The Oxytocin Receptor (OXTR) Contributes to Prosocial
Fund Allocations in the Dictator Game and the Social
Value Orientations Task
Salomon Israel1, Elad Lerer2, Idan Shalev3, Florina Uzefovsky1, Mathias Riebold2, Efrat Laiba2, Rachel
Bachner-Melman1, Anat Maril1, Gary Bornstein1,4, Ariel Knafo1, Richard P. Ebstein1,5*
1Department of Psychology, The Hebrew University of Jerusalem, Jerusalem, Israel, 2Department of Human Genetics, The Hebrew University of Jerusalem, Jerusalem,
Israel, 3Brain and Behavior Science, The Hebrew University of Jerusalem, Jerusalem, Israel, 4Center for the Study of Rationality and Interactive Decision Theory, Jerusalem,
Israel, 5S. Herzog Memorial Hospital, Jerusalem, Israel
Background: Economic games observe social decision making in the laboratory that involves real money payoffs. Previously
we have shown that allocation of funds in the Dictator Game (DG), a paradigm that illustrates costly altruistic behavior, is
partially determined by promoter-region repeat region variants in the arginine vasopressin 1a receptor gene (AVPR1a). In
the current investigation, the gene encoding the related oxytocin receptor (OXTR) was tested for association with the DG
and a related paradigm, the Social Values Orientation (SVO) task.
Methodology/Principal Findings: Association (101 male and 102 female students) using a robust-family based test
between 15 single tagging SNPs (htSNPs) across the OXTR was demonstrated with both the DG and SVO. Three htSNPs
across the gene region showed significant association with both of the two games. The most significant association was
observed with rs1042778 (p=0.001). Haplotype analysis also showed significant associations for both DG and SVO.
Following permutation test adjustment, significance was observed for 2–5 locus haplotypes (p,0.05). A second sample of
98 female subjects was subsequently and independently recruited to play the dictator game and was genotyped for the
three significant SNPs found in the first sample. The rs1042778 SNP was shown to be significant for the second sample as
well (p=0.004, Fisher’s exact test).
Conclusions: The demonstration that genetic polymorphisms for the OXTR are associated with human prosocial decision
making converges with a large body of animal research showing that oxytocin is an important social hormone across
vertebrates including Homo sapiens. Individual differences in prosocial behavior have been shown by twin studies to have a
substantial genetic basis and the current investigation demonstrates that common variants in the oxytocin receptor gene,
an important element of mammalian social circuitry, underlie such individual differences.
Citation: Israel S, Lerer E, Shalev I, Uzefovsky F, Riebold M, et al. (2009) The Oxytocin Receptor (OXTR) Contributes to Prosocial Fund Allocations in the Dictator
Game and the Social Value Orientations Task. PLoS ONE 4(5): e5535. doi:10.1371/journal.pone.0005535
Editor: Paul Zak, Claremont Graduate University, United States of America
Received June 24, 2008; Accepted April 10, 2009; Published May 20, 2009
Copyright: ? 2009 Israel et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This study was supported in part by grants from the Hebrew University (BINCA: GB, RPE, AK, AM), the Israel Science Foundation founded by the Israel
Academy of Sciences and Humanities (RPE, AK & GB) and the Israel National Institute for Psychobiology (AK). SI is the recipient of a doctoral fellowship awarded by
the Israel Ministry of Finance and the Ministry of Industry, Trade and Labor. The funders had no role in study design, data collection and analysis, decision to
publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: email@example.com
The nascent field of neuroeconomics has in a short time made
major strides in elucidating the neurobiological underpinnings of
social decision making. Neuroeconomics seeks to explain the brain
mechanisms underlying a broad range of human behaviors
exhibited not only by the marketplace but also by the matrix of
human relationships that constitute more complex community
interactions. An important contribution of neuroeconomics is the
demonstration that the classical economic assumption of self-
interest cannot explain the richness of human social behavior .
Economic games, which observe human decision making in the
laboratory and involve the benefits of real outcomes (real money)
and quantifiability, provide a more coherent framework to
benchmark social behavior. Combined with brain imaging and
pharmacological manipulations, economic games have provided
critical insights into the neural circuits which drive economic and
social decision cognition . In addition, these findings have
suggested an evolutionary basis for many of our affective and
cognitive biases, particularly other-regarding behavior.
In a seminal study  we have shown that the powerful
techniques of molecular genetics can also be employed to further
clarify the mechanisms by which the social brain formulates
economic decisions. We demonstrated that allocation of funds in
the Dictator Game (DG) were in part determined by length of the
arginine vasopressin 1a (AVPR1a) RS3 promoter region repeat
region. Additionally, the length of the RS3 repeat region was
correlated with increased amounts of AVPR1a mRNA in
PLoS ONE | www.plosone.org1 May 2009 | Volume 4 | Issue 5 | e5535
hippocampal post-mortem specimens. The DG is a simple one-
shot, two player game. The first player, or ‘‘Dictator’’, makes a
unilateral decision regarding the split of a fixed sum of money
between herself and a second player. The second player, or
‘‘Recipient’’, must accept the result. Fund allocation in the
Dictator Game is considered a real-life test of altruism since it
involves a ‘put your money where your mouth is’ decision where
giving to another is both costly and not subject to strategic
considerations of reciprocity; this is particularly the case when
‘‘Dictator’’ and ‘‘Recipient’’ identities are made anonymous .
Moreover, allocations in the DG underscore the universal nature
of human altruism since the canonical model - based on self-
interest – has been shown to fail in all societies studied .
Our first studies of the role of the AVPR1a gene in social
cognition exemplified by a simple laboratory paradigm converges
with a wide range of both animal and human experimentation
showing that arginine vasopressin (AVP) and the closely related
oxytocin (OT) are key neuropeptides that facilitate social
communication, affiliative behaviors and social cognition across
mammals [6,7] . Not only has the role of AVP and OT been
independently shown to enhance a range of social behaviors across
the class of Mammalia including humans, but the interaction
between these two social hormones has also been broadly
documented [8–10]. OT and AVP have overlapping functions
in part mediated by mutual actions on their respective receptors
[11–15], possibly based on receptor affinities . Moreover,
behavioral and tissue binding studies suggest that considerable
cross-communication is possible between the functional effects of
OT and AVP[14,17–19]. We hypothesized, based on our first
findings showing that the length of the AVPR1a RS3 promoter
region repeat predicted Dictator Game allocations , that the
OXTR gene would also likely modulate other-regarding behavior
in this simple laboratory game.
Intriguingly, intranasal application of OT has been shown to
increase trusting behavior in humans [20,21], identification of
others’ affective states  and amygdala responses to emotional
faces . Administration of OT also improves social cognition in
autism . In addition to their roles in enhancing social
behaviors, the AVPR1a and oxytocin receptor (OXTR) genes
have been shown by us [25,26] and others [27–30] to contribute to
autism spectrum disorders (ASD), a behavioral disorder charac-
terized by deficits in social communication and bonding. A recent
study by Zak et al., had subjects indicate transfer amounts for
both the DG and the Ultimatum game (where the recipient has the
opportunity to refuse the allocation, in which case both players
receive zero) after undergoing intranasal administration of OT.
Oxytocin significantly increased transfer amounts in the Ultima-
tum game as compared to placebo, whereas no effect in the DG
was observed. An effect the authors attributed to the increased
perspective taking in the ultimatum game, since in addition to
indicating transfer amounts allocators also had to take into
account the threat of punishment.
Genetic association studies for complex phenotypes such as
social behavior can be highly sensitive to measurement construct.
For example, a meta-analysis reviewing association for a serotonin
transporter gene polymorphism (5-HTTLPR) and anxiety related
traits demonstrated how choice of measurement scale may be a
critical factor in moderating phenotypic contours. To this end,
we also explored allocation of funds in a second, related paradigm,
Social Value Orientations (SVO). We selected this paradigm
since it addresses related, yet distinct aspects of prosocial behavior
than the Dictator Game. While the Dictator is a zero-sum game
with a fixed pie, the SVO enables individuals to allocate funds with
relatively small costs, tapping into a broad motivational orientation
that is largely independent of the costs of sharing. Social Value
Orientations (SVO), use a set of interdependent payouts to
categorize individuals based on two different types of other
regarding preferences: concern for the well-being of others, and
concern for equality. In SVO experiments, subjects are asked to
make choices regarding the distribution of points (which in the
present study had real monetary value) between themselves and an
‘other’. Consistent value orientations classify players as prosocial-
maximizing joint outcomes, individualistic- maximizing personal
outcomes or competitive-maximizing the difference between
individual and ‘other’ outcome. In making comparisons between
orientations, it is also common to group ‘competitive’ and
‘individualistic’ subjects into a general ‘pro-self’ cohort. We
hypothesized that positive associations for both high giving in the
Dictator Game and prosociality in the Social Value Orientations
would provide additional evidence for the robustness of SNP
association beyond a specific, singular measure. Altogether, we
tested association between 16 tagging SNPs (htSNPs) across the
entire OXTR gene region and individual distribution of endow-
ments in these two games. Our selection of all tagging SNPs listed
for OXTR in HapMap represents an optimal strategy for detecting
association across this gene region .
Fund allocation in Dictator Game and Social Values
Allocation amounts in the DG were dichotomized into low
(,25 NIS, or less than 50% of the pie) and high givers ($25
NIS) ; 25 NIS was the modal value in this distribution and
was used as the cutoff point to divide participants into low and
high allocators (see Figure 1). Altogether, 47% of the participants
were designated as high allocators. Based on their responses in
the SVO task, participants were categorized into one of three
groups: prosocial, proself (made up of both pro-individuals and
competitors) and undefined (participants that had less than six
responses consistent with any one category); 58% of participants
were categorized as prosocial, 36% as proself (which includes
1% competitive and 35% individualistic) and 6% as undefined.
Similar to dictator giving, prosocial responses also fell into a bi-
modal distribution with a plurality (38%) of participants selecting
9/9 prosocial responses and 24% selecting 0/9. There is
considerable similarity between individual allocation of funds in
the Dictator Game and the Social Values Orientation (SVO)
task (Figure 2, Pearson Chi-Square=28.18, DF=1 p,0.0001).
In the SVO prosocial group, 64.7% of respondents are also high
allocators ($25 NIS) in the Dictator Game. Among proself SVO
individuals, only 26.5% are high allocators in the Dictator
Association between OXTR htSNPs and fund allocation
All of the SNPs tested were in Hardy-Weinberg Equilibrium
(p.0.01). Of the 16 tagging SNPs, one of them, rs2139184, had a
very low rate of heterozygosity, and was not included in the
analysis. We next examined association using PBAT between 15
single tagging SNPs (htSNPs) across the OXTR (Figure 3) and
allocation of funds in the Dictator Game and the Social Values
Orientation task (Table 1), results are given using PBAT’s
dominant model, which assumes a dominant mode of inheritance.
Of the 15 htSNPs across the gene region, 8 htSNPs showed
nominal significant association (before correction for multiple
testing) with one or both of the game paradigms The most
significant association, which is also robust to Bonferroni
correction, was observed with rs1042778 for both the Dictator
OXTR and Social Behavior
PLoS ONE | www.plosone.org2 May 2009 | Volume 4 | Issue 5 | e5535
Game (p=0.001, p corrected 0.016) and SVO (p=0.002, p
corrected=0.031). There were also SNPs whose effects appeared
to be gender specific e.g. rs237897 (nominally significant in male
subjects) and rs9840864 (nominally significant in female subjects).
6 of the 8 htSNPs showed a consistent association for both
Dictator Game and SVO allocations. For the three most
significant SNPs (rs1042778, rs2268490, and rs237887), an
allele-based comparison of the number of prosocial responses for
the SVO is displayed in Figure 4.
Since haplotypes are often more informative than individual
SNPs , haplotype analysis was carried out. We first used
UNPHASED to test associations between 2–8 locus haplotypes
Figure 1. Distribution of Dictator allocations for student and mother samples. Allocation sums by participants grouped in
5<$1.17). For both samples, the modal value of 25 was used as the cutoff point to divide participants into low and high allocators.
Figure 2. Cross tabulation of Dictator and SVO allocation of funds. Cross tabulation of Dictator giving with Social Value Orientation (SVO).
High givers in the DG were significantly more likely to maintain prosocial value orientations as compared to low givers.
OXTR and Social Behavior
PLoS ONE | www.plosone.org3 May 2009 | Volume 4 | Issue 5 | e5535
since this program features a convenient, automatic sliding
window approach that minimizes multiple testing by only testing
adjacent SNPs (see FIGURE 5, only significant haplotypes are
shown). Haplotype analysis using UNPHASED calculates overall
global p values that corrects for multiple testing of individual
haplotypes. Global-P-values are determined in UNPHASED to
compare the maximum observed SNP score in the observed data
against each maximum of the SNP score in the permuted data set.
Figure 3. Visual schematic of the oxytocin receptor. Schematic representation of chromosome 3 and the oxytocin receptor (OXTR) gene with
the location of the 16 tagging single-nucleotide polymorphisms (SNPs).
Table 1. Association between OXTR SNPs and allocation of funds in the Dictator Game (high versus low allocators) and SVO
RS number SNP
Minor allele frequency in
sample populationDG OverallDG Males DG Females SVO OverallSVO Males SVO Females
rs4564970C/G C : 0.13
rs237897A/G A : 0.360.0052 0.0019
rs13316193C/TC : 0.27 0.0508
rs237889C/T T : 0.39
rs2254298A/G A : 0.190.04550.0548
rs2268494A/T A : 0.11
rs4686301C/T T : 0.30
rs9840864C/T C : 0.350.0184 0.0357
rs237888 C/T C : 0.08
rs2268490C/T T : 0.22 0.01110.0265 0.0398 0.00830.0295
rs237887A/G G : 0.490.0048 0.00680.02500.0144
rs237885G/T G : 0.480.0194 0.0163
rs1042778G/T T : 0.290.0011 0.00460.0021 0.0048
DG=Dictator Game, categorized as high (.=25) and low (,25) allocators.
SVO=Social Value Orientation, categorized as prosocial vs proself orientations.
OXTR and Social Behavior
PLoS ONE | www.plosone.org4 May 2009 | Volume 4 | Issue 5 | e5535
Figure 4. Comparison of SVO Prosocial Responses by OXTR SNPs. Average number of prosocial responses in the Social Value Orientation
task categorized by the three most significant OXTR SNPs. A family-based analysis for the three SNPs showed significant association with prosocial
value orientations (PBAT dominant model: rs1042778 p=0.001, rs2268490 p=0.011, and rs237887 p=0.005).
Figure 5. Association between OXTR haplotypes and allocation of funds in the Dictator Game (high versus low allocators) and Social
Value Orientations. Haplotypes are indicated in outlined blocks.1UNPHASED global p-value, additive model2UNPHASED permutation test
for the haplotype block3PBAT multi-variate statistic, dominant model.
OXTR and Social Behavior
PLoS ONE | www.plosone.org5 May 2009 | Volume 4 | Issue 5 | e5535
For Dictator Giving, significant associations were observed for 2–8
locus haplotypes (global p,0.05), many of which were also
significant for the SVO.
We then used the stringent permutation test option provided in
the UNPHASED software package to further void spurious results
and correct for multiple testing. Following permutation testing,
significance was observed for 2–5 locus haplotypes (p,0.05) and
approached significance for the 7 and 8 locus haplotypes
(permutation test p,0.1). See FIGURE 5, ‘permuted’ column.
Next, to rigorously test the robustness of the associations
observed with UNPHASED we employed the multivariate statistic
(FBAT-GEE) utilized in the PBAT software package for the
haplotype blocks assessed in UNPHASED. Simulation experi-
ments involving quantitative traits show that the multivariate
FBAT clearly outperforms permutation tests and individual
FBATs with corrections for multiple testing. Lange et al. 
has proposed a natural extension of the FBAT statistics which are
based on the general likelihood-score test approach  to
multivariate phenotypes based on GEE scores. Without making
any assumptions about the distributions of the phenotypes, the
multivariate FBAT test allows us to test the null hypothesis that the
marker locus is not linked to any genetic locus that has an
influence on the selected phenotypes.
When several phenotypes are selected, all phenotypes are tested
simultaneously using FBAT-GEE. For FBAT-GEE, both binary
and continuous phenotypes can be entered. The program can
combine phenotypes with different distributions (e.g. continuous
and ordinal). Generally, the FBAT-GEE statistic can handle a
moderate amount of any type of multivariate data, including
groups of dichotomous phenotypes. This is an attractive
methodology, given the increasingly complex traits, such as
prosocial behavior, which we have investigated. Notably, all
significant (global p,0.05) haplotype blocks that withstood
permutation tests in UNPHASED for Dictator Giving and the
SVO were also significantly associated using the PBAT multivar-
iate statistic which considers the DG and SVO simultaneously
(p,0.05) (see FIGURE 5, FBAT-GEE column) strongly confirm-
ing the results observed in UNPHASED. Notably, the 4 locus
significant for Dictator Giving (FBAT-GEE=0.005), SVO
(FBAT-GEE=0.028) and when both Dictator Giving and SVO
are jointly considered (p=0.038).
After completion of the first study discussed above, a second
group of 98 female subjects participating in a separate protocol in
the psychology department became available, and were recruited
to play the dictator game. To limit the conundrum of multiple
testing, we used a more stringent criteria for SNP selection in the
second sample. Subjects were genotyped solely for the three
SNPs found to be significant in both the SVO and Dictator in the
first sample (rs2268490, rs237887 and rs1042778).
Dictator giving was considerably higher in the second sample
(One-way ANOVA F1,296=9.934 p=0.002), with giving amounts
4.97 higher in the second sample (see Figure 1).
Using the same dominant model of inheritance as in the first
sample, a significant association was observed with the G allele of
the rs1042778 SNP and high/low giving in the (p=0.004, Fisher’s
exact test). There was also a trend for mean differences in overall
dictator giving (F1,96p=0.07), with TT genotypes allocating on
average 18.3 shekels to the ‘other’ and GG+GT genotypes
allocating on average 25 shekels to the other, an effect size of
approximately 0.5 SD. Notably, SNP rs1042778 was the most
significant SNP associated with DG and SVO in the first sample
and this association with DG allocations was replicated in the
second, independently recruited group of subjects.
This is the second report in our ongoing investigation of the role
of common genetic polymorphisms in shaping human altruism as
measured in experimental economic games. We have employed a
candidate gene approach towards understanding the molecular
architecture of altruism and focused our first studies on the AVP –
OT pathway, nonapeptides that have been elegantly shown in
lower mammals to profoundly impact on so-called ‘social circuits’
. Neural networks employing AVP and OT uniquely influence
both social recognition and the expression of appropriate social
responses. It is increasingly evident from our own studies [41–
45] and those of other groups [21,46–54] that these two ‘social
hormones’ also contribute to a range of normal human social
We had previously shown that the promoter region repeat
length of the AVPR1a RS3 microsatellite polymorphism contrib-
utes to individual differences in Dictator Game allocation of real
money and correlates with hippocampal mRNA levels . We
now extend these findings and demonstrate that a close,
phylogenetically related receptor, OXTR, also impacts on money
allocations in laboratory-based economic games that model real
life prosocial versus proself behaviors. Additionally, our finding of
a significant association for both the Dictator game and SVO,
decision tasks that are significantly correlated with one another,
provides supporting evidence for the robustness of the current
A previous study by Zak et al.  noted a lack of effect of
intranasal administration of oxytocin on DG giving, however, in
the case of the Ultimatum Game, where participants also have to
consider the second player’s perspective, oxytocin increased
generosity levels. The lack of convergence between the results
obtained by sniffing oxytocin and the currently reported results
using a genetic association study to establish the neurochemical
pathways underlying Dictator Game behavior appear to be at
odds. However, it should be considered that the pharmacological
approach used by Zak et al  and the molecular genetic
approach used in the current research represent very different
experimental modalities. Perhaps the resolution of these apparent
discrepant results awaits a combined pharmacogenetic strategy
where subjects (administered oxytocin or placebo) play in the
Dictator Game and are stratified by genotype.
We have reported association OXTR htSNPs previously in an
association study of autism spectrum disorder (ASD) . Notably,
three of the SNPs (rs2268490, rs237887 and rs1042778) associated
most robustly with the two economic decision tasks reported in the
current investigation were also significantly associated with
ASD. One of the triad of core deficits in ASD is social
communication, and it makes sense that a gene contributing to
prosocial behavior in nonclinical populations may also contribute
to deficits in social behavior in clinical groups such as ASD. Of
particular interest is rs1042778, which is significant both in our
prior autism study, the 1stsample for both Dictator and SVO
and in the second independently recruited DG sample. Interest-
ingly, this SNP is located in the 39 untranslated region. 39
untranslated regions (39 UTRs) often contain several regulatory
elements that govern the spatial and temporal expression of an
mRNA.  Furthermore, several translational regulators are
highly conserved and seem to control many different mRNAs. We
suggest, based on our observation that this SNP shows significant
association with altruism and autism, the possibility that rs1042778
OXTR and Social Behavior
PLoS ONE | www.plosone.org6May 2009 | Volume 4 | Issue 5 | e5535
may play an important regulatory role in OXTR transcription and
A regulatory role for the intronic (3rdintron) htSNPs (or SNPs in
LD with these htSNPs) analyzed in this study should also be
considered. Suggestion of a functional character of SNPs located
in the 3rdintron is based on the finding of a specific motif of about
10–15 nucleotides close to the middle of the 3rdintron that
specifically binds nuclear proteins correlating with the down-
regulated state of the gene . Indeed, Mizumoto et al 
suggested that this intronic element is specifically binding nuclear
protein(s) associated with a suppression of OXTR gene activity. In
addition, to this intronic regulation site, a number of 59 upstream
regulatory sites have been identified for the OXTR promoter
region . In fact, tagging SNPs analyzed in the current
investigation may be detecting signals from other regulatory
elements along the OXTR extended gene region.
Although twin studies underscore the importance of the genetic
contribution to prosocial behavior [58,59], untangling the
contribution of specific polymorphisms to this and other complex
traits is difficult. Complex traits including human altruism are
molded by both environment and multiple genes and such
phenotypes pose special challenges for genetic analysis due to
gene-gene and gene-environment interactions, genetic heteroge-
neity, low penetrance, and limited statistical power . That
variations in these two receptor genes, OXTR and AVPR1a, jointly
contribute to a complex human behavioral phenotype is probably
the rule and not the exception. We are interested in testing other
elements of the OT-AVP system that we conjecture are also worth
examining within the context of a broad candidate gene-neural
network approach towards understanding the fine contours of how
these two neurohormones mold social behavior in our own species.
Materials and Methods
The participants in our ongoing studies of personality, primarily
college students and their families, were recruited by word of
mouth and advertisements on campus, as previously described
[45,61]. They were contacted by telephone and asked to
participate in online versions of economic games. 93% of
contacted individuals agreed to participate. The sample consisted
of 102 men and 101 women whose average age was 26.00
(SD=3.73). Parents of the student sample were also genotyped to
allow for a family based analysis, however parents did not
participate in the games.
A second sample of 98 subjects, all mothers, average age 34.47
(SD=4.76) were recruited to play the Dictator Game. Mothers
were part of an ongoing twin study examining the heritability of
prosocial behavior; time constraints of their participation
precluded playing in the SVO.
All subjects gave informed written consent and the genetic study
was approved by the local University and Hospital Internal
Review Board and the Israeli Ministry of Health (Genetics
In the student sample, subjects were provided with a login and
password to an online site and upon entering were provided with
instructions for the DG and SVO. Ordering of DG and SVO was
randomized and there was no effect of game order on average
giving amount or value orientation. In the second sample, subjects
played the dictator game at the Hebrew University Social
Development lab as part of an ongoing study on genetic and
environmental influences on prosocial behavior in early childhood.
Game instructions were identical for both groups.
Dictator Game instructions to participants were as follows:
In this task you will be asked to make a decision which can
earn you some money. The task will take place in pairs, in
which case one of the participants will be Player A, and the
other participant will be Player B. The assignment of Player
A and Player B will be done randomly by the computer. You
do not know the other participant and will not knowingly
meet them in the future.
In this task there are 50 points that Player A must decide
how to distribute between himself and Player B. That is,
Player A decides how many points he will keep for himself
and how many points Player B will receive. For every point
that he keeps for himself, Player A will receive 1
every point that he gives to Player ‘B’, Player ‘B’ will receive
This completes the task.
Please press the button to continue.
Upon pressing the button, subjects were notified if they were
Player ‘A’ or Player ‘B’. In reality, all subjects were selected as
Player ‘A’. Upon completing the experiment, subjects were paid
via check to their home address.
Social Value Orientations Task
A measure developed by Van Lange and colleagues (23) to
measure social value orientations by using a series of decomposed
games, involving choices among combinations of outcomes for
oneself and for another person, was adapted in this study. In the
current study, instead of assigning hypothetical points to the
participant, each point had real monetary value, as described to
participants in the following instructions:
In this task you will be partnered with a random additional
person whom which we will call the ‘‘other’’. You do not
know this person and will not knowingly meet them in the
future. You will be asked to make a number of decisions in
which you will be required to select between three options
‘A’ ‘B’ or ‘C’. Your selection will provide points for you and
for the ‘‘other’’. At the completion of the task we will sum
the number of points you have accumulated for yourself and
Table 2. Social Value Orientation Example Task
Other Receives100500 300
Based on this example, if you were to select:
‘A’ you would receive 500 points=5
‘B’ you would receive 500 points=5
‘C’ you would receive 550 points=5.5
and the ‘‘other’’ would receive 100
and the ‘‘other’’ would receive 500
and the ‘‘other’’ would receive 300
OXTR and Social Behavior
PLoS ONE | www.plosone.org7 May 2009 | Volume 4 | Issue 5 | e5535
for the other and will pay you 1
have collected. Similarly, we will pay the ‘‘other’’ 1
each 100 points as well.
See Table 2 for an example of the task.
for each 100 points you
As you can see, your decisions affect both the number of
points you receive and the number of points the ‘‘other’’
receives. Before you begin your selections, realize that there
are no right or wrong answers. Pick the selections that you
prefer for whatever reason. As such, remember that the
points have a real money value: 100 points are equal to 1
Payment will transpire according to the aforementioned
rules and will be transferred to via check to you and the
‘‘other’’ a few weeks after completion of the experiment.
The ‘‘other’’ does not have the opportunity to identify you.
Game results are recorded in a way to assure anonymity.
You will only have one opportunity to make your selections.
Please weigh your decisions carefully; you will not have an
opportunity to go back.
The OXTR was genotyped as previously described . DNA
was extracted by Master Pure kit (Epicentre, Madison, WI, USA).
Genotyping of all SNPs was performed using the SNaPshot
Method (Applied BioSystems, Foster City, CA, USA) which relies
upon the extension of a primer immediately adjacent to the SNP
using fluorescently labeled ddNTPs. The fluorescently labeled
extension primers can then be visualized by electrophoresis on a
capillary ABI PRISM 310 automated sequencer. OXTR SNP
regions were amplified using the pairs of primers listed in Table 3.
PCR cycling conditions in the SNaPshot Method were as follows:
samples were initially heated at 94uC for 5 min followed by 35
cycles of 94uC (30 s), 55uC (30 s) 72uC (90 s) and a final extension
step of 72uC for 5 min. After the first PCR cycle, the PCR product
was cleaned with ExoSAP for 37uC for 30 min and then at 80uC
for 15 min. The conditions for the second PCR were as follows:
96uC (10 s), 50uC (5 s) and 60uC (30 s) for 25 cycles. The second
PCR product was cleaned using shrimp alkaline phosphatase
(SAP) initially at 37uC for 1 h followed by 72uC for 15 min.
All tagging SNPs listed in HapMap and selected using the
Haploview program http://www.broad.mit.edu/mpg/haploview/
were genotyped. Most association studies rely on the use of
surrogate markers, single-nucleotide polymorphisms (SNPs) being
the most suitable due to their abundance and ease of scoring (see
discussion in . SNP marker selection is aimed to increase the
chances that at least one typed SNP would be in linkage
disequilibrium (LD) with the disease causative variant, while at
the same time controlling the cost of the study in terms of the
number of markers genotyped and sampled. Empirical studies
reporting block-like segments in the genome with high LD and low
haplotype diversity have motivated a marker selection strategy
whereby subsets of SNPs that ‘tag’ the common haplotypes of a
region are picked for genotyping, avoiding typing redundant SNPs.
We tested for association between OXTR polymorphisms and
money allocations in both the Dictator Game and Social Values
Orientation using the ETDT, a logistic based variant of the
transmission disequilibrium test (TDT) which assesses for associa-
tion (and linkage) without the confounding effect of population
stratification. The TDT, in its simplest version, compares, for one
allele, the number of times this allele is transmitted to the number of
times where this allele is not-transmitted to an affected offspring.
Notethat only heterozygous parents areinformative. Thisapproach
can be extended to haplotypes. For the second sample, a Fisher’s
exact test was used to examine the significance of the association
(contingency) between genotypes and high/low giving amounts.
The various tests we used are implemented in the program
Table 3. Primers used for genotyping of OXTR tagged SNPs
PCR I (59 to 39) ForwardPCR I (59 to 39) ReverseSNP PCR II (primer extension 59 to 39)
TTGTAATTCTAATGCACCCTCA AAAGTTGAAAGATCCAAAGAGTAAArs237887 AGCTTTGCAATGAGGTAG
ATTGTCTCGGTGCCATTTGT CCATCAGAAAGAATAAAATAGGAArs9872310 TCAATTGACCGTAAGTAT
AGAGCTGCCTGCCAAATG GACCCCCGGAGAAGGTGCTrs9840864 (T)12CAAATGTGCAAAGGCCAG
AAGGGAGGGTCAAAATCAGC GTTAGACGGGGAAGGACCAG rs4564970(T)18CCCTCAGCATATCCACCT
TGGGTTCAGGGTGGTAGAAG AGGCTGTGCTGGCATAAGTGrs1042778 (T)12TGAAGCCACCCCAAGGAG
OXTR and Social Behavior
PLoS ONE | www.plosone.org8May 2009 | Volume 4 | Issue 5 | e5535
unphased/) , a suite of programs for association analysis of
multilocus haplotypes from unphased genotype data. Permutation
test correction was performed using 1000 iterations (random
permutations) and applied to correction of global P-values (the P-
value as an overall test of each block of haplotypes). In nuclear
families, the permutations are generated by randomizing the
transmission status of the parental haplotypes. In unrelated
subjects, the trait values are randomly shuffled between subjects.
In each permutation, the minimum P-value is compared to the
minimum P-value over all the analyses in the original data. This
allows for multiple- testing corrections over all tests performed in a
run. This procedure corrects for multiple testing but accounts for
correlation between markers, hence is less conservative than a
Bonferroni correction, which is appropriate for independent tests
such as unlinked markers.
PBAT (Helix Tree)
The PBAT software package provides a unique set of tools for
complex family-based association analysis at a genome-wide level
[65,66]. PBAT can handle nuclear families with missing parental
genotypes, extended pedigrees with missing genotypic information,
analysis of single nucleotide polymorphisms (SNPs), haplotype
analysis, quantitative traits, multivariate/longitudinal data and
time to onset phenotypes. The data analysis can be adjusted for
covariates and gene/environment interactions. Haplotype-based
features include sliding windows and the reconstruction of the
haplotypes of the probands.
The family-based association test (FBAT) statistic  compris-
es a linear combination of observed offspring genotypes and traits.
We used the HelixTree version of PBAT (http://www.goldenhelix.
com/pharmhelixtreefeatures.html). The test statistic is defined as:
S=gijTijXij, where Xijdenotes a coding of the marker genotype of
the jthoffspring in family i. The coded trait of the jthoffspring in the
ithfamily is defined by Tij. In general, the trait will be mean-
centered. The test statistic, a score function, in large samples, is
defined as: Z=gijTij(Xij2E(Xij))/Var(S)1/2,N(0,1), where E(Xij) is
the expected value of the coding function for the offspring genotype,
conditional on the parental genotype and parental/offspring
phenotypes, which are assumed to follow Mendelian segregation
under the null hypothesis. This methodology extends readily to
scenarios in which the parental genotypes are not known, using the
approach of Rabinowitz and Laird . The FBAT statistic has an
additional advantage in that it is robust against population
admixture and stratification .
Conceived and designed the experiments: SI AM GB AK RPE. Performed
the experiments: SI EL IS FU MR EL RBM RPE. Analyzed the data: SI
EL IS FU MR EL RBM GB AK RPE. Contributed reagents/materials/
analysis tools: RPE. Wrote the paper: SI GB AK RPE.
1. Loewenstein G, Rick S, Cohen J (2007) Neuroeconomics. Annu Rev Psychol.
2. Fehr E, Camerer CF (2007) Social neuroeconomics: the neural circuitry of social
preferences. Trends Cogn Sci 11: 419–427.
3. Knafo A, Israel S, Darvasi A, Bachner-Melman R, Uzefovsky F, et al. (2007)
Individual differences in allocation of funds in the dictator game associated with
length of the arginine vasopressin 1a receptor RS3 promoter region and
correlation between RS3 length and hippocampal mRNA. Genes Brain
4. Eckel CC, Grossman PJ (1996) Altruism in Anonymous Dictator Games. Games
and Economic Behavior 16: 181–191.
5. Henrich J, Boyd R, Bowles S, Camerer C, Fehr E, et al. (2005) "Economic man"
in cross-cultural perspective: behavioral experiments in 15 small-scale societies.
Behav Brain Sci 28: 795–815; discussion 815–755.
6. Bartz AJ, Hollander E (2006) The neuroscience of affiliation: Forging links
between basic and clinical research on neuropeptides and social behavior. Horm
Behav 50: 518–528.
7. Young LJ, Wang Z (2004) The neurobiology of pair bonding. Nat Neurosci 7:
8. Prichard ZM, Mackinnon AJ, Jorm AF, Easteal S (2007) AVPR1A and OXTR
polymorphisms are associated with sexual and reproductive behavioral
phenotypes in humans. Mutation in brief no. 981. Online. Hum Mutat 28: 1150.
9. Carter CS (2006) Sex differences in oxytocin and vasopressin: Implications for
autism spectrum disorders? Behav Brain Res.
10. Storm EE, Tecott LH (2005) Social circuits: peptidergic regulation of
mammalian social behavior. Neuron 47: 483–486.
11. Cho MM, DeVries AC, Williams JR, Carter CS (1999) The effects of oxytocin
and vasopressin on partner preferences in male and female prairie voles
(Microtus ochrogaster). Behav Neurosci 113: 1071–1079.
12. Hirasawa M, Mouginot D, Kozoriz MG, Kombian SB, Pittman QJ (2003)
Vasopressin differentially modulates non-NMDA receptors in vasopressin and
oxytocin neurons in the supraoptic nucleus. J Neurosci 23: 4270–4277.
13. Landgraf R, Neumann ID (2004) Vasopressin and oxytocin release within the
brain: a dynamic concept of multiple and variable modes of neuropeptide
communication. Front Neuroendocrinol 25: 150–176.
14. Pedersen CA, Boccia ML (2006) Vasopressin interactions with oxytocin in the
control of female sexual behavior. Neuroscience 139: 843–851.
15. Ragnauth AK, Goodwillie A, Brewer C, Muglia LJ, Pfaff DW, et al. (2004)
Vasopressin stimulates ventromedial hypothalamic neurons via oxytocin
receptors in oxytocin gene knockout male and female mice. Neuroendocrinology
16. Carter CS (1998) Neuroendocrine perspectives on social attachment and love.
Psychoneuroendocrinology 23: 779–818.
17. Bales KL, Kim AJ, Lewis-Reese AD, Sue Carter C (2004) Both oxytocin and
vasopressin may influence alloparental behavior in male prairie voles. Hormones
and Behavior 45: 354–361.
18. Barberis C, Mouillac B, Durroux T (1998) Structural bases of vasopressin/
oxytocin receptor function. J Endocrinol 156: 223–229.
19. Bossmar T, Akerlund M, Szamatowicz J, Laudanski T, Fantoni G, et al. (1995)
Receptor-mediated uterine effects of vasopressin and oxytocin in nonpregnant
women. Br J Obstet Gynaecol 102: 907–912.
20. Baumgartner T, Heinrichs M, Vonlanthen A, Fischbacher U, Fehr E (2008)
Oxytocin shapes the neural circuitry of trust and trust adaptation in humans.
Neuron 58: 639–650.
21. Kosfeld M, Heinrichs M, Zak PJ, Fischbacher U, Fehr E (2005) Oxytocin
increases trust in humans. Nature 435: 673–676.
22. Domes G, Heinrichs M, Michel A, Berger C, Herpertz SC (2006) Oxytocin
Improves "Mind-Reading" in Humans. Biol Psychiatry.
23. Domes G, Heinrichs M, Glascher J, Buchel C, Braus DF, et al. (2007) Oxytocin
attenuates amygdala responses to emotional faces regardless of valence. Biol
Psychiatry 62: 1187–1190.
24. Hollander E, Bartz J, Chaplin W, Phillips A, Sumner J, et al. (2007) Oxytocin
increases retention of social cognition in autism. Biol Psychiatry 61: 498–503.
25. Lerer E, Levi S, Salomon S, Darvasi A, Yirmiya N, et al. (2008) Association
between the oxytocin receptor (OXTR) gene and autism: relationship to
Vineland Adaptive Behavior Scales and cognition. Mol Psychiatry 13: 980–988.
26. Yirmiya N, Rosenberg C, Levi S, Salomon S, Shulman C, et al. (2006)
Association between the arginine vasopressin 1a receptor (AVPR1a) gene and
autism in a family-based study: mediation by socialization skills. Mol Psychiatry
27. Kim SJ, Young LJ, Gonen D, Veenstra-VanderWeele J, Courchesne R, et al.
(2002) Transmission disequilibrium testing of arginine vasopressin receptor 1A
(AVPR1A) polymorphisms in autism. Mol Psychiatry 7: 503–507.
28. Wassink TH, Piven J, Vieland VJ, Pietila J, Goedken RJ, et al. (2004)
Examination of AVPR1a as an autism susceptibility gene. Mol Psychiatry 9:
29. Wu S, Jia M, Ruan Y, Liu J, Guo Y, et al. (2005) Positive association of the
oxytocin receptor gene (OXTR) with autism in the Chinese Han population.
Biol Psychiatry 58: 74–77.
30. Jacob S, Brune CW, Carter CS, Leventhal BL, Lord C, et al. (2007) Association
of the oxytocin receptor gene (OXTR) in Caucasian children and adolescents
with autism. Neurosci Lett 417: 6–9.
31. Zak PJ, Stanton AA, Ahmadi S (2007) Oxytocin increases generosity in humans.
PLoS ONE 2: e1128.
32. Schinka JA, Busch RM, Robichaux-Keene N (2004) A meta-analysis of the
association between the serotonin transporter gene polymorphism (5-HTTLPR)
and trait anxiety. Mol Psychiatry 9: 197–202.
33. Van Lange PAM, De Bruin EMN, Otten W, Joireman JA (1997) Development
of prosocial, individualistic, and competitive orientations: Theory and
preliminary evidence. Journal of Personality and Social Psychology 73: 733–746.
34. Van Lange PAM (1999) The pursuit of joint outcomes and equality in outcomes:
An integrative model of social value orientation. Journal of Personality and
Social Psychology 77: 337–349.
35. Halldorsson BV, Istrail S, De La Vega FM (2004) Optimal selection of SNP
markers for disease association studies. Hum Hered 58: 190–202.
OXTR and Social Behavior
PLoS ONE | www.plosone.org9 May 2009 | Volume 4 | Issue 5 | e5535
36. Crawford DC, Nickerson DA (2005) Definition And Clinical Importance Of Download full-text
Haplotypes. Annu Rev Med 56: 303–320.
37. Zeger SL, Liang KY (1986) Longitudinal data analysis for discrete and
continuous outcomes. Biometrics 42: 121–130.
38. Laird NM, Horvath S, Xu X (2000) Implementing a unified approach to family-
based tests of association. Genet Epidemiol 19 Suppl 1: S36–42.
39. Herbert A, Gerry NP, McQueen MB, Heid IM, Pfeufer A, et al. (2006) A
common genetic variant is associated with adult and childhood obesity. Science
40. Bielsky IF, Young LJ (2004) Oxytocin, vasopressin, and social recognition in
mammals. Peptides 25: 1565–1574.
41. Israel S, Lerer E, Shalev I, Uzefovsky F, Reibold M, et al. (2008) Molecular
genetic studies of the arginine vasopressin 1a receptor (AVPR1a) and the
oxytocin receptor (OXTR) in human behavior: From autism to altruism with
some notes in between. Prog Brain Res in press.
42. Granot R, Frankel Y, Gritsenko V, Lerer E, Gritsenko I, et al. (2007) Provisional
evidence that the arginine vasopressin 1a receptor gene is associated with
musical memory. Evolution and Human Behavior in Press.
43. Bachner-Melman R, Zohar AH, Elizur Y, Nemanov L, Gritsenko I, et al. (2004)
Association between a vasopressin receptor AVPR1A promoter region
microsatellite and eating behavior measured by a self-report questionnaire
(Eating Attitudes Test) in a family-based study of a nonclinical population.
Int J Eat Disord 36: 451–460.
44. Bachner-Melman R, Zohar AH, Bacon-Shnoor N, Elizur Y, Nemanov L, et al.
(2005) Linkage between vasopressin receptor AVPR1A promoter region
microsatellites and measures of social behavior in humans. Journal of Individual
Differences 26: 2–10.
45. Bachner-Melman R, Dina C, Zohar AH, Constantini N, Lerer E, et al. (2005)
AVPR1a and SLC6A4 Gene Polymorphisms Are Associated with Creative
Dance Performance. PLoS Genet 1: e42.
46. Fries AB, Ziegler TE, Kurian JR, Jacoris S, Pollak SD (2005) Early experience in
humans is associated with changes in neuropeptides critical for regulating social
behavior. Proc Natl Acad Sci U S A 102: 17237–17240.
47. Thompson R, Gupta S, Miller K, Mills S, Orr S (2004) The effects of
vasopressin on human facial responses related to social communication.
Psychoneuroendocrinology 29: 35–48.
48. Meinlschmidt G, Heim C (2007) Sensitivity to intranasal oxytocin in adult men
with early parental separation. Biol Psychiatry 61: 1109–1111.
49. Guastella AJ, Mitchell PB, Dadds MR (2007) Oxytocin Increases Gaze to the
Eye Region of Human Faces. Biol Psychiatry.
50. Zak PJ, Fakhar A (2006) Neuroactive hormones and interpersonal trust:
International evidence. Econ Hum Biol 4: 412–429.
51. Kirsch P, Esslinger C, Chen Q, Mier D, Lis S, et al. (2005) Oxytocin modulates
neural circuitry for social cognition and fear in humans. J Neurosci 25:
52. Bartels A, Zeki S (2004) The neural correlates of maternal and romantic love.
Neuroimage 21: 1155–1166.
53. Gordon I, Zagoory-Sharon O, Schneiderman I, Leckman JF, Weller A, et al.
(2008) Oxytocin and cortisol in romantically unattached young adults:
Associations with bonding and psychological distress. Psychophysiology.
54. Savaskan E, Ehrhardt R, Schulz A, Walter M, Schachinger H (2008) Post-
learning intranasal oxytocin modulates human memory for facial identity.
55. Kuersten S, Goodwin EB (2003) The power of the 39 UTR: translational control
and development. Nat Rev Genet 4: 626–637.
56. Mizumoto Y, Kimura T, Ivell R (1997) A genomic element within the third
intron of the human oxytocin receptor gene may be involved in transcriptional
suppression. Mol Cell Endocrinol 135: 129–138.
57. Blanks AM, Shmygol A, Thornton S (2007) Regulation of oxytocin receptors
and oxytocin receptor signaling. Semin Reprod Med 25: 52–59.
58. Knafo A, Plomin R (2006) Parental discipline and affection and children’s
prosocial behavior: genetic and environmental links. J Pers Soc Psychol 90:
59. Knafo A, Plomin R (2006) Prosocial behavior from early to middle childhood:
genetic and environmental influences on stability and change. Dev Psychol 42:
60. Glazier AM, Nadeau JH, Aitman TJ (2002) Finding genes that underlie complex
traits. Science 298: 2345–2349.
61. Bachner-Melman R, Gritsenko I, Nemanov L, Zohar AH, Dina C, et al. (2005)
Dopaminergic polymorphisms associated with self-report measures of human
altruism: a fresh phenotype for the dopamine D4 receptor. Mol Psychiatry 10:
62. Knafo A, Zahn-Waxler C, Davidov M, Van Hulle C, Robinson J, Rhee SH.
Empathy in early childhood: Genetic, environmental and affective contributions.
Annals of the New York Academy of Sciences: in press.
63. Dudbridge F (2003) Pedigree disequilibrium tests for multilocus haplotypes.
Genet Epidemiol 25: 115–121.
64. Koeleman BP, Dudbridge F, Cordell HJ, Todd JA (2000) Adaptation of the
extended transmission/disequilibrium test to distinguish disease associations of
multiple loci: the Conditional Extended Transmission/Disequilibrium Test. Ann
Hum Genet 64: 207–213.
65. Steen KV, Lange C (2005) PBAT: A comprehensive software package for
genome-wide association analysis of complex family-based studies. Hum
Genomics 2: 67–69.
66. Lange C, DeMeo D, Silverman EK, Weiss ST, Laird NM (2004) PBAT: tools
for family-based association studies. Am J Hum Genet 74: 367–369.
67. Rabinowitz D, Laird N (2000) A unified approach to adjusting association tests
for population admixture with arbitrary pedigree structure and arbitrary missing
marker information. Hum Hered 50: 211–223.
OXTR and Social Behavior
PLoS ONE | www.plosone.org10 May 2009 | Volume 4 | Issue 5 | e5535