Neuroscience Letters 437 (2008) 158–161
Contents lists available at ScienceDirect
Neuroscience Letters
journal homepage: www.elsevier.com/locate/neulet
Relationship between platelet serotonin content and rejections
of unfair offers in the ultimatum game
Enzo Emanuele a,∗ , Natascia Brondino b , Marco Bertona a , Simona Re a , Diego Geroldi a
a
b
Interdepartmental Center for Research in Molecular Medicine (CIRMC), University of Pavia, Viale Taramelli 24, I-27100 Pavia, Italy
Department of Applied Health and Behavioural Sciences, Section of Psychiatry, University of Pavia, Pavia, Italy
a r t i c l e
i n f o
Article history:
Received 7 February 2008
Received in revised form 27 March 2008
Accepted 2 April 2008
Keywords:
Neuroeconomics
Bargaining
Ultimatum game
Serotonin
a b s t r a c t
The ultimatum game (UG), a well-studied decision task used in experimental neuroeconomics, represents
a simple two-person bargaining between a proposer and a responder. The proposer offers the responder
how to split a sum of money. The responder decides whether to accept or reject the offer. When the
responder accepts it, each player earns money according to the proposer’s offer. If the offer is rejected,
neither player gets anything. Rejections of “free” money in the UG represent a deviation from the standard
economic model of rationality. This behaviour could be linked to adverse psychological reactions to unfair
offers, including anger, hostility and impulsiveness. Currently, it is believed that the most plausible biological system related to anger and impulsivity is the serotonergic system. We hypothesize that serotonergic
activity, as measured by platelet serotonin levels, will differentiate subjects who either reject or accept
low UG offers. A sample of 60 economy students (31 males and 29 females, mean age: 24.4 ± 2.3 years)
was investigated. As predicted, the mean platelet serotonin level was significantly lower in participants
who reject unfair offers (D 1 out of D 10) than in those who accept (2.86 ± 0.13 versus 3.48 ± 0.11 nmol/109
platelets, respectively, p < 0.001). We conclude that low platelet serotonin may serve as a reliable biomarker
to identify people who are more likely to reject unfair ultimatum offers in an experimental neuroeconomic
setting. Our pilot data seem to indicate that the serotonergic system may play an important role in the
UG rejection behaviour.
© 2008 Elsevier Ireland Ltd. All rights reserved.
One of the most extensively studied decision tasks in the neuroeconomic literature is the ultimatum game (UG) [21,8,20,1,23,
10,15,18,14]. In the UG there are two players and an endowment.
The first of two players (the proposer) divides an amount of money
between the two players. Payment of the allocated money is conditional on the second player’s (the responder) acceptance of the proposer’s decision. If the offer is rejected, neither player gets anything.
A large body of experimental evidence suggests that unfair ultimatum offers are often rejected [15,14]. Such rejections of “free”
money in the UG conflict with the predictions of the standard economic model of rationality. Indeed, a true rational responder should
accept any nonzero amount in keeping with the rational maximization of utility [17]. Situations in which rationality fails to predict
actual behaviour often involve a strong emotional component [8].
In ultimatum experiments, unfair offers often inspire anger and
a desire for revenge [16,22]. Moreover, impulsiveness may affect
decision-making [24], and impulsive behaviour could be involved
in the rejection of remunerative, but unfair, offers.
∗ Corresponding author. Tel.: +39 0382 528 341; fax: +39 0382 528 341.
E-mail address: [email protected] (E. Emanuele).
0304-3940/$ – see front matter © 2008 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.neulet.2008.04.006
Currently, it is believed that the most plausible biological system related to negative emotional reactivity and impulsiveness is
the serotonergic system [11,4]. Accordingly, it has been suggested
that serotonin dysfunction may play a role in the expression of irritability and anger [9]. Moreover, impulsiveness has been shown
to be associated with low brain serotonin turnover [3]. Taken
together, these findings seem to suggest that serotonergic activity may account at least in part for the emotional component of
human economic decision-making. To the best of our knowledge,
however, serotonergic function has never been studied in relation
to the UG behaviour. In the present study, we hypothesize that serotonergic activity, as measured by platelet serotonin content, will
differentiate subjects who either accept or reject unfair UG offers.
Specifically, the focus of the experiment was on the potential role of
serotonin in rejection behaviour. We used platelet serotonin level
as a marker of serotonergic function because it is non-invasively
recovered, inexpensive, easy to obtain in an experimental context,
and can be considered as a simplified model for serotonergic neurons [6,7,13].
Subjects were 60 undergraduate economy students enrolled in
Northern Italy. This sample consisted of 31 males and 29 females.
The age range of participants was 22–28 years, with a mean age
E. Emanuele et al. / Neuroscience Letters 437 (2008) 158–161
of 24.4 (S.D. = 2.3) years. The study protocol complied with the
Declaration of Helsinki and followed the guidelines of our internal
review board.
In order to ensure that all subjects were free of any psychiatric
disorders that could bias the results, all eligible subjects were first
screened by detailed structured psychiatric interviews (Structured
Clinical Interview for DSM-IV Disorders of Axis I and Structured
Clinical Interview for DSM-IV Axis II Personality Disorders) and
through a series of rating scales that included the Beck Depression Inventory (BDI), the State-Trait Anxiety Inventory (STAI), and
the Yale-Brown Obsessive-Compulsive Rating Scale (Y-BOCS). AxisI and axis-II disorders, axis-I diagnosis of first degree relatives and
psychotropic medication intake led to immediate exclusion from
the study. In addition, only subjects who did not show an abnormal
scoring at psychometric instruments were considered for participation. The cutoff points used for defining abnormal scores were in
accordance with previous research [5]. Specifically, all study participants showed BDI scores <16, STAI scores <46, and Y-BOCS scores
lower than or equal to 7.
Participants were instructed to distribute D 10 between themselves and their randomly assigned, anonymous, partner. They were
told that their partner could either accept or reject their offer. If the
offer was accepted, the money would be divided accordingly. If the
offer was rejected, both agents received nothing. All aspects of the
game were public knowledge without using deception. Subjects
were asked for their behaviour as either proposers or responders.
The games were played in two rounds and all the subjects participate in both roles (proposer and responder). Since the focus of the
experiment was on the potential role of serotonin on the rejection
behaviour, ultimatum offers were constrained to be D 5 (proposer
keeps D 5) or D 1 (proposer keeps D 9) to generate a roughly even
split between fair (5:5) and unfair (1:9) offers, in keeping with
previous methodology [2].
Prior to the bargaining experiments, blood samples were drawn
from all participants and collected in Vacutainer tubes (BectonDickinson, Meylan Cedex, France) containing 0.12 mL (0.34 mol/L)
EDTA solution. Five milliliters of blood were centrifuged for 5 min
at 10,000 × g at 4 ◦ C, to obtain platelet-rich plasma. After a platelet
count was obtained, platelet serotonin concentration was determined according to a previous method using an HPLC system [19].
Concentrations of serotonin were expressed as nmol/109 platelets.
Platelet serotonin was determined in a quality control sample with
within-series and between-series coefficients of variation of 2.9
and 4.3%, respectively. Since laboratory personnel were blinded to
the participants’ UG behaviour, any possible measurement error
was likely to be non-differential.
Statistical analyses were performed using SPSS statistical software (SPSS for Windows, release 11.0, SPSS, Chicago, IL, USA).
Descriptive statistics are presented as mean ± S.D. or counts, as
appropriate. Differences between groups were compared with
parametric t-test because data had a Gaussian distribution. Categorical data were analyzed with the 2 -test. The level of statistical
significance was taken as p < 0.05 (two-tailed).
Two rounds of UG experiments were performed. The results
of the first round are depicted in Fig. 1. In the second round
(Fig. 2), participants who were in the proposer role during the
first round played the responder role. As shown in Table 1, mean
platelet serotonin levels, as determined with HPLC, were significantly lower in participants who reject unfair offers (D 1 out of D 10)
than in those who accept (2.86 ± 0.13 versus 3.48 ± 0.11 nmol/109
platelets, respectively, t = 3.501, p < 0.001, Fig. 3). No differences
in platelet count and mean platelet volume were observed
between the two groups (Table 1). There was no statistically significant association between platelet serotonin levels and age
and scores on the psychometric indices. Similarly, no differ-
159
Fig. 1. Results of the first round of the UG experiment.
Fig. 2. Results of the second round of the UG experiment. Herein participants who
were in the proposer role during the first round played the responder role.
ences with regard to platelet serotonin content were found
among males (n = 8, 2.94 ± 0.14 nmol/109 platelets) as compared
to females (n = 10, 2.77 ± 0.12 nmol/109 platelets, p = 0.72) who
rejected unfair offers. Notably, samples taken from individuals who
made unfair offers (n = 28, 2.99 ± 0.14 nmol/109 platelets) showed
a slight, but significant (t = 2.431, p < 0.05), reduction of mean
platelet serotonin levels compared with those who made fair offers
(n = 32, 3.43 ± 0.11 nmol/109 platelets). No differences with regard
to platelet serotonin content were found among males (n = 18,
Table 1
General characteristics of individuals according to UG rejection behaviour
Age (years)
Males/females
BDI
STAI trait
STAI state
Y-BOCS
Platelet serotonin,
nmol/109 platelets
Platelet count (109 /L)
Mean platelet volume (fL)
Reject D 1 out
of D 10 (n = 18)
Accept D 1 out
of D 10 (n = 10)
p
24.9 ± 2.6
8/10
4.8 ± 1.1
29.13 ± 12.57
26.13 ± 10.57
2.4 ± 0.9
2.86 ± 0.13
24.1 ± 2.0
6/4
5.2 ± 0.9
28.63 ± 11.57
25.95 ± 11.02
2.6 ± 1.1
3.48 ± 0.11
0.81
0.64
0.72
0.31
0.59
0.88
<0.001
0.421 ± 0.125
8.04 ± 1.06
0.433 ± 0.118
7.95 ± 1.04
0.49
0.61
BDI: Beck Depression Inventory; STAI: State-Trait Anxiety Inventory; Y-BOCS: YaleBrown Obsessive Compulsive Scale.
160
E. Emanuele et al. / Neuroscience Letters 437 (2008) 158–161
Fig. 3. Box plots of the platelet serotonin levels in subjects who either reject or
accept low UG offers. The bottom and top of the box plots are the 25th and 75th
percentiles. The median is the horizontal line within the box. Platelet serotonin was
significantly lower in responders who rejected unfair offers when compared to those
who accepted (p < 0.001).
3.55 ± 0.13 nmol/109 platelets) as compared to females (n = 14,
3.24 ± 0.10 nmol/109 platelets, p = 0.59) who made fair offers.
In recent years, ultimatum game experiments have been widely
used to investigate the mechanisms of human economic decisionmaking [8,23]. Low offers in the UG induce conflict in the responder
between accepting an unfair division and foregoing financial gain
[16]. To gain a better understanding of the UG rejections is
important since traditional economic theory fails to predict this
behaviour. It has been recently suggested that anger and negative
emotional reactivity may play an important role in determining the outcome of the UG [8,22,16,24]. Individual differences
in emotional reactivity and impulse control are associated with
diminished serotonergic function, as seen in a variety of clinical and
forensic populations [11,4,9,3]. We investigated the relationship
between platelet serotonin content and rejections of low offers in
the UG under the hypothesis that the serotonergic system plays an
important role in emotional reactivity, impulsiveness and anger. As
predicted, our data provide evidence that low serotonergic activity
may be involved in the normal, albeit irrational, rejections of low UG
offers. Accordingly, the main result of this study is that subjects who
reject low UG offers have a lower platelet serotonin content compared with those who accept them. Interestingly, samples taken
from individuals who made unfair offers showed a slight, but significant, reduction of mean platelet serotonin levels compared with
those who made fair offers.
It is generally accepted that variability in central serotonergic
activity underlies, in part, individual differences in the constraint of
anger and impulse, with reduced serotonergic function associated
with actions committed in haste [11,4]. Notably, previous studies
have shown that central serotonergic activity can be influenced
both environmentally (e.g., by social stress) and by genetic factors
[12]. Interestingly, a very recent study has shown a genetic loading on human UG behaviour [23]. Although our findings provide no
direct evidence of a genetic influence on platelet serotonin content,
they are at least consistent with this possibility. Accordingly, a 44base pair polymorphism in the regulatory region of the serotonin
transporter gene has been associated with peripheral serotonin
indices [4].
Platelet serotonin has been previously associated to psychological trait markers, but not with specific nosological constructs [12],
and in keeping with this, no any significant correlation between
platelet serotonin and psychometric scores was evident in our
healthy nonclinical sample of economy students. Platelet serotonin
levels might be affected by different factors [13]; however, the
effect of age and sex was neglected in this study as there was
no significant relation between age, gender and platelet serotonin
in our healthy subjects. Diurnal variation or seasonal variations
were also excluded since collection of blood was carried out at
the same times; therefore these sources of variation are unlikely to
have biased the results. Moreover, the platelet serotonin content as
observed in the entire study cohort was in keeping with the normal
range reported in the literature [6,7]. It could be also argued that the
selection of subjects without psychiatric comorbidities could have
diminished any potential relationship detectable between serotonin and UG behaviour. Even though such bias could exist, we
believe that such an error would underestimate rather than overestimate the relationship between reduced platelet serotonin content
and rejection of unfair offers in the UG as detected in our study.
In conclusion, our findings provide preliminary evidence of a
potential association between UG bargaining behaviour and levels
of a peripheral biological marker of serotonergic function. These
data add to the growing evidence that the economic behaviour
might be influenced by biological factors, including hormonal [2]
and genetic influences [23].
References
[1] N. Agay, S. Kron, Z. Carmel, S. Mendlovic, Y. Levkovitz, Ultimatum bargaining behaviour of people affected by schizophrenia, Psychiatry Res. 157 (2008)
39–46.
[2] T.C. Burnham, High-testosterone men reject low ultimatum game offers, Proc.
Biol. Sci. 274 (2007) 2327–2330.
[3] L. Buydens-Branchey, M. Branchey, J. Hudson, P. Fergeson, Low HDL cholesterol,
aggression and altered central serotonergic activity, Psychiatry Res. 93 (2000)
93–102.
[4] P. Castrogiovanni, M.R. Capone, I. Maremmani, D. Marazziti, Platelet serotonergic markers and aggressive behaviour in healthy subjects, Neuropsychobiology
29 (1994) 105–107.
[5] E. Emanuele, P. Politi, M. Bianchi, P. Minoretti, M. Bertona, D. Geroldi, Raised
plasma nerve growth factor levels associated with early-stage romantic love,
Psychoneuroendocrinology 31 (2006) 288–294.
[6] N. Eynard, E. Flachaire, C. Lestra, M. Broyer, R. Zaidan, B. Claustrat, C. Quincy,
Platelet serotonin content and free and total plasma tryptophan in healthy
volunteers during 24 hours, Clin. Chem. 39 (1993) 2337–2340.
[7] E. Flachaire, C. Beney, A. Berthier, J. Salandre, C. Quincy, B. Renaud, Determination of reference values for serotonin concentration in platelets of healthy
newborns, children, adults, and elderly subjects by HPLC with electrochemical
detection, Clin. Chem. 36 (1990) 2117–2120.
[8] K.M. Harlé, A.G. Sanfey, Incidental sadness biases social economic decisions in
the ultimatum game, Emotion 7 (2007) 876–881.
[9] R. Kavoussi, P. Armstead, E. Coccaro, The neurobiology of impulsive aggression,
Psychiatr. Clin. North Am. 20 (1997) 395–403.
[10] M. Koenigs, D. Tranel, Irrational economic decision-making after ventromedial
prefrontal damage: evidence from the ultimatum game, J. Neurosci. 27 (2007)
951–956.
[11] M. Krakowski, Violence and serotonin: influence of impulse control, affect regulation, and social functioning, J. Neuropsychiatry Clin. Neurosci. 15 (2003)
294–305.
[12] S.B. Manuck, J.D. Flory, J.M. McCaffery, K.A. Matthews, J.J. Mann, M.F. Muldoon,
Aggression, impulsivity, and central nervous system serotonergic responsivity
in a nonpatient sample, Neuropsychopharmacology 19 (1998) 287–299.
[13] E. Maurer-Spurej, C. Pittendreigh, S. Misri, Platelet serotonin levels support
depression scores for women with postpartum depression, J. Psychiatry Neurosci. 32 (2007) 23–29.
[14] M.A. Nowak, K.M. Page, K. Sigmund, Fairness versus reason in the ultimatum
game, Science 289 (2000) 1773–1775.
[15] Y. Ohmura, T. Yamagishi, Why do people reject unintended inequity? Responders’ rejection in a truncated ultimatum game, Psychol. Rep. 96 (2005) 533–
541.
[16] J.J. Rotemberg, Minimally acceptable altruism and the ultimatum game, J. Econ.
Behav. Organ. 66 (2008) 457–476.
[17] A. Rubistein, Perfect equilibrium in a bargaining model, Econometrica 50 (1982)
97–109.
E. Emanuele et al. / Neuroscience Letters 437 (2008) 158–161
[18] A.G. Sanfey, J.K. Rilling, J.A. Aronson, L.E. Nystrom, J.D. Cohen, The neural basis
of economic decision-making in the ultimatum game, Science 300 (2003)
1755–1758.
[19] L.G. Schmidt, P. Dufeu, A. Heinz, S. Kuhn, H. Rommelspacher, Serotonergic
dysfunction in addiction: effects of alcohol, cigarette smoking and heroin on
platelet 5-HT content, Psychiatry Res. 72 (1997) 177–185.
[20] T. Takahashi, Economic decision-making in the ultimatum game by smokers,
Neuro Endocrinol. Lett. 28 (2007) 659–661.
[21] B. Van den Bergh, S. Dewitte, Digit ratio (2D:4D) moderates the impact of sexual cues on men’s decisions in ultimatum games, Proc. Biol. Sci. 273 (2006)
2091–2095.
161
[22] M. van’t Wout, R.S. Kahn, A.G. Sanfey, A. Aleman, Affective state and
decision-making in the ultimatum game, Exp. Brain Res. 169 (2006) 564–
568.
[23] B. Wallace, D. Cesarini, P. Lichtenstein, M. Johannesson, Heritability of ultimatum game responder behaviour, Proc. Natl. Acad. Sci. U.S.A. 104 (2007)
15631–15634.
[24] M. Wittmann, M.P. Paulus, Decision making, impulsivity and time perception,
Trends Cogn. Sci. 12 (2008) 7–12.