Prosocial Behavior Leads to Happiness in a Small-Scale Rural Society
Lara B. Aknin & Tanya Broesch
Simon Fraser University
J. Kiley Hamlin & Julia W. Van de Vondervoort
University of British Columbia
Word Count: 5847
Lara B. Aknin
Simon Fraser University
Department of Psychology, 8888 University Drive, Burnaby, British Columbia, V5A 1S6
Humans are extraordinarily prosocial, and research conducted primarily in North America
indicates that giving to others is emotionally rewarding. To examine whether the hedonic
benefits of giving represent a universal feature of human behavior, we extended upon previous
cross-cultural examinations by investigating whether inhabitants of a small-scale, rural, and
isolated village in Vanuatu, where villagers have little influence from urban, Western culture,
survive on subsistence farming without electricity, and have minimal formal education, report or
display emotional rewards from engaging in prosocial (vs. personally beneficial) behavior. In
Study 1, adults were randomly assigned to purchase candy for either themselves or others and
then reported their positive affect. Consistent with previous research, adults purchasing goods for
others reported greater positive emotion than adults receiving resources for themselves. In Study
2, 2- to 5-year-old children received candy and were subsequently asked to engage in costly
giving (sharing their own candy with a puppet) and non-costly giving (sharing the
experimenter’s candy with a puppet). Emotional expressions were video-recorded during the
experiment and later coded for happiness. Consistent with previous research conducted in
Canada, children displayed more happiness when giving treats away than when receiving treats
themselves. Moreover, the emotional rewards of giving were largest when children engaged in
costly (vs. non-costly) giving. Taken together, these findings indicate that the emotional rewards
of giving are detectable in people living in diverse societies and support the possibility that the
hedonic benefits of generosity are universal.
KEYWORDS: Prosocial behavior, well-being, happiness, generosity, psychological universal
To a far greater extent than any other species, humans help unrelated others even when
doing so is personally costly. But why? While numerous theories underscore the ultimate
rewards of prosocial behavior for evolutionary fitness (reputational concerns, Benabou & Tirole,
2006; kin selection, Hamilton, 1963; indirect reciprocity, Nowak, 2006; direct reciprocity,
Trivers, 1971), an additional possibility is that humans give to others because giving feels good.
Indeed, a growing body of evidence supports the possibility that prosocial behavior
promotes positive emotional rewards for the giver. Numerous correlational studies have
documented a positive link between generous behaviors, such as offering time or money, and
well-being. For instance, Borgonovi (2008) reported a positive relationship between volunteering
and well-being across 29 states in America: the more people volunteered, the happier they
reported being, even while accounting for a number of demographic (e.g., age, marital status)
and socio-economic (e.g., income, education) factors. Moreover, the relationship between
generous behavior and happiness is causal: North American university students randomly
assigned to spend a small windfall on others were significantly happier at the end of the day than
students assigned to spend money on themselves (Dunn, Aknin & Norton, 2008). The emotional
benefits derived from giving can even be detected when givers have no direct contact with the
beneficiary, and when experimenters are unaware of condition assignment, suggesting that
happiness is not simply a result of building social relationships or anticipating social praise
(Aknin, Fleerackers & Hamlin, 2014).
Even young children are motivated to help others in need (Hepach, Vaish & Tomasello,
2012; Warneken & Tomasello, 2006) and find giving rewarding. Twenty-two-month old toddlers
given edible treats (e.g., teddy graham crackers) and subsequently asked to give some away were
rated as happier when giving treats away than when receiving treats, regardless of whether the
treats belonged to themselves or to an experimenter (Aknin, Hamlin & Dunn, 2012). In fact,
toddlers were rated as happier when they gave away their own treat than when giving away a
treat that belonged to the experimenter, suggesting that young children find giving to others
rewarding even (or, especially) if it is personally costly. But is the warm glow of giving detected
among adults and toddlers in North America a universal feature of human prosociality?
Though these findings paint a promising portrait of the emotional rewards of engaging in
prosocial behavior, research to date has been conducted almost exclusively in Western,
Educated, Industrialized, Rich and Democratic (WEIRD) societies, whose people represent just
12% of the world’s population (Henrich, Heine & Norenzayan, 2010a, 2010b). Although studies
conducted in North America and other WEIRD populations are valuable in their own right, it
may be problematic to utilize findings from WEIRD samples alone to draw conclusions about
human beings in general. Indeed, it has recently been demonstrated that non-WEIRD populations
respond differently than WEIRD populations in some common psychological tasks; for example,
individuals from certain hunter-gatherer groups fail to show the Muller-Lyer visual illusion,
traditionally considered to reflect universal characteristics of the visual system (McCauley &
Henrich, 2006). More relevant to the domain of prosocial behavior, there is now evidence for
considerable cross-cultural variation in humans’ distributive and punitive tendencies during
economic games (Henrich et al., 2010a). In these studies, prosocial responding (as measured by
fair distributions, third-party punishment, etc.) has been shown to vary predictably with the level
of market-integration and religiosity in a given society, suggesting that what have traditionally
been considered universal forms of prosocial behavior vary considerably across cultures.
Together, these findings highlight the importance of exploring whether and how phenomena
observed in WEIRD populations generalize to humanity more broadly.
To address the possibility that the relationship between happiness and giving previously
demonstrated in North American adults is a phenomenon of a WEIRD environment, Aknin and
colleagues (2013) examined the relationship between financial generosity and well-being in
adults from both Western and non-Western and rich and poor countries including Canada, India,
South Africa, and Uganda. Results replicated earlier findings demonstrating that spending money
on others led to greater happiness than spending money on oneself, suggesting that deriving
emotional rewards from giving might represent a psychological universal, or “core mental
attributes shared by humans everywhere” (Norenzayan & Heine, 2005, p. 763). Importantly,
however, several of these studies utilized university student populations and adults recruited via
an online marketplace. Thus, because these samples had access to formal education and/or the
internet, it is possible that participants were at least somewhat WEIRD: They were presumably
1) influenced by Western culture and ideals, 2) relatively wealthy within their societies, 3) raised
in more urban and less-collective living environments, 4) living amongst a large population, and
5) living in societies with high levels of marketplace integration. Therefore, it currently remains
unclear whether the cross-cultural similarity in happiness from giving observed by Aknin and
colleagues (2013) reflects the existence of a universal mechanism that encourages prosociality,
or rather stems from various experiences shared with individuals in the Western world.
The current investigation builds upon this earlier work by examining the emotional
rewards of generosity in remote, small-scale rural villages on Tanna Island in Vanuatu, a small
island nation in the South Pacific. In these villages, life is notably non-WEIRD, and inhabitants
live in ways that more resemble the practices of our human ancestors. They survive on
subsistence-based living – eating a plant-based diet filled with foods farmed in a large
community garden or grown wild in nearby areas – and reside in homes built from local, natural
materials including earth floors and bamboo, sugarcane, or palm walls. Since the villages are
inland and there is no running water, locals drink rainwater collected in the one functional water
tank. Homes have no electricity, and there are no televisions, sharply limiting villager’s access to
Western or urban cultures. Traditional forms of learning are practiced and there is little emphasis
on formal education. Most villagers rarely travel to the ‘main island’ of Efate is because travel
consists of expensive airfare or long trip by boat. As such, this population diverges from
previously studied populations in that they have minimal influence from Western culture (and
even outright rejection of Western customs in “Kastom” villages), monetary scarcity, a rural,
inter-dependent lifestyle, small population size, and very little marketplace integration. Since
these dimensions have been shown to shape human culture and psychology (Henrich, et al.,
2001; Henrich, Ensimger, et al., 2010), the current examination offers a particularly strong test of
the generalizability of earlier findings.
In the current studies, we explore whether prosocial behavior leads to emotional gains in
both adults and 2- to 5-year-old Tannese children, utilizing two methodologies previously
validated in Western contexts with similar age groups (e.g., Aknin et al., 2012; 2013). In Study
1, adults were provided with the opportunity to purchase candy for either themselves or others
before reporting their positive and negative affect. In Study 2, young children between the ages
of 2-5 years were given candy and then asked to share some of their own or the experimenter’s
candies with a third-party while their emotional expressions were captured on videotape and later
coded for happiness. If the emotional benefits of giving are detectable among adults and children
in this small-scale, rural society, this would provide additional support for the possibility that
people from diverse cultural contexts find generosity rewarding.
Twenty-six adults living in small island villages on Tanna Island, Vanuatu (Mage = 45, 15
female) participated in this study in exchange for 100 Vatu (approximately 1 US dollar), or about
half a day’s wage. Adults were recruited by word of mouth. This sample size reflects the total
number of adults that we were able to recruit during our time in the village; there was no
stopping rule other than participant availability. The study was conducted in a quiet local
dwelling in the village. The experimenter was a local female adult, trained by the first author and
blind to the experimental hypothesis. Since literacy was variable among participants, materials
were administered verbally in Bislama, a linking language understood and spoken by most adults
on Tanna Island. Materials were back-translated to ensure accurate translation (i.e., documents
were translated from English to Bislama by one research assistant and then translated from
Bislama to English by a second research assistant who had not seen or heard the intended
Adults were asked to report their current happiness and hunger on a 3-point scale (Do
you feel happy right now? Do you feel hungry right now? 1= not at all; 2 = a little; 3 = yes);
hunger was included as a filler question. Next, adults were told that they had earned an additional
100 Vatu, presented in the form of a 100 Vatu voucher, for their participation in the experiment.
(making their total earnings 200 Vatu). Adults were given the 100 Vatu voucher and asked to put
the voucher away in their possession (e.g., put it in their pocket, if applicable) to encourage
ownership of it. The researcher explained that the voucher was worth 100 Vatu so that
participants who could not read were aware of its value. Then, adults were told that they could
use their additional 100 Vatu voucher to purchase candy, a rare commodity in the village.
Participants were shown the candy (cookies and lollipops) and were informed that it was valued
at 200 Vatu, but that it would cost only their 100 Vatu voucher. Adults were randomly assigned
to either the personal spending condition (n = 13) in which they were informed they could
purchase candy for themselves, or the prosocial spending condition (n = 13), in which they were
informed they could purchase candy for friends or family. Adults in both conditions were given
the option of selecting between cookies, lollipops, or both, and made their selection by informing
the research assistant who immediately handed the items to the participant. Adults in both
conditions were also informed that they could opt out of purchasing candy, and instead redeem
the voucher for 100 Vatu in cash for themselves approximately one week later. Participants were
offered this opportunity to opt out of buying candy so that those in the prosocial spending
condition would not feel forced to engage in a generous act, as past research has shown that the
emotional benefits of giving are eliminated when people feel forced to give (Weinstein & Ryan,
2010). The one-week time delay was included to subtly discourage participants from selecting
the cash option (see Aknin et al., 2013 for a similar method). No one in the prosocial condition
opted to take the cash for themselves, but three adults in the personal condition spending
condition did; adults who selected the cash value in the personal spending condition were
included in analyses because participants received a form of self gain. Results remain unchanged
if these individuals are excluded1.
After making a purchasing decision, adults reported their current positive and negative
affect (specifically: happiness, pride, strength, sadness, and anger2) on a 1-11 scale. In
consultation with an expert in cross-cultural research, we decided to collect ratings using an
image of a ladder rather than a verbal or written assessment of affect in order to clearly convey
the linear nature of the scale; local research assistants indicated that this method was appropriate
for respondents in the village. Specifically, participants were shown a picture of a wooden ladder
with 11 rungs total, and were informed that higher rungs on the ladder indicate a lot of an
emotion, whereas lower rungs indicate very little of an emotion. Participants were instructed to
point to the rung on the ladder that best expressed their current level of a given affect variable.
The affect items were chosen from one of the most frequently used affect measures (the
Positive and Negative Affect Schedule; Watson, Clark & Tellegen, 1988) and/or for relevance to
the research question (i.e. “happiness”) in consultation with an expert in cross-cultural research.
In addition, in order to gain a thorough understanding of how our participants understood the
positive and negative affect terms a Tannese research assistant provided a definition of each term
and examples of when the emotion might be experienced. The research assistant explained the
affect terms as follows: Happiness was defined as “happy” or “pleasure”, an emotion that may be
prompted by community, social connection, and enjoyable experiences (e.g., dancing, marriage,
or gathering at the Nakamal- the village gathering spot). Strong was defined as “brave” or “a
person with courage”, an emotion that may be prompted by completing a task or achieving a
goal. Consistent with research on the two facets of the pride expression (e.g., authentic and
hubristic pride; Tracy & Robins, 2007a; 2007b), pride was defined as both “proud” and “show-
off”, an emotion that occurred when people feel proud because they or their close others
achieved important things that others have not, such as receiving a scholarship to study overseas.
Sadness was defined as “no good” or “bad”, an emotion that someone would experience when a
fellow villager dies or if they are missing something important to them. Finally, Anger was
defined as “angry”, an emotion someone may experience if someone had stolen from them or
their house was burned down by their family member as a result of land dispute.
Consistent with previous research (e.g., Dunn et al., 2008), the positive items happiness,
pride, and strength were correlated and so were averaged to form an index of positive affect
(alpha = .56); sadness and anger were not correlated and therefore were retained separately.
Finally, adults reported their sex and year of birth.
Adults randomly assigned to buy candy for others reported higher levels of positive affect
(M = 9.67, SE = .21) than adults assigned to buy candy for themselves (M = 8.51, SE = .44),
t(24) = 2.383, p < .03 (two-tailed), d = .93 (see Figure 1). Results remained relatively unchanged
when baseline levels of happiness were added as a covariate, F(1,23) = 4.364, p < .05 (two-
tailed), partial ƞ2 = .16 and were similar for men and women; when sex was added as a between
subjects factor, neither the main effect of sex nor the interaction of sexXcondition was
significant (Fs < 2.0, ps > .20).
To explore whether these results are similar in direction and magnitude to previous
research conducted in Canada and South Africa, we combined data from the present study with
those collected by Aknin and colleagues (2013) into one data set. We analyzed positive affect
reports standardized within each sample using a 2(condition) X 3(society) ANOVA and found a
significant main effect of condition, F(1,217) = 15.609, p < .001 (two-tailed), partial ƞ2 = .07,
such that participants who bought candy for someone else reported higher levels of positive
affect than those who bought candy for themselves. Importantly, there was no interaction by
society, F < .50, p > .60, indicating that the emotional rewards of prosocial spending did not
differ across sample. Thus, these findings replicate previous research conducted in South Africa
and Canada using a similar experimental design in a small-scale, rural society, and add to the
growing body of evidence indicating that people throughout the world experience hedonic
benefits from engaging in prosocial actions.
As a previous examination demonstrated that being prosocial makes young children
happy as early as 22 months of age (Aknin et al., 2012), in Study 2 we explored the emotional
benefits of giving and receiving among young children from the same village in Vanuatu.
Children were provided with several pieces of candy and subsequently asked to share some of
their candy with a puppet. Children’s facial reactions to receiving and providing treats were
captured on video and later coded for happiness. This investigation provides a convergent test of
the emotional rewards of giving in Vanuatu with a different sample (children instead of adults)
and different paradigm – using a within-subjects design and distinct dependent variable (smiling
instead of self-reported affect).
Twenty children from the same small-scale, rural villages on Tanna Island in Vanuatu (14
boys, Xage = approximately 2 years and 1 month, age range: 2 years, 4 months – 4 years, 8
months)3 participated in this experiment. A sample size of 20 children was determined in
advance, as consistent with previous research utilizing the same paradigm (Aknin et al., 2012), to
provide enough statistical power to detect a medium to large effect. Children were recruited by
word of mouth at a village meeting and through the local childcare facility. The current sample
included a wider age range than previously studied by Aknin and colleagues (2012) because the
villages had an insufficient number of toddlers between 22- to 24- months. The experiment was
conducted in the local village language. Translation accuracy was ensured through back-
translations as in Study 1. Children sat on a caregiver’s lap throughout the study.
Warm-up. The warm-up phase was designed to familiarize children with the testing
situation, to introduce them to puppets, and to show them that puppets enjoyed eating treats
(fruit-flavored candies). Each child was introduced to a plush dog puppet and encouraged to
wave to or touch him. The experimenter told the child that the puppet liked eating candies. Next,
the experimenter gave the child and puppet their own empty bowl; the puppet’s bowl had a false
bottom that was used to create the illusion that the dog could eat. The experimenter then gave a
candy to the puppet and then to the child while saying “Look! I’m going to give one of these
candies to Dog, and I’m going to give one of these candies to you!” The dog “ate” the treat
placed in its bowl by leaning its head in the bowl, making eating noises (“mmmm!” and “yum
yum yum!”), and pushing the candy through the false bottom. The experimenter then placed a
third “common” bowl with two additional candies next to the child’s bowl and said “Now it’s
your turn! Do you want to give Dog a candy from this?” while pointing to the “common” bowl.
Once the child did so, the dog puppet “ate” his candy just as when the experimenter gave him
one. The research assistant then said, “Do you want to eat your candy too?”, indicating that the
final candy was for the child.
Testing. After the warm up, children participated in a five-part testing phase identical to
that used in Aknin, Hamlin and Dunn (2012). In Phase 1, children were introduced to a new
puppet (“Monkey”), encouraged to wave at or to touch Monkey, and told Monkey liked candies.
This phase allowed children to interact with the puppet if they wished. The experimenter then
said, “Both you and Monkey have no candies right now.” In Phase 2, the experimenter then
“found” eight candies, and said “Oh look! I found some more candies. I’m going to give them all
to you” and poured them into the child’s bowl, providing the child with valuable resources. The
next three phases (3-5) were presented in counterbalanced order. In Phase 3, the experimenter
“found” another candy (hidden in opaque bowls on the side of the table) and gave it to the puppet
after saying “Oh look! I found one more candy. I’m going to give it to Monkey!” This phase
allowed the child to see the puppet receive a treat but did not require the child to interact with the
puppet or to forfeit any of his or her own resources. In Phase 4, the experimenter “found”
another candy and asked the child to give it to Monkey by saying “Oh look! I found one more
candy. Will you give it to Monkey?” In Phase 5, the experimenter asked the child to give one of
his or her candies to Monkey by saying “I don’t see any more candies. Will you give one of your
candies to Monkey?” Phase 4 was designed to provide an instance of non-costly giving because
the child was provided with a candy to give to Monkey; Phase 5 was designed to provide an
instance of costly-giving, whereby giving to Monkey involved a personal sacrifice. Monkey’s
reaction to receiving candies was always the same: he pushed the treat through the false bottom
of the bowl with his nose and excitedly said “yum yum yum!” If children hesitated to engage in
any of the requested actions, the experimenter asked again and provided prompts if necessary.
Afterward, all children were presented with a final candy and asked “Do you want to take this
candy for you? Or would you like to give this candy to Monkey?”
Children were videotaped during all phases. Emotional expressions were later coded from
videos for happiness by four coders (average α =.90; range = 83-.95). Three coders were from
Canada and one was native to the region in Vanuatu; the coder from Vanuatu was included to
confirm that Canadian coders could reliably identify facial expressions of happiness of children
in Vanuatu. Coders were kept blind to experimental hypotheses and the Canadian coders
watched audio-free video clips displaying only the child’s face in each phase; these precautions
were taken to ensure that contextual cues did not sway happiness ratings. All coders were
instructed to code the child’s emotional response to each action on a seven point Likert scale (1-
Extremely unhappy, 7- Extremely happy); the four coders’ ratings for each phase were averaged
Using two-tailed paired samples t-tests, we investigated the emotional rewards of giving
and receiving. Specifically, using each child as their own control, we looked at within-participant
differences in happiness displayed as children received treats from the experimenter and gave
their own treat to Monkey. Children displayed significantly more happiness when engaging in
costly giving – providing their own candy to Monkey (M = 5.24, SE =.18) – than when they
received candies themselves (M = 4.53, SE = .23), t(16) = 3.747, p < .005, d = .83 (see Figure
2)5. Similarly, children displayed more happiness when they engaged in a non-costly act of
giving – provided the experimenter’s candy to Monkey (M = 4.88, SE = .20) – versus received
candies from the experimenter, t(17) = 2.415, p < .05, d = .46. Finally, consistent with previous
research conducted with toddlers under 2 years in Canada (Aknin et al., 2012), engaging in
costly giving was more rewarding than engaging in non-costly giving: children displayed more
happiness when giving their own candy to Monkey than when giving an identical candy provided
by the experimenter to Monkey, t(15) = 2.511, p < .03, d = .30.
We examined prosocial tendencies by looking at what children chose to do with the final
candy offered at the end of the experimental session. Two-thirds (13 out of 20) of children opted
to give the final candy to Monkey while one-third (7 out of 20) children took the candy for
themselves. Although more children selected the prosocial option, choosing to give the last
candy away rather than keep it for themselves, this difference was not significant; X2 = 1.800, p
= .18 (two-tailed). The tendency to give the final candy to Monkey was not associated with age
or sex (rs < .15, ps > .35).
To explore whether these findings are similar in direction and magnitude to previous
research conducted in North America, we combined data from the current investigation with
previous research conducted by Aknin and colleagues (2012) into one data set and ran repeated
measures analyses using sample as a between subjects variable. Analyses revealed that children
in both samples were rated as happier after sharing treats – either their own, F(1,35) = 27.465, p
< .001 (two-tailed), partial ƞ2 = .44, or the experimenter’s, F(1,36) = 7.010, p < .02 (two-tailed),
partial ƞ2 = .16 – than they were receiving treats themselves, with no interaction by sample, Fs <
1.1, ps > .30. Moreover, children in both samples were rated as happier after engaging in costly
giving than non-costly giving, F(1,34) = 31.373, p < .001 (two-tailed), partial ƞ2 = .48. The only
significant difference between the happiness ratings between past research in Canada and the
current research in Vanuatu was that Canadian children displayed larger happiness boosts for
costly versus non-costly giving than did Vanuatuan children, as captured by the significant
interaction between costly vs. non-costly giving and sample, F(1,34) = 8.218, p < .01 (two-
tailed), partial ƞ2 = .20. This difference could be due to any number of differences between the
two studies, including the difference in age of participants, culture, and the desirability of
resource that children were asked to share. Nonetheless, children in both investigations displayed
more happiness when giving than receiving and when engaging in costly (vs. non-costly) giving,
suggesting that the emotional rewards of prosocial behavior can be detected in children from
The findings reported here suggest that both adults and young children from an isolated,
rural village in Vanuatu experience greater emotional rewards from engaging in an act of
generosity than from engaging in an act that benefits themselves. Specifically, in Study 1 we
found that similarly to adults previously tested in North America and elsewhere, adults in
Vanuatu reported greater happiness after using money to purchase a gift for someone else than
after using money to purchase a gift for themselves. Notably, the emotional benefits of giving
were detectable despite the fact that adults from Vanuatu were gifting a relatively rare and
desirable commodity equivalent to a day’s wage. Similarly, in Study 2, young children displayed
more happiness when giving a candy away than when receiving candies themselves. Critically,
the emotional rewards of giving were greater when children gave their own candy away as
opposed to an identical treat that did not belong to them.
These findings support the possibility that the emotional rewards of giving may be a
universal feature of human behavior. Although detecting the warm glow of giving in Vanuatu
does not provide conclusive evidence for a psychological universal, these findings add to the
growing body of research demonstrating the emotional benefits of prosocial behavior can be
detected in various countries and cultures throughout the world (Aknin et al., 2013). These
findings also beg the question of what ontogenetic and/or phylogenetic mechanisms support the
link between prosociality and positive affect. First, it is possible that humans everywhere
socialize their offspring to be helpful members of society and, in turn, children learn to apply
these same principles of generosity to others (Eisenberg & Mussen, 1989) via norms of
prosociality (Silk & House, 2011). Since very young children in Vanuatu and North America
already demonstrate these effects, this socialization process may take place extremely early in
human development (e.g., Dahl, in press). Alternatively or in addition, it is possible that humans
evolved to find giving to others rewarding. Several evolutionary theorists have argued that
cooperation has allowed humans to thrive (Darwin, 1871/1982; Henrich & Henrich, 2006;
Tomasello, 2009; Wilson, 1975); perhaps positive emotion serves as a proximate mechanism for
promoting cooperation by attenuating the sting of engaging in costly prosocial acts. These
mechanisms are not mutually exclusive: it may be that universal socialization processes serve to
accentuate an evolved mechanism.
Although we were able to conduct two controlled experiments in a remote, rural, non-
Western village, both studies have limitations. For instance, in Study 1, materials were
administered verbally and as such the local experimenter was not blind to condition, making it
possible that the experimenter swayed participants’ emotional responses. We argue that this is
unlikely because the local experimenter followed a script and was not informed of the
experimental hypothesis. In addition, the number of participants in Study 1 is on the low end of
acceptable sample sizes (Simmons, Nelson & Simonsohn, 2011). We argue that this concern is
minimized for at least two reasons. First, the large effect size detected here is consistent with
previous research (Aknin et al., 2013), which suggests that our results reflect a robust
relationship between generous spending and well-being. Second, all adults in the village
available during the testing period were included in our examination, meaning that we did not
artificially restrict our sample or stop data collection upon attaining favorable results. As others
have pointed out (Norenzayan & Heine, 2005), studying behavior in remote locations may
require moderation of some research standards, including large samples, in service of the larger
search for generalizability that helps to put theories of human behavior on more solid ground.
An alternative explanation for the findings from Study 1 is that participants in the
personal spending condition may have worried about receiving negative evaluations and
treatment from fellow villagers after receiving candy for themselves, whereas participants in the
prosocial spending condition may have anticipated social praise. While this alternative
explanation is possible, previous work suggests that even anonymous prosocial behavior can be
emotionally rewarding, suggesting that happiness does not solely result from anticipating praise
(Aknin et al., 2013; 2014). In addition, we did not detect differences in the two negative affect
items - anger and sadness - between conditions. If participants in the personal condition were
concerned about the social consequences of receiving an item that benefitted themselves, this
likely would have manifested in a between-condition difference in one or both of these items. As
such, we argue that our results reflect the benefits of engaging in prosocial behavior, rather than
anticipating praise or sanction.
In addition, it is important to note that we did not follow adults in Study 1 as they left the
testing site to confirm that candy was distributed in line with either their assigned personal or
prosocial directions. While this methodological feature is consistent with previous work (e.g.,
Aknin, Dunn, Sandstrom & Norton, 2013; Dunn et al., 2008), several outcomes suggest that
participants adhered to instructions. First, although not asked or required, many participants in
the personal condition decided to eat some or all of their candy while in the testing room,
confirming the candy was for themselves. Second, several of the participants in the prosocial
condition spontaneously mentioned who they would give the candy to, with the most frequent
targets being their immediate family, such as their spouse or children. Finally, had participants
decided to ignore their assigned purchasing directions, we should not have been able to detect the
predicted differences in well-being between across conditions.
In interpreting the results of Study 2, it is important to note that it is unclear how children
in Vanuatu understood their interaction with our puppet recipient. Although we opted to use a
puppet to maximize methodological similarity with previous research performed in North
America (e.g., Aknin et al., 2012) in which most children are very familiar with inanimate toys
and willing to treat puppets as interaction partners (Rakoczy & Tomasello, 2006), puppets do not
exist in rural Tanna. Therefore, it is possible that children’s reactions were influenced by this
novel experience. However, the data suggest that extreme responses due to novelty were not
solely responsible for the observed results. Specifically, if seeing or interacting with the puppet
had led children to become either very afraid or very excited, we would have expected either
floor or ceiling effects in happiness ratings across all phases of the study that included
interactions with monkey, as opposed to the within-participant phase differences in happiness
that were observed. Indeed, that children in Vanuatu and Canada showed similar responses to
giving and receiving treats despite differences in the novelty of the interaction suggests there is a
robust relationship between happiness and giving.
On a related note, it is possible that children were particularly excited when Monkey
“ate” his treats, and that this led children to display happiness, as opposed to the act of sharing
itself. While it is possible that actively giving treats led children to smile more than did passively
receiving treats this does not explain why children smiled more when giving identical treats to
Monkey from their own (costly giving) and the experimenter’s (non-costly giving) supply.
Indeed, that children smile more when engaging in costly versus non-costly giving suggests that
it is prosocial behavior in particular, and not (for example) a social interaction, seeing Monkey
express happiness while eating a candy, nor the act of moving a candy that drives children’s
positive responses following prosociality.
Due to the wide age difference between participants in Studies 1 and 2, we assessed the
happiness consequences of prosocial behavior using quite different methodologies. Consistent
with a large number of studies exploring predictors of well-being (e.g., Emmons & McCullough,
2003; Lyubomirsky, Sheldon & Schkade, 2005; Nelson, Kushlev, English, Dunn &
Lyubomirsky, 2013), adults’ happiness was assessed via self-report, whereas children’s’ facial
expressions were captured on videotape and later coded for happiness by a team of trained
coders. Although we assumed that both measures assess the same underlying happiness
construct, they may not do so in equivalent ways. For instance, although classic research on
emotion demonstrates that people tend to smile when they are happy (Ekman, Friesen & Ancoli,
1980), smiling is not the only indicator of happiness (i.e. people can be happy when they are not
smiling), nor is happiness the only reason people smile (e.g., nervous laughter; see Ekman &
Friesen, 1982). That said, past research has demonstrated that naïve coder ratings correlate
highly (r > .95) with validated measures of emotion coding, including Baby FACS (Oster, 1995),
suggesting that coders could have been able to accurately assess children’s happiness.
While adults in the prosocial condition were able to select their beneficiary, children in
Study 2 were repeatedly asked to share with an unfamiliar puppet. To the extent that children
believed that Monkey was an independent being, these findings demonstrate the emotional
rewards of giving to an unknown target who may not reciprocate in the future. However, if
children believed that they were giving treats to the female experimenter, participants may have
enjoyed giving because they thought the researcher would reciprocate their kind act in the future.
Interestingly, however, expectations of reciprocation do not appear to fully explain the emotional
rewards of giving (Aknin et al., 2013; 2014), suggesting that even if the latter were true, giving
treats should still be rewarding if the recipient is unlikely to return the favor.
More broadly, this cross-cultural investigation adds to the larger initiative to consider
whether findings initially demonstrated within WEIRD societies replicate in non-WEIRD
populations (Henrich et al., 2010; Norenzayan & Heine, 2005). Despite the numerous differences
between the initial Canadian participants and those assessed here, the results were quite similar,
suggesting not only that the emotional rewards of generosity may be shared by humans around
the world, but that initial investigations conducted in WEIRD samples can unearth psychological
universals. Indeed, the larger message is not that investigations conducted in WEIRD samples
cannot reveal phenomena that may be universal to all humans, but rather that claims of
universality should be explored with cross-cultural investigation, such as this examination
conducted in Vanuatu.
In conclusion, this work adds to the growing body of research demonstrating the hedonic
rewards of prosocial behavior. Detecting the emotional payoffs of generosity (vs. self gain) in
both an adult and child sample from a remote, non-Western society supports the claim that
humans around the world find giving rewarding and provides firmer grounds for conclusions of
Aknin, L. B., Barrington-Leigh, C. P., Dunn, E. W., Helliwell, J. F., Burns, J., Biswas-Diener,
R., Kemeza, I., Nyende, P., Ashton-James, C. E., & Norton, M. I. (2013). Prosocial
spending and well-being: Cross-cultural evidence for a psychological universal. Journal
of Personality and Social Psychology, 104, 635-652.
Aknin, L. B., Dunn, E. W., Sandstrom, G. M., & Norton, M. I. (2013). Does social connection
turn good deeds into good feelings?: The value of putting the "social" in prosocial
spending. International Journal of Happiness and Development, 1, 155-171.
Aknin, L.B. Fleerackers, A.L. & Hamlin, J. K. (2014). Can third-party observers detect the
emotional rewards of generous spending? Journal of Positive Psychology, 9(3), 198-203.
Aknin, L. B., Hamlin, J.K. & Dunn, E. W. (2012). Giving leads to happiness in young children.
PLoS ONE, 7(6), e39211.
Benabou, R. & Tirole, J. (2006). Incentives and prosocial behavior. The American Economic
Review, 96, 1652-1678.
Borgonovi, F. (2008). Doing well by doing good. The relationship between formal volunteering
and self-reported health and happiness. Social Sciences & Medicine, 66, 2321-2334.
Dahl, A. (in press). The developing social context of infant helping in two U.S. samples. Child
Darwin, C. (1871/1982). The descent of man, and selection in relation to sex. Princeton, NJ:
Princeton University Press.
Dunn, E. W., Aknin, L. B., & Norton, M. I. (2008). Spending money on others promotes
happiness. Science, 319, 1687-1688.
Eisenberg, N., & Mussen, P. H. (Eds.). (1989). The roots of prosocial behavior in children.
Cambridge, MA: Cambridge University Press.
Ekman, P., & Friesen, W. V. (1982). Felt, false, and miserable smiles. Journal of Nonverbal
Behavior, 6(4), 238-252.
Ekman, P., Freisen, W. V., & Ancoli, S. (1980). Facial signs of emotional experience. Journal of
Personality and Social Psychology, 39(6), 1125-1134.
Emmons, R. A., & McCullough, M. E. (2003). Counting blessings versus burdens: An
experimental investigation of gratitude and subjective well-being in daily life. Journal of
Personality and Social Psychology, 84(2), 377- 389.
Hamilton, W. D. (1963). The evolution of altruistic behavior. The American Naturalist, 97, 354-
Hepach, R., Vaish, A., & Tomasello, M. (2012). Young children are intrinsically motivated to
see others helped. Psychological Science, 23, 967-972.
Henrich, J., Boyd, R., Bowles, S., Camerer, C., Fehr, E., Gintis, H., & McElreath, R. (2001). In
search of homo economicus: Behavioral experiments in 15 small-scale societies. American
Economic Review, 91, 73-78.
Henrich, J., Ensimger, J., McElreath, R., Barr, A., Barrett, C., Bolyanatz, A., Cardenas, J.C.,
Gurven, M., Gwako, E., Henrich, N., Lesorogol, C., Marlowe, F., Tracer, D., & Ziker, J.
(2010). Markets, religion, community size, and the evolution of fairness and punishment.
Science, 327, 1480-1484.
Henrich, J., Heine, S.J., & Norenzayan, A. (2010a). The weirdest people in the world?
Behavioural and Brain Sciences, 1-75.
Henrich, J., Heine, S., & Norenzayan, A. (2010b). Most people are not WEIRD. Nature, 466, 29.
Henrich, J., & Henrich, N. (2006). Culture, evolution and the puzzle of human cooperation.
Cognitive Systems Research, 7, 221-245.
Lyubomirsky, S., Sheldon, K. M., & Schkade, D. (2005). Pursuing happiness: The architecture
of sustainable change. Review of General Psychology, 9(2), 111-131.
McCauley, R., Henrich, J. (2006). Susceptibility to the Muller-Lyer Illusion, Theory-Neutral
Observation, and the diachronic penetrability of the Visual Input System. Philosophical
Psychology, 19, 1-23.
Nelson, S. K., Kushlev, K., English, T., Dunn, E. W., & Lyubomirsky, S. (2013). In defense of
parenthood: Children are associated with more joy than misery. Psychological
Science, 24(1), 3-10.
Norenzayan, A., & Heine, S. J. (2005). Psychological universals: What are they and how can we
know? Psychological Bulletin, 131(5), 763-784.
Nowak, M. (2006). Five rules for the evolution of cooperation. Science, 314, 1560-1563.
Oster, H. (2003). Emotion in the infant’s face. Annals of the New York Academy of Sciences,
Rakoczy, H., & Tomasello, M. (2006). Two‐year‐olds grasp the intentional structure of pretense
acts. Developmental Science, 9(6), 557-564.
Silk, J. B., & House, B. R. (2011). Evolutionary foundations of human prosocial
sentiments. Proceedings of the National Academy of Sciences, 108, 10910-1091.
Simmons, J. P., Nelson, L. D., & Simonsohn, U. (2011). False-positive psychology: Undisclosed
flexibility in data collection and analysis allows presenting anything as significant.
Psychological Science, 22(11), 1359-1366.
Tomasello, M. (2009). In B. Galef, & K. Laland (Eds.), The Question of Animal Culture (pp.
198-221). USA: Academic Press.
Tracy, J. L., & Robins, R. W. (2007a). Emerging insights into the nature and function of
pride. Current Directions in Psychological Science, 16, 147-150.
Tracy, J. L., & Robins, R. W. (2007b). The psychological structure of pride: A tale of two
facets. Journal of Personality and Social Psychology, 92, 506-525.
Trivers, R. L. (1971). The evolution of reciprocal altruism. The Quarterly Review of Biology, 46,
Warneken, F. & Tomasello, M. (2006). Altruistic helping in human infants and young
chimpanzees. Science, 311, 1301- 1303.
Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures
of positive and negative affect: The PANAS scales. Journal of Personality and Social
Psychology, 54, 1063-1070.
Weinstein, N., Ryan, R. M. (2010). When helping helps: Autonomous motivation for prosocial
behaviour and its influence on well-being for the helper and recipient. Journal of
Personality and Social Psychology, 98, 222-244.
Wilson, D.S. (1975). A theory of group selection. Proceedings of the National Academy of
Sciences, 72, 143-146.
1 The results of Study 1 remain the same when the three participants who opted to take the cash
voucher instead of buying candy for themselves are removed from analyses, t(21) = 2.550, p <
.02, d = 1.02.
2 Participants were also asked to report their current level of excitement using the same 11-step
ladder but this item was removed from the composite positive affect score because back-
translation of the term used in Bislama did not translate to “excitement” and because participant
ratings on this item did not correlate with other positive affect items. Adding this question to the
positive affect composite does not alter the findings of Study 1; participants purchasing treats for
others were significantly happier (M = 9.731, SE = .17) than participants purchasing treats for
themselves (M = 8.872, SE = .33), t(17.83) = 2.312, p = .033 (two-tailed), d = .91.
3 The age of six children in Vanuatu was estimated because parents did not have a copy of their
child’s birth certificate (if the child was born in a hospital) and/or could not remember their
child’s birthday. If this was the case, age was estimated based on peers and parents’ recollections
of their child’s birth in comparison to other village members.
4 When ratings from only the Canadian (context-free) coders are used, results are as follows:
Costly giving vs. receiving, t(16) = 2.725, p < .02, d = .65; Non-costly giving vs. receiving, t(18)
= 1.789, p = .091, d = .26; Costly giving vs. non-costly giving, t(16) = 2.148, p < .05, d = .38.
5A reduced sample size is reported when children did not complete the requested action (3
children did not give their own treat) or their facial expression could not be seen clearly (e.g.,
hand blocking their mouth or face moved out of video for several seconds).
Figure 1. Average positive affect reported by adults in the personal and prosocial spending
conditions in Study 1. Error bars display standard error of the mean.
Figure 2. Average happiness ratings as reported by coders across the five phases in Study 2.
Error bars display standard error of the mean.