could donate 

to a recovered bone cancer patient who was seeking financial help. In the Beauty condition, 

participants saw the original photo of either an attractive man or an attractive woman from a 

fundraising website. In the Control condition, participants saw an edited, less attractive version 

of the same man or woman. Donors decided whether to donate their $.25 bonus to the recipient 

or not. 

Following procedures from Study 1B, participants then rated how happy, pleased, and 

content they felt as a result of considering the donation request (“satisfaction” score α = .90) as 

well as how happy, sad, needy, and beautiful the recipient looked. 

Results. Participants were more likely to donate in the Beauty condition (59%) than in the 

Control condition (45%, χ
2
(1, N = 238) = 4.45, p < .05). This effect did not differ between 

female and male photos (p > .9); moving forward we collapse results across recipient gender. 

The beauty effect remained significant when controlling for the recipient’s judged 

happiness, sadness, neediness, or all three measures at once (in simultaneous logistic regressions, 

ps < .05). When controlling for perceived beauty of the recipient, however, the effect of beauty 



17 

became non-significant (p > .2); this result is consistent with the notion that perceived beauty, 

but not facial expression, drives donation differences between the Beauty and Control conditions.  

A simultaneous mediation analysis confirmed that perceived beauty uniquely accounted 

for the influence of beauty on donations. We conducted a bootstrapping mediation analysis with 

5,000 samples using Hayes’s (2013) PROCESS Macro (Model 4), which included measures of 

perceived happiness, sadness, neediness, and beauty. Only perceived beauty significantly 

mediated the relationship between the Beauty condition and donation choice, 95% CI = .15–.63. 

In this model, the direct effect was no longer significant (p = .09), and no other potential 

mediators were significant (ps > .2). We note that this particular mediation analysis does not 

suggest a psychological mediator for the charity beauty premium (Zhao, Lynch, and Chen 2010) 

but instead serves as a rigorous manipulation check. 

Participants also reported marginally significantly greater satisfaction in the Beauty 

condition than in the Control condition (MBeauty = 3.27, MControl = 2.98, t(235) = 1.93, p < .06). A 

simultaneous mediation model analogous to the one described above confirmed that perceived 

beauty uniquely accounted for the influence of the Beauty condition on satisfaction, 95% CI = 

.12–.40. In this analysis, the direct effect was no longer significant (p = .91), and no other 

potential mediators, including facial expression, were significant. 

Discussion 

In Study 1A, donors were more likely to sponsor beautiful animals, even when beautiful 

animals were not the most endangered. In Study 1B, donors selected the neediest child as long as 

a beautiful (but less needy) recipient was not an option; once a beautiful recipient was an option, 

she became the most popular recipient and increased donor satisfaction (see also Web Appendix 

B). In Study 1C, this preference for a beautiful recipient, and the subsequent greater donation 



18 

satisfaction, persisted when participants made real donations to a single adult recipient, and it 

was not explained by recipients’ facial expression. 

  Although a preference for beautiful recipients is consistent with some prior evidence 

about beauty and giving (Mims, Hartnett, and Nay 1975; West and Brown 1975), it is not 

consistent with donors’ explicit statements (reported in our initial survey) that neediness is the 

most important criterion for charitable allocations. It is also not consistent with recent evidence 

that finds the opposite pattern, namely, that less attractive recipients receive more donations 

(Fisher and Ma 2014, Study 4). 

A common feature of Studies 1A through 1C is that participants’ decisions were 

relatively intuitive. Studies 1A and 1B involved a sponsorship selection from a small set of 

pictures; Study 1C included a simple yes/no choice of whether to donate a bonus payment. 

Previous research documenting a beauty premium in prosocial domains also relied on choices 

that are likely intuitive, such as whether agreeing or not to help an attractive, versus unattractive, 

experimenter with an additional task (Mims, Hartnett, and Nay 1975). 

In this research, we propose that intuitive versus deliberative decision modes can lead to 

diverging recipient preferences. More specifically, we hypothesize that intuitive processing leads 

to preferences for beautiful recipients whereas deliberative, cognitively effortful processing leads 

to preferences for needy recipients. Studies 2 through 4 directly test this hypothesis. 

 

STUDY 2 

 

Study 2 tests whether intuitive versus deliberative decision modes influence the charity 

beauty premium by directly asking participants to let either deliberation or intuition guide a 



19 

charity recipient choice. We predict that when deciding intuitively, participants will prefer the 

most beautiful recipient, but when deciding deliberatively, they will not. 

Method. Two hundred forty Mechanical Turk participants (139 female, 101 male; MAge = 

33) participated online in exchange for a small payment. Participants first read the scenario from 

Study 1B. In the Intuitive condition, participants also read, “Please make a quick choice, letting 

intuition guide which child you would sponsor. Be sure to rely on your instincts.” In the 

Deliberative condition, participants read, “Please think carefully about your choice, thoroughly 

considering which child you would sponsor. Be sure to rely on logical deliberation.” All 

participants then viewed the photos from the Beauty condition from Study 1B (excluding any 

additional text such as names). Participants first chose a child to sponsor and then rated to what 

extent their sponsorship choice was based on intuition versus deliberation (1–5 rating scale with 

1 = “entirely on intuition” and 5 = “entirely on deliberation”). We also measured participants’ 

response time from when they loaded the survey screen with recipient photos until they made the 

final decision click on the same screen. 

Results 

Manipulation checks. Participants in the Deliberative condition indicated that their choice 

was based more on deliberation than did participants in the Intuitive condition (MDeliberative = 

3.06, MIntuitive = 2.05, t(238) = 7.62, p < .0005). Response time in seconds also was higher in the 

Deliberative condition than in the Intuitive condition (MedDeliberative = 20.76, MedIntuitive = 11.59, 

Mann-Whitney U = 3,600, p < .0005). 

Main results. An omnibus chi-square test revealed a marginally significant difference 

between conditions in sponsorship choices, χ
2
(3, N = 240) = 7.36, p = .06 (see Figure 3). 

Participants in the Intuitive condition showed a preference for Angelica, the most attractive and 



20 

least needy of the four children according to the Study 1B pre-test (Angelica, 49%; Vera, 23%, p 

< .001; Maria, 12%, p < .0001; Silva, 16%, p < .0001; all p-values relative to Angelica’s choice 

share). 

***** Insert Figure 3 about here ***** 

 

In the Deliberative condition, Angelica was no longer a favorite (Angelica, 33%; Vera, 

28%, p = .55; Maria, 18%, p < .05; Silva, 22%, p = .13; all p-values relative to Angelica’s choice 

share). In the Deliberative condition, compared to the Intuitive condition, participants were less 

likely to choose Angelica, the most beautiful and least needy recipient, for their allocation 

decision (33% vs. 49%, z(239) = 2.63, p < .01); consequently, they were more likely to choose 

needier recipients. 

Discussion 

In Study 2, when donors decided intuitively, they were most likely to choose a beautiful 

recipient; however, an explicit request to choose deliberatively eliminated this preference. In the 

Deliberative condition, preferences shifted toward the less beautiful, but needier, recipients 

instead, providing initial evidence that the charity beauty premium emerges from intuitive, but 

not deliberative, decisions. 

 

STUDY 3 

 

In Study 3, we use an alternative method to manipulate deliberation. We draw upon 

findings that decision makers rely more heavily on logic and deliberation when they make 

choices for others compared to when they make choices for themselves. For example, consumers 



21 

are less loss averse when choosing for others than when choosing for themselves (Polman 2012), 

and they seek greater variety when choosing for others because it is a more objectively 

compelling strategy (Choi et al. 2006; Ratner and Kahn 2002). 

In Study 3, we manipulate whether donors make a sponsorship choice for themselves or 

on behalf of someone else while serving as an adviser. We predict that asking donors to serve as 

advisers, and thus choosing on behalf of someone else, will encourage them to override intuitive 

preferences for beauty in favor of deliberative preferences for neediness. 

Method 

Two hundred sixty-four participants from Mechanical Turk (105 female, 159 male; MAge 

= 32) participated online in exchange for a small payment. Participants read the scenario from 

Study 1B. In the Intuitive condition, participants chose one child to support as a sponsor. In the 

Deliberative condition, participants served as an adviser, choosing “which child will be 

sponsored by an anonymous donor.” All participants viewed the same pictures and short 

recipient descriptions from the Beauty condition from Study 1B and chose a child, in the role 

either of sponsor (Intuitive condition) or adviser (Deliberative condition). 

Results 

An omnibus chi-square test revealed a difference between conditions in sponsorship 

choices χ
2
(4, N = 264) = 8.99, p = .06 (see Figure 4). Participants in the Intuitive condition 

sponsored Angelica, the most attractive and least needy of the four children, most often 

(Angelica, 42%; Vera, 25%, p < .03; Mariana, 20%, p = .001; Silva, 12%, p < .0001; all p-values 

relative to Angelica’s choice share). 

***** Insert Figure 4 about here ***** 



22 

In the Deliberative condition, in which participants served as an adviser, Angelica was no 

longer the favorite (Angelica, 30%; Vera, 39%, p = .24; Mariana, 18%, p < .05; Silva, 13%, p < 

.01; all p-values relative to Angelica’s choice share). In the Deliberative condition, compared to 

the Intuitive condition, participants were less likely to choose Angelica for their allocation 

decision (30% vs. 42%, z(263) = 2.04, p < .05), and they were also more likely to choose Vera, 

the neediest child (39% vs. 25%, z(263) = 2.45, p < .05). 

In a posttest, we verified that making sponsorship choices on behalf of someone else 

prompts deliberation compared to making sponsorship choices for oneself. Previous research has 

shown that accountability to others can heighten expectations of justifying one’s decision (Lerner 

and Tetlock 1999) and induce controlled information processing (Bazerman et al. 1998). Thus, 

we asked 113 Mechanical Turk participants to read either the Intuitive or the Deliberative 

scenario and to rate the extent to which they would expect to need to justify the decision to 

others. Participants who read the Deliberative condition scenario were more likely to anticipate 

justifying their choice than were participants who read the Intuitive condition scenario 

(MDeliberative = 2.66, MIntuitive = 1.98, t(111) = 2.97, p < .01). 

Discussion 

In Study 3, donors who were placed in a more deliberative state by serving as advisers 

were less likely to choose Angelica, the most attractive but least needy recipient. These 

participants also were more likely to act in line with an explicit prioritization of neediness by 

choosing Vera, the child who was rated as needier than the others. 

We note that the current result is incompatible with the notion that donors choose 

beautiful recipients because beautiful individuals are more worthy or more promising. If 

expectations of beautiful recipients’ success drive the charity beauty premium, then we should 



23 

observe it both when participants make sponsorship choices for themselves and when they make 

sponsorship recommendations to others. However, we observe that when participants serve as 

advisers, the charity beauty premium weakens and, in fact, reverses; this pattern is consistent 

with an account that donors prefer beautiful recipients because of an intuitive preference for 

aesthetic appeal and not because these recipients are justified to be more likely to succeed.  

Study 4 tests an additional way of increasing deliberation: asking participants to rate 

recipients on several dimensions before making sponsorship choices. 

 

STUDY 4 

 

 Previous research documenting that donors allocate more funds to less attractive, needier 

recipients than to more attractive, less needy recipients finds this pattern after participants first 

explicitly rate charity recipients on several dimensions including neediness, attractiveness, and 

empathy (Fisher and Ma 2014, Study 4). It is possible that such a rating task encourages 

participants to thoroughly consider recipients, prompting deliberative decision making. In Study 

4, we directly manipulate whether such a rating task steers donors toward a deliberative choice of 

needy recipients and away from an intuitive choice of beautiful recipients. 

Study 4 includes an Intuitive condition identical to that in Study 3 and also includes a 

new Deliberative condition in which, before making sponsorship choices, participants explicitly 

rate each recipient on dimensions identical to those measured by Fisher and Ma (2014): 

attractiveness, emotional expression, social competence, perceived need, and empathy. 



24 

Method 

 Three hundred six Mechanical Turk participants (189 female, 117 male; MAge = 33) 

completed the study in exchange for a small payment. The experiment included two conditions: 

the Intuitive condition from Study 3 and a new Deliberative condition in which participants rated 

each child on attractiveness (three items, 5-points scales), emotional expression (four items, 5-

point scales), social competence (seven items, 5-point scales), perceived need (four items, 5-

points scales), and empathy (five items, 7-point scales), all following procedures used by Fisher 

and Ma (2014), before making their sponsorship choice. 

Results 

Rating results. Similar to the pretest described before Study 1B, participants in the 

Deliberative condition, who rated each child on multiple dimensions before making the 

allocation choice, rated Angelica as more attractive (MAngelica = 4.20, MVera = 3.42, MMaria = 3.43, 

MSilva = 3.39, ps < .0005) and less needy (MAngelica = 2.39, MVera = 3.28, MMaria = 2.80, MSilva = 

3.10, ps < .0005) than any of the other three children; they also judged Vera to be needier than 

any of the other three children, ps < .0005). In addition, participants reported either directionally 

or significantly less empathy for Angelica (MAngelica = 5.23, MVera = 5.52, p < .0005; MMaria = 

5.32, p = .19, MSilva = 5.38, p = .05) and significantly more empathy for Vera compared to the 

other children (ps < .01). These empathy ratings are consistent with the central finding from 

Fisher and Ma (2014, Studies 1–3) that participants express more empathy for a less beautiful 

than for a more beautiful charity recipient.
5
 

Main results. An omnibus chi-square test showed a difference between conditions when 

choosing which child to support, χ
2
(3, N = 304) = 12.30, p < .01 (see Figure 5). 

***** Insert Figure 5 about here ***** 



25 

In between-subjects comparisons, participants in the Deliberative condition were less 

likely to choose Angelica, the most beautiful recipient, than were participants in the Intuitive 

condition (21% vs. 39%, z(302) = 3.42, p < .001), and more likely to choose Vera, the neediest 

recipient (42% vs. 30%, z(302) = 2.21, p < .03). Within the Deliberative condition, Vera was 

selected the most frequently (Vera, 42%; Angelica, 21%, p < .001; Mariana, 20%, p < .0005; 

Silva, 18%, p <.0005; all p-values relative to Vera’s choice share), whereas within the Intuitive 

condition, Angelica was selected the most frequently (Angelica, 39%; Vera, 30%, p = .23; Maria, 

18%, p < .0005; Silva, 13%, p < .0005; all p-values relative to Angelica’s choice share). 

Discussion 

The Deliberative condition from Study 4 demonstrates that a rating task, which prompts a 

thorough consideration of each recipient, pushes donors to choose needier, less beautiful, 

recipients. This study also sheds light on previous results by showing that such a rating task can 

trigger a reversal of the charity beauty premium (Fisher and Ma 2014; Study 4). Although donors 

gravitate toward beautiful recipients when choices are intuitive, increasing deliberation by asking 

participants to rate recipients on several attributes (including both beauty and neediness) steers 

them to allocate in accordance with their explicit priority of neediness. Across studies 2 through 

4, donors were more likely to choose beautiful recipients when deciding intuitively and needy 

recipients when deciding deliberatively. 

Rating results from Study 4 also show that participants report greater empathy for a less 

beautiful than for a more beautiful recipient. This result is consistent with the central finding 

from Fisher and Ma (2014, Studies 1–3). Despite greater empathy for less beautiful recipients, 

however, we find that donors choose these recipients only when the choice is deliberative, but 



26 

not when the choice is intuitive. In the next study, we further explore the complex role of 

empathy in the relationship between donors’ preferences for beautiful versus needy recipients. 

 

STUDY 5 

 

Study 5 explores the role of empathy in the charity beauty premium. In Study 4, we see 

that empathy only predicts giving when participants process information in a deliberative, 

System 2 mode. In many respects, this result is puzzling. Empathy and sympathy are often 

presumed to be affective responses and, therefore, associated with System 1 intuitive processing 

(Batson 1998; Loewenstein and Small 2007). Yet, empathy and sympathy are unique affective 

influences, because they also often correlate with deliberative judgments of neediness (Batson et 

al. 2005; Fisher and Ma 2014; Study 4 ratings from the current paper). Moreover, past research 

proposes a deliberative component of such sympathetic emotions (Hatfield, Cacioppo, and 

Rapson 1994). According to this account, empathy entails taking another person’s perspective, a 

process involving reflection and effort (Wispé 1986) that relies on deliberative cognition. This 

characterization of empathy is consistent with our findings thus far that donors are more likely to 

favor empathetic recipients when processing deliberatively. 

Nevertheless, empathy and sympathy have potential to activate intense affective 

responses as well. Such intense affective responses may actually oppose, and even overwhelm, 

donors’ intuitive want preference for beauty. We posit that very high levels of need may trigger 

strong emotional empathy toward recipients (Loewenstein and Small 2007; Small and Cryder 

2016), which in turn can override beauty preferences due to a competing System 1 influence. We 

test this hypothesis in Study 5. 



27 

Method 

Seven hundred seventy Mechanical Turk participants (438 female, 332 male; MAge = 34) 

participated in exchange for a small payment. Similar to Study 1C, participants learned that they 

would receive a 20-cent bonus that they could donate to charity during the study. 

The study included a 2 (Photo: Control, Beautiful) × 2 (Empathy: Control-Empathy, 

High-Empathy) between-subjects experimental design. The Beautiful versus Control photo 

manipulation was identical to that in Study 1C (randomizing female and male photos within 

condition). In the Control-Empathy condition, participants read the recipient description from 

Study 1C portraying a recipient who has recovered from bone cancer and is seeking financial 

help to move forward in life. In the High-Empathy condition, the recipient was described as even 

higher in need, struggling to recover from bone cancer and no longer able to work and support a 

family (please see Web Appendix C for full wording). 

As in Study 1C, the main dependent variable was whether participants decided to donate 

their participation bonus to the recipient or not, after which they responded to five items from 

Fisher and Ma (2014; see also Study 4 of the current paper) designed to measure empathy: the 

extent to which participants felt sympathetic, compassionate, softhearted, warm, and moved (1–5 

point scales, items combined for single empathy score, α = .94). 

Results 

Manipulation check. Participants in the High-Empathy condition reported greater 

empathy than did participants in the Control-Empathy condition (MHighEmpathy = 3.61, 

MControlEmpathy = 3.35, t(768) = 3.53, p < .0005). 

Main results. A binary logistic regression simultaneously tested the influence of beauty, 

empathy, and their interaction on donation likelihood. We observed a marginally significant 



28 

positive main effect of beauty, Exp(B) = 1.46, p = .06, and a significant positive main effect of 

empathy, Exp(B) = 1.88, p < .01. Most importantly, we observed a significant interaction 

between beauty and empathy, Exp(B) = .53, p < .03. Within the Control-Empathy condition, 

participants who saw a beautiful photo were marginally significantly more likely to donate 

(59%) than those who saw a control photo (50%, χ
2
(1, N = 396) = 3.4, p = .06), replicating Study 

1C’s findings. Within the High-Empathy condition, however, participants who saw a beautiful 

photo were not more likely to donate (59%) than were participants who saw a control photo 

(65%; χ
2
(1, N = 374) = 1.6, p > .2). 

To more firmly understand the phenomenon from Study 5, we showed a new set of 

Mechanical Turk participants (N = 72) the high empathy description from Study 5 and told them 

that it prompted many people to donate. Participants then indicated whether they thought people 

donated because of 1) logical thoughts that the person was needy, 2) emotional reactions to the 

person’s situation, or 3) other reasons (with an option to type a response; the order of the first 

two response options was counterbalanced). A strong majority, 89%, of participants indicated 

that emotional reactions drove the response, providing evidence that the high-empathy 

manipulation was indeed an affective, System 1–based manipulation. 

Discussion 

 In Study 5 we observe that, when faced with recipients in severe need, donors experience 

high empathy and a tendency to help regardless of beauty. This result complements that from 

Fisher and Ma (2014) showing that when recipient need and donor empathy are high, donors feel 

compelled to help regardless of recipient beauty. In both sets of findings, empathy overwhelms 

beauty; however, we note that the natures of these two findings differ. In the case of Fisher and 

Ma (2014, p. 438), high empathy trumps a System 2 deliberative process that favors less 



29 

beautiful recipients; in the present study, high empathy trumps another System 1 process, an 

instinctive preference for beauty. 

We therefore have found that multiple preference patterns can result from intuitive 

System 1 processing; first, we have observed an instinctive draw toward beauty that leads to 

feelings of happiness and satisfaction (Studies 1B and 1C); second, we have observed that very 

high levels of empathy can push donors to choose needy recipients, regardless of their beauty 

(Study 5). Because there are multiple types of preferences that can arise from intuitive (versus 

deliberative) decisions, in a final study we test whether beautiful versus needy recipients 

specifically represent donors’ want versus should preferences. 

 

STUDY 6 

 

In Studies 2 through 4 donors were more likely to choose beautiful recipients when 

deciding intuitively but more likely to choose needy recipients when deciding deliberatively. 

However, there are multiple types of preference patterns that can result from intuitive (vs. 

deliberative) decision modes (Morewedge and Kahneman 2009; Stanovich 2009), including, for 

example, stronger sensitivity to affect-rich options (Rottenstreich and Hsee 2004) or greater 

sensitivity to emotional influences like empathy (Loewenstein and Small 2007). In Study 6, we 

test whether donors’ preferences for beautiful versus needy recipients map onto their want versus 

should desires respectively. 

Method 

Two hundred and forty-eight Mechanical Turk participants (146 female, 102 male; MAge 

= 33) participated in the study online in exchange for a small payment. Participants viewed the 



30 

sponsorship scenario and photos from the Beauty condition from Study 1B and then answered 

two questions on the same page: “Which one of these children do you think the donors would 

want to choose?” and “Which one of these children do you think the donors should choose?” 

(the words “want” and “should” were bolded in the actual questions; cf. Milkman, Rogers, and 

Bazerman 2009). 

Results  

A within-subjects omnibus McNemar-Bowker test showed a difference between 

conditions in which child was selected, χ
2
(6, n = 247) = 79.8, p < .0005. When participants 

considered which child donors want to give to, they chose Angelica (54% of choice share), the 

most attractive and least needy child (Vera 24%, Maria 14%, Silva 7%, all ps < .0001 relative to 

Angelica’s choice share; see Figure 6). 

Results were different when participants considered which child donors should give to. 

With this question, Angelica was no longer the most frequent choice, now only receiving 13% of 

choice share, whereas Vera, who was previously rated as needier than the other children, was 

chosen the most frequently (Vera, 48%; Maria, 24%; Silva, 14%; all ps < .0001 relative to Vera’s 

choice share). 

In between-questions analysis, participants were more likely to choose Angelica when 

answering the “want” compared to “should” question (54% vs. 13%, z(247) = 9.67, p < .0001); 

participants were more likely to choose Vera when answering the “should” compared to “want” 

question (48% vs. 24%, z(247) = 5.57, p < .0001). 

***** Insert Figure 6 about here ***** 



31 

Discussion 

Study 6 lends direct support to the idea that donors choose beautiful charitable recipients 

to satisfy want preferences whereas they choose needy charitable recipients to satisfy should 

preferences. When asked which recipient donors want to choose, participants selected Angelica 

(the most beautiful, least needy recipient) most often, whereas when asked which recipient 

donors should choose, they selected Vera (the neediest recipient) most often. Participants were 

naturally able to express charity recipient preferences in want and should categories. 

We also once again see evidence that the charity beauty premium cannot be attributed to 

beliefs that beautiful recipients have more potential than their less beautiful counterparts 

(Langlois et al. 2000). If beliefs about recipient success drive the charity beauty premium, then 

participants ought to indicate that they should give to beautiful recipients. Instead, participants 

indicate that they should give to needier, but less beautiful, recipients instead. 

 

GENERAL DISCUSSION 

 

Donors explicitly endorse neediness as a top donation priority; however, we observe that 

they often choose beautiful, less needy recipients instead. We test a hypothesis that donors 

simultaneously hold a want preference to give to beautiful recipients and a should preference to 

give to needy recipients. We further test whether preferences for beautiful recipients are most 

likely to emerge when decisions are intuitive whereas preferences for needy recipients are most 

likely to emerge when decisions are deliberative. In Studies 1A through 1C, when beautiful 

recipients are included in donation recipient choice sets, they are chosen most frequently and 

increase donor satisfaction. In Studies 2 through 4, prompting deliberative, compared to intuitive, 



32 

decision making eliminates the charity beauty premium and directs donors toward the neediest 

recipients. In Study 5, we observe that inducing high levels of empathy for all recipients can 

overwhelm the charity beauty premium. Finally, in Study 6, we find that the precise nature of 

donors’ preferences for beautiful versus needy recipients corresponds with their want versus 

should desires: participants explicitly state that donors want to give to beautiful recipients but at 

the same time that they should give to less beautiful, needier ones. It seems that donors hold both 

sets of preferences, and that contextual factors favoring intuitive versus deliberative processing 

drive which preferences they will act upon. 

The present research reconciles seemingly contradictory previous findings about beauty 

and giving. Although a charity beauty premium is consistent with some past research showing a 

positive effect of beauty on helping (Mims, Hartnett, and Nay 1975; West and Brown 1975), 

recent research uncovered a negative effect of beauty on empathy and helping (Fisher and Ma 

2014). We bring this work together by identifying the circumstances under which each pattern 

occurs: when decisions are intuitive, attractive recipients receive more help; when decisions are 

deliberative, unattractive recipients receive more help. 

Our findings suggest that when donors decide intuitively, charities may benefit by using 

beautiful photos, whereas when donors decide deliberatively, charities may benefit by 

emphasizing recipient neediness instead. Intuitive System 1 processing often is considered to be 

the default decision mode, routinely guiding thoughts and actions that are accepted by the 

deliberative system when processing under low effort (Evans and Stanovich 2013; Kahneman 

2011). A frequent predominance of intuition over deliberation, in combination with the present 

findings, sheds light on why needy recipients may be neglected in the real world; if donors 

predominantly process information intuitively, beauty is likely to be favored. Further, these 



33 

patterns suggest that charities might be best off using beautiful photos as a default unless they 

believe their donors will be particularly deliberative when choosing (though we emphasize that 

all organizations should test patterns amongst their own donors and donation contexts). 

Study 5 identifies a situation when System 1 processing does not result in a charity 

beauty premium: when emotional empathy is high, donors no longer prefer beautiful recipients. 

Although one reaction to this finding is that beauty might have minimal influence in the real 

world because of an overriding influence of strong empathy, we caution against such an 

interpretation. First, as we mention in our introduction, real world donations show that the 

highest-need causes, which have potential to trigger high levels of empathy, do not receive a 

substantial share of charitable contributions (Giving USA 2015). Second, empathy and sympathy 

are often characterized as negative emotional states (Cialdini et al. 1987; Small and Verrochi 

2009) and recent neuroimaging research that predicts real world giving finds stronger effects of 

positive rather than negative donor emotional responses (Genevsky and Knutson 2015; 

Genevsky, et al. 2013). Finally, recent field evidence suggests that donors may be adept at 

avoiding situations where they expect to feel high empathy. A fascinating field study showed 

that when Salvation Army solicitors working outside stores made a direct plea to potential 

donors, looking them in the eye and asking “please give today,” donations increased. However, 

at the same time, traffic patterns into and out of the stores where solicitors operated also changed 

so that potential donors avoided encountering the solicitor altogether (Andreoni, Rao, and 

Trachtman 2011). These results and others suggest that, once potential donors’ attention is 

captured, high empathy appeals can be effective (Dickert and Slovic 2009); however, when 

given the opportunity, potential donors may actively attempt to avoid often uncomfortable 

empathetic appeals. 



34 

An additional preliminary study that we conducted sounds a cautionary note about 

increasing deliberation to encourage donors to choose the neediest recipients. In this study, 

designed similarly to Study 4, participants were randomly assigned to an Intuitive condition in 

which they simply chose a recipient to sponsor, or a Deliberative condition in which they rated 

each potential recipient on neediness before selecting one to sponsor. As in Study 4, the 

Deliberative condition effectively prompted donors to select a needier, less beautiful recipient 

compared to the Intuitive condition. However, later in the same session we asked participants to 

predict how much they would donate to the overall charity organization one year from now. 

Deliberative condition participants indicated that they would donate significantly less money one 

year from now compared to those in the Intuitive condition, who had been more likely to initially 

choose beauty. In other words, participants who were prompted to choose needy recipients over 

beautiful ones did so, but showed less willingness to donate in the future. Deliberation, therefore, 

appears to be a double-edged sword in that it increases priority for the neediest causes in the 

short term but may decrease kindness in response to future requests. Choosing in favor of 

neediness might satisfy a deliberative priority but might also diminish the immediate 

gratification that individuals derive from their want charity choices, suppressing future 

generosity. Taken together, this evidence again tentatively suggests that charities might benefit 

by relying on beautiful photos unless they have a specific reason to believe their donors decide 

deliberatively when donating. 

In conclusion, scholars have attempted for many years to understand the puzzle of why 

highly important campaigns remain underfunded while other, less desperate ones receive an 

outpouring of support. One of the most notable insights from this research is that focusing on the 

large scope of a problem, a natural way to present the world’s neediest causes, can undermine 



35 

fundraising efforts by blocking donors’ capacity for sympathy (Kogut and Ritov 2005; Small, 

Loewenstein, and Slovic 2007; Smith, Faro, and Burson 2013) and by obscuring a donor’s sense 

of impact (Cryder, Loewenstein, and Scheines 2013; Grant 2007). In this research, we posit 

another reason why donors do not favor the world’s neediest recipients: they are drawn to 

beautiful and pleasing ones instead. We examine donors’ intuitive preference for beauty and find 

that it represents a want desire to give to instinctively pleasing targets. We posit that multiple 

other want charity preferences exist as well, including preferences to give to high-status causes 

such as prestigious universities or art museums, or personally relevant causes such as charities 

that have benefited friends or family (Small and Simonsohn 2008). We believe that 

understanding donors’ desires to give to what feels good versus what feels optimal, as well as the 

contextual factors that favor one set of preferences over the other, present a fruitful research path 

for both scientists and practitioners.  



36 

REFERENCES 

Aharon, Itzhak, Nancy Etcoff, Dan Ariely, Christopher F. Chabris, Ethan O’ Connor, and Hans 

C. Breiter (2001), “Beautiful Faces have Variable Reward Value: fMRI and Behavioral 

Evidence,” Neuron, 32 (3), 537–51. 

Andreoni, James (1990), “Impure Altruism and Donations to Public Goods: A Theory of Warm-

Glow Giving,” The Economic Journal, 100 (401), 464–77. 

———, Justin M. Rao, and Hannah Trachtman (2011), “Avoiding the Ask: A Field Experiment 

on Altruism, Empathy, and Charitable Giving,” National Bureau of Economic Research, 

No. w17648. 

Batson, C. Daniel (1991), The Altruism Question: Toward a Social-psychology Answer, New 

York: Lawrence Erlbaum Associates, Inc. 

——— (1998), “Altruism and Prosocial Behavior,” The Handbook of Social Psychology, Vol. 2, 

Daniel T. Gilbert, Susan T. Fiske, and Gardner Lindzey, eds. New York: McGraw-Hill, 

282–316. 

———, David A. Lishner, Jennifer Cook, and Stacey Sawyer (2005), “Similarity and 

Nurturance: Two Possible Sources of Empathy for Strangers,” Basic and Applied Social 

Psychology, 27 (1), 15–25. 

Bazerman, Max H., Ann E. Tenbrunsel, and Kimberly Wade-Benzoni (1998), “Negotiating with 

Yourself and Losing: Making Decisions with Competing Internal Preferences,” Academy 

of Management Review, 23 (2), 225–41. 

Bitterly, T. Bradford, Robert Mislavsky, Hengchen Dai, and Katherine L. Milkman (2014), 

“Dueling with Desire: A Synthesis of Past Research on Want/Should Conflict,” Available 

at SSRN 2403021. 



37 

Bloch, Peter H., Frederic F. Brunel, and Todd J. Arnold, (2003), “Individual Differences in the 

Centrality of Visual Product Aesthetics: Concept and Measurement,” Journal of 

Consumer Research, 29 (4), 551–65. 

Buss, David, and David Schmidt (1993), “Sexual Strategies Theory: An Evolutionary 

Perspective on Human Mating,” Psychological Review, 100 (2), 204–32. 

Cain, Daylian M., Jason Dana, and George E. Newman (2014), “Giving vs. Giving In,” Academy 

of Management Annals, 8, 505–33. 

Chaiken, Shelly, and Yaacov Trope (1999), Dual-process Theories in Social Psychology, New 

York: Guilford Press. 

Choi, Jinhee, B. Kyu Kim, Incheol Choi, and Youjae Yi (2006), “Variety-seeking Tendency in 

Choice for Others: Interpersonal and Intrapersonal Causes,” Journal of Consumer 

Research, 32 (4), 590–95. 

Cialdini, Robert B., Mark Schaller, Donald Houlihan, Kevin Arps, Jim Fultz, and Arthur L. 

Beaman (1987), “Empathy-based Helping: Is it Selflessly or Selfishly motivated?” 

Journal of Personality and Social Psychology, 52, 749–58. 

Cryder, Cynthia E., George Loewenstein and Richard Scheines (2013), “The Donor is in the 

Details,” Organizational Behavior and Human Decision Processes, 120, 15–23. 

Dion, Karen K., Ellen Berscheid, and Elaine Walster (1972), “What is Beautiful is Good,” 

Journal of Personality and Social Psychology, 24 (3), 285–90. 

Dunn, Elizabeth W., Lara B. Aknin, and Michael I. Norton (2008), “Spending Money on Others 

Promotes Happiness,” Science, 319 (5870), 1687–88. 

Evans, Jonathan S. B., and Keith E. Stanovich (2013), “Dual-process Theories of Higher 

Cognition Advancing the Debate,” Perspectives on Psychological Science, 8 (3), 223–41. 



38 

Fisher, Robert J., and Yu Ma (2014), “The Price of Being Beautiful: Negative Effects of 

Attractiveness on Empathy for Children in Need,” Journal of Consumer Research, 41 (2), 

436–50. 

Genevsky, Alexander, and Brian Knutson (2015), “Neural Affective Mechanisms Predict 

Market-level Microlending,” Psychological Science, 26 (9), 1411–22. 

Genevsky, Alexander, Daniel Västfjäll, Paul Slovic, and Brian Knutson (2013), “Neural 

Underpinnings of the Identifiable Victim Effect: Affect Shifts Preferences for Giving,” 

The Journal of Neuroscience, 33 (43), 17188–17196. 

“GiveWell: Our Criteria” (2013), http://www.givewell.org/criteria. 

Giving USA: The Annual Report on Philanthropy for the year 2014 (2015). Chicago: Giving 

USA Foundation. 

Grant, Adam (2007), “Relational Job Design and The Motivation to Make a Prosocial 

Difference,” Academy of Management Review, 32, 393–417. 

Halberstadt, Jamin, and Gillian Rhodes (2000), “The Attractiveness of Nonface Averages: 

Implications for an Evolutionary Explanation of the Attractiveness of Average Faces,” 

Psychological Science, 11 (4), 285–9. 

Hamermesh, Daniel S., and Jeff E. Biddle (1994), “Beauty and the Labor Market,” American 

Economic Review, American Economic Association, 84 (5), 1174–94. 

Hatfield, Elaine, John T. Cacioppo and Richard L. Rapson (1994), Emotional contagion.  

Cambridge, England: Cambridge University Press. 

“IUCN Red List 2014” (2014), http://www.iucnredlist.org/. 

Kahneman, Daniel (2011), Thinking, Fast and Slow. New York, NY: Farrar, Strauss and Giroux. 

http://www.givewell.org/criteria
http://www.iucnredlist.org/


39 

——— and Shane Frederick (2002), “Representativeness Revisited: Attribute Substitution in 

Intuitive Judgment,” in Heuristics and Biases: The Psychology of Intuitive Judgment, 

Thomas Gilovich, Dale Griffin, and Daniel Kahneman, eds. Cambridge, UK: Cambridge 

University Press, 49–81. 

Kogut, Tehila, and Ilana Ritov (2005), “The Identified Victim Effect: An Identified Group, or 

Just a Single Individual,” Journal of Behavioral Decision Making, 18 (3), 157–67. 

Kubovy, Michael (2000), “Visual Aesthetics,” Encyclopedia of Psychology, 8, 188–93. 

Kurt, Didem, and J. Jeffrey Inman (2013), “Mispredicting Others’ Valuations: Self-Other 

Difference in the Context of Endowment,” Journal of Consumer Research, 40 (1), 78–89. 

Landy, David, and Harold Sigall (1974), “Beauty is Talent: Task Evaluation as a Function of the 

Performer’s Physical Attractiveness,” Journal of Personality and Social Psychology, 29 

(3), 299–304. 

Landry, Craig, Andreas Lange, John A. List, Michael K. Price, and Nicholas G. Rupp (2005), 

“Toward an Understanding of the Economics of Charity: Evidence from a Field 

Experiment,” National Bureau of Economic Research, No. w11611. 

Langlois, Judith H., Lisa Kalakanis, Adam J. Rubenstein, Andrea Larson, Monica Hallam, and 

Monica Smoot (2000), “Maxims or Myths of Beauty? A Meta-Analytic and Theoretical 

Review,” Psychological Bulletin, 126 (3), 390–423. 

Leder, Helmut, Benno Belke, Andries Oeberst, and Dorothee Augustin (2004), “A Model of 

Aesthetic Appreciation and Aesthetic Judgments,” British Journal of Psychology, 95 (4), 

489–508. 

Lerner, Jennifer S., and Philip E. Tetlock (1999), “Accounting for the Effects of Accountability,” 

Psychological Bulletin, 125 (2), 255–75. 



40 

Loewenstein, George (1996), “Out of Control: Visceral Influences on Behavior,” Organizational 

Behavior and Human Decision Processes, 65 (3), 272–92. 

——— and Deborah A. Small (2007), “The Scarecrow and the Tin Man: The Vicissitudes of 

Human Sympathy and Caring,” Review of General Psychology, 11 (2), 112–26. 

MacAskill, William (2015), Doing Good Better: How Effective Altruism Can Help You Make a 

Difference, New York: Gotham. 

Milkman, Katherine L., Dolly Chugh, and Max H. Bazerman (2009), “How can Decision 

Making be Improved?” Perspectives on Psychological Science, 4 (4), 379–83. 

———, Todd Rogers, and Max H. Bazerman (2008), “Harnessing our Inner Angels and 

Demons: What we have Learned about Want/Should Conflicts and how that Knowledge 

can Help us Reduce Short-sighted Decision Making,” Perspectives on Psychological 

Science, 3 (4), 324–38. 

———, ———, and ——— (2009), “Highbrow Films Gather Dust: Time-inconsistent 

Preferences and Online DVD Rentals,” Management Science, 55 (6), 1047–59. 

Miller, Arthur G. (1970), “Role of Physical Attractiveness in Impression Formation,” 

Psychonomic Science, 19 (4), 241–3. 

Mims, Patricia R., John J. Hartnett, and W. Robert Nay (1975), “Interpersonal Attraction and 

Help Volunteering as a Function of Physical Attractiveness,” Journal of Psychology, 89 

(1), 125–31. 

Mobius, Markus M., and Tanya S. Rosenblat (2006), “Why Beauty Matters,” The American 

Economic Review, 96 (1), 222–35. 

Morewedge, Carey K., and Daniel Kahneman (2010), “Associative Processes in Intuitive 

Judgment,” Trends in Cognitive Sciences, 14 (10), 435–40. 



41 

Mulford, Matthew, John Orbell, Catherine Shatto, and Jean Stockard (1998), “Physical 

Attractiveness, Opportunity, and Success in Everyday Exchange,” American Journal of 

Sociology, 103 (6), 1565–92. 

Newman, George E., and Paul Bloom (2012), “Art and Authenticity: The Importance of 

Originals in Judgments of Value,” Journal of Experimental Psychology: General, 141 

(3), 558. 

Norman, Donald A. (2004), Emotional Design: Why we Love (or Hate) Everyday Things, New 

York: Basic Books. 

Oppenheimer, Daniel M., Tom Meyvis, and Nicolas Davidenko (2009), “Instructional 

Manipulation Checks: Detecting Satisficing to Increase Statistical Power,” Journal of 

Experimental Social Psychology, 45 (4), 867–72. 

Owen, David (2013), “Bears Do It,” The New Yorker, September 2, 26–31. 

Polman, Evan (2012), “Self-other Decision Making and Loss Aversion,” Organizational 

Behavior and Human Decision Processes, 119 (2), 141–50. 

Ratner, Rebecca K., and Barbara E. Kahn (2002), “The Impact of Private versus Public 

Consumption on Variety-Seeking Behavior,” Journal of Consumer Research, 29 (2), 

246–57. 

Ratner, Rebecca K., and Dale T. Miller (2001), “The Norm of Self-interest and Its Effects on 

Social Action,” Journal of Personality and Social Psychology, 81 (1), 53–62. 

Reber, Rolf, Norbert Schwarz, and Piotr Winkielman (2004), “Processing Fluency and Aesthetic 

Pleasure: Is Beauty in the Perceiver’s Processing Experience?” Personality and Social 

Psychology Review, 8 (4), 364–82. 



42 

Reimann, Martin, Judith Zaichkowsky, Carolin Neuhaus, Thomas Bender, and Bernd Weber 

(2010), “Aesthetic Package Design: A Behavioural, Neural, and Psychological 

Investigation,” Journal of Consumer Psychology, 20 (4), 431–41. 

Rhodes, Gillian (2006), “The Evolutionary Psychology of Facial Beauty,” Annual Review of 

Psychology, 57, 199–226. 

Rottenstreich, Yuval, and Christopher K. Hsee (2001), “Money, Kisses, and Electric Shocks: On 

the Affective Psychology of Risk,” Psychological Science, 12 (3): 185–90. 

Singer, Peter (2009), The Life You Can Save: Acting Now to End World Poverty, New York: 

Random House, Inc. 

——— (2013), ‘The Why and How of Effective Altruism,’ TED2013. 

http://www.ted.com/talks/peter_singer_the_why_and_how_of_effective_altruism. 

Small, Deborah A., and Cynthia Cryder (2016), “Prosocial Consumer Behavior,” Current 

Opinion in Psychology, 10, 107–11. 

———, George Loewenstein, and Paul Slovic (2007), “Sympathy and Callousness: The Impact 

of Deliberative Thought on Donations to Identifiable and Statistical Victims,” 

Organizational Behavior and Human Decision Processes, 102 (2), 143–53. 

———, and Uri Simonsohn (2008), “Friends of Victims: Personal Experience and Prosocial 

Behavior,” Journal of Consumer Research, 35, 532–42. 

———, and Nicole M. Verrochi (2009),”The Face of Need: Facial Emotion Expression on 

Charity Advertisements,” Journal of Marketing Research, 46 (6), 777–87. 

Smith, Robert W., David Faro, and Katherine A. Burson (2013), “More for the Many: The 

Influence of Entitativity on Charitable Giving,” Journal of Consumer Research, 39 

(February), 961–76. 

http://www.ted.com/talks/peter_singer_the_why_and_how_of_effective_altruism


43 

Stanovich, Keith E., and Richard F. West (2000), “Advancing the Rationality Debate,” 

Behavioral and Brain Sciences, 23 (5), 701–17. 

Stanovich, K. (2009), Distinguishing the Reflective, Algorithmic, and Autonomous Minds: Is It 

Time for a Tri-process Theory. In In Two Minds (Evans, J.S.B.T. and Frankish, K., eds), 

pp. 33–54, Oxford University Press. 

Townsend, Claudia, and Suzanne B. Shu (2010), “When and How Aesthetics Influences 

Financial Decisions,” Journal of Consumer Psychology, 20 (4), 452–8. 

———, and Sanjay Sood (2012), “Self-Affirmation Through the Choice of Highly Aesthetic 

Products,” Journal of Consumer Research, 39 (2), 415–28. 

UNICEF (2014), http://www.unicef.org. 

West, Stephen G., and T. Jan Brown (1975), “Physical Attractiveness, the Severity of the 

Emergency and Helping: A Field Experiment and Interpersonal Simulation,” Journal of 

Experimental Social Psychology, 11 (6), 531–38. 

Wispé, Lauren (1986), “The Distinction between Sympathy and Empathy: To Call Forth a 

Concept a Word Is Needed,” Journal of Personality and Social Psychology, 50 (2), 314–

21. 

Zaatari, Darine, Brian G. Palestis, and Robert Trivers (2009), “Fluctuating Asymmetry of 

Responders Affects Offers in the Ultimatum Game Oppositely According to 

Attractiveness or Need as Perceived by Proposers,” Ethology, 115 (7), 627–32. 

Zhao, Xinshu, John G. Lynch, and Qimei Chen (2010), “Reconsidering Baron and Kenny: Myths 

and Truths about Mediation Analysis,” Journal of Consumer Research, 37 (2), 197–206.  

http://www.unicef.org/


44 

Footnotes 

1 
Recent conceptualizations of want versus should preferences characterize them as tradeoffs that 

occur over time, with want options being those that are immediately gratifying and should 

options being those that are gratifying over the long term (e.g., Bitterly et al. 2014; Milkman, 

Rogers, and Bazerman 2008). In the current research, we rely on previous broader 

conceptualizations of want versus should tradeoffs in which want preferences are infused with a 

notion of desire even if that desire does not have a clear intertemporal component (e.g., 

Bazerman et al. 1998). 

 
2
All photo stimuli are available via e-mail from the first author. 

 
3
Sample sizes for all experiments were determined ex ante. At the end of all experiments, we 

included an “instructional manipulation check” (IMC; Oppenheimer, Meyvis, and Davidenko 

2009) to identify inattentive participants. Participants who failed this attention check were 

removed before any analyses were conducted; thus, all reported results exclude them. Here we 

report the numbers of participants excluded from each experiment based on this criterion: Study 

1B: 117; Study 1C: 70; Study 2: 55; Study 3: 75; Study 4: 112; Study 5: 281; Study 6: 165. 

 
4
We conducted a similar study, reported in Web Appendix B, using a choice set including four 

boys rather than four girls. 

 
5
We report social competence and emotional expression ratings from Study 4 here. Consistent 

with findings from Fisher and Ma (2014) and other research about the beauty premium (Langlois 

et al. 2000), participants in the Deliberative condition rated Angelica, the most attractive 

recipient, as also the most sociable (MAngelica = 4.26, MVera = 3.07, MMaria = 3.75, MSilva = 3.01, ps 

< .0005), most helpful (MAngelica = 4.10, MVera = 3.57, MMaria = 4.01, MSilva = 3.39, ps < .05), most 

intelligent (MAngelica = 3.61, MVera = 3.09, MMaria = 3.15, MSilva = 2.89, ps < .0005), and as having 

the least negative facial expression (MAngelica = 1.33, MVera = 3.11, MMaria = 1.49, MSilva = 3.24, ps 

< .0005). 

  



45 

TABLE 1 

Study 1A: Beauty Ratings, Actual Endangerment, and Perceived Endangerment Ratings 

Animals 

Average 

beauty ratings 

(SE) 

Actual 

endangerment status 

(numerical level 

from IUCN 2015) 

Average perceived 

level of 

endangerment (SE) 

Giraffe 5.81 (.15) Least Concern (1) 3.66 (.15) 

Zebra 5.12 (.15) Least Concern (1) 3.49 (.16) 

Flamingo 4.79 (.14) Near Threatened (2) 4.02 (.20) 

Elephant 4.66 (.16) Endangered (4) 4.95 (.16) 

Chimpanzee 4.42 (.15) Endangered (4) 4.47 (.16) 

Lemur 4.24 (.16) Endangered (4) 3.93 (.17) 

Orangutan 4.07 (.16) Endangered (4) 4.77 (.16) 

Penguin 4.06 (.16) Vulnerable (3) 3.87 (.17) 

  



46 

FIGURE 1 

STUDY 1A: PERCENTAGE OF SPONSORSHIPS FOR THE MOST AND LEAST 

BEAUTIFUL ANIMALS 

 

  

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

50%

55%

60%

65%

70%

4 least beautiful animals 4 most beautiful animals

S
p

o
n

s
o

rs
h

ip
 a

ll
o

c
a

ti
o

n
s

 

Chimpanzee

Elephant

Flamingo

Giraffe

Lemur

Orangutan

Penguin

Zebra



47 

FIGURE 2 

STUDY 1B: PERCENTAGE OF ALLOCATIONS RECEIVED BY EACH CHILD BASED ON 

CONTROL VERSUS BEAUTY CONDITION 

 

**p < .01 

***p < .001 

  

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

50%

Vera Maria Silva Angelica/YaoLin

D
o

n
a

ti
o

n
 A

ll
o

c
a

ti
o

n
s

 

Beauty Control

** 
*** 



48 

FIGURE 3 

STUDY 2: PERCENTAGE OF ALLOCATIONS RECEIVED BY EACH CHILD BASED ON 

INTUITIVE VERSUS DELIBERATIVE CONDITION 

 

**p < .01 

 

  

0%

10%

20%

30%

40%

50%

Vera Maria Silva Angelica

D
o

n
a

ti
o

n
 A

ll
o

c
a

ti
o

n
s

 

Intuitive Deliberative

** 



49 

FIGURE 4 

STUDY 3: PERCENTAGE OF ALLOCATIONS RECEIVED BY EACH CHILD BASED ON 

INTUITIVE VERSUS DELIBERATIVE CONDITION 

 

*p < .05 

  

0%

10%

20%

30%

40%

50%

Vera Maria Silva Angelica

D
o

n
a

ti
o

n
 A

ll
o

c
a

ti
o

n
s

 

Intuitive (Self) Deliberative (Adviser)

* * 



50 

FIGURE 5 

STUDY 4: PERCENTAGE OF ALLOCATIONS RECEIVED BY EACH CHILD BASED ON 

INTUITIVE VERSUS DELIBERATIVE CONDITION 

 

*p < .05 

***p < .001 

  

0%

10%

20%

30%

40%

50%

Vera Maria Silva Angelica

D
o

n
a

ti
o

n
 A

ll
o

c
a

ti
o

n
s

 

Intuitive (Control) Deliberative (Rating)

*** * 



51 

FIGURE 6 

STUDY 6: CHOICES BASED ON WANT VERSUS SHOULD PREFERENCES 

 

 

***p < .001 

0%

10%

20%

30%

40%

50%

60%

Vera Maria Silva Angelica

R
e
c

ip
ie

n
t 

C
h

o
ic

e
 

Should Want

*** ***