Psychiatry Research 185 (2011) 9296

Contents lists available at ScienceDirect

Psychiatry Research
j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / p s yc h r e s

Interactive decision-making in people with schizotypal traits: A game

a r t i c l e i n f o a b s t r a c t

Article history: Studies that have investigated whether decits in social cognition observed in schizophrenia are also present
Received 16 August 2009 in schizotypal individuals have largely been inconclusive, and none of these studies have examined social
Received in revised form 12 May 2010 interactive behavior. Here, we investigated interactive decision-making behavior in individuals differing in
Accepted 17 May 2010
the amount of schizotypal symptoms using tasks derived from Game Theory. In total 1691 undergraduate
students were screened with the Schizotypal Personality Questionnaire-Brief version. We selected 69 people
Keywords:
Schizotypy
distributed across the full schizotypal continuum to participate in Ultimatum and Dictator Games in which
Schizophrenia they played against human and non-human, computer partners. The results showed that higher levels of
Social cognition schizotypal symptoms, particularly positive and disorganized schizotypy, were related to proposing higher
Social functioning offers to all partners. Additionally, the amount of interpersonal schizotypal symptoms was associated with
Decision-making an increased acceptance rate of very unfair offers from human partners, possibly reecting a blunted
Game theory emotional response to such offers. We conclude that positive and disorganized schizotypal symptoms are
Ultimatum game associated with less adequate bargaining behavior, similar to what has been recently observed in patients
Dictator game
with schizophrenia. The observed similarities on Ultimatum Game behavior between patients with
schizophrenia and individuals with more schizotypal symptoms contribute to the growing evidence that
social cognitive decits may represent a marker of vulnerability to schizophrenia.
Published by Elsevier Ireland Ltd.

1. Introduction 2002; Baas et al., 2008) and social knowledge or understanding social
schema (Corrigan and Addis, 1995; Penn et al., 2002).
Apparent social dysfunction is one of the major classications of Equivalent aberrations, although to a lesser degree, have been
the diagnosis of schizophrenia (A.P.A., 2000) and these dysfunctions reported in individuals with schizophrenia-like, or so-called schizo-
often precede the onset of the illness (Hans et al., 1992; Baum and typal symptoms (Poreh et al., 1994; Langdon and Coltheart, 1999;
Walker, 1995), which have led to the suggestion that social Waldeck and Miller, 2000; Henry et al., 2008). However, ndings are
dysfunction can serve as vulnerability marker for schizophrenia. inconsistent, and other studies have reported no decits in social
Given that the perception, interpretation and processing of social emotional processing in schizotypal individuals (Toomey and Schuld-
emotional information is crucial for appropriately navigating social berg, 1995; Van 't Wout et al., 2004; Jahshan and Sergi, 2007).
environments, numerous studies have examined socialemotional Schizotypy is the basis of schizotypal personality disorder (Raine,
information processing in schizophrenia (see Green et al., 2005 for an 1991) and is thought to be a continuous phenomenon that can be
overview of social cognition domains investigated in schizophrenia). detected in the general population. Schizotypal traits have been
These studies reported social cognitive decits measured with various associated with increased risk for developing psychotic disorders
tasks and include impairments in emotion processing such as including schizophrenia (Lenzenweger, 1994) and relatives of
recognition of facial emotional expressions and tone of voice patients with schizophrenia report having more schizotypal symp-
(Edwards et al., 2002; Kohler and Brennan, 2004; Van 't Wout et al., toms (Vollema et al., 2002). The examination of individuals with high
2007); theory of mind and the attribution of mental states to others levels of schizotypal symptoms is informative as it allows researchers
(Corcoran et al, 1995; Greig et al., 2004) social perceptions and to study schizophrenia-related traits without confounding factors
impaired social judgments (Corrigan and Green, 1993; Toomey et al., such as medication, severe psychopathology, and institutionalization.
Additionally, (cognitive) aberrations in schizotypal individuals that
parallel those seen in schizophrenia might point to a vulnerability
Corresponding author. Brown University, 190 Thayer Street, P.O. Box 1978,
marker for the disease (Gottesman and Gould, 2003).
Providence, RI, USA. Tel.: + 1 401 8636489; fax: + 1 401 8632255. Despite the important ndings of socialemotional processing
E-mail address: [email protected] (M. van 't Wout). decits in schizophrenia and schizotypal individuals, the majority of

0165-1781/$ see front matter. Published by Elsevier Ireland Ltd.

studies have investigated people placed outside an actual social apparent when overly fair offers are proposed in the Ultimatum Game
interaction. In these situations, actions have no direct consequences as compared to offers proposed in the Dictator Game. In contrast,
for participants or others, for instance when asked to label facial offering fair amounts in both games could reect other goals, i.e.
affective expressions. However, social behavior can best be thought of inequity aversion or altruism.
as an interaction between two or more individuals in which the The purpose of this study was to examine interactive decision-
outcome of the interaction has consequences for all parties. making behavior in non-patient individuals across the range of
Consequently, an important, though currently understudied, question schizotypal symptoms. Using multiple Game Theory tasks we were
is whether schizophrenia and schizotypy can be associated with able to test whether schizophrenia phenomena are associated with
abnormalities in social interactive behavior in which decisions made aberrant social bargaining behavior. We hypothesized that individuals
have actual consequences. with more schizotypal symptoms would propose a higher share to
To investigate consequential social interactive behavior, the eld another person only when their partners can reject unfair offers
of Economic Game Theory provides a set of useful, widely-used tasks (parallel to the results of Agay et al., 2008 in patients with
that allow the investigation of strategic decision-making in a social schizophrenia). We further expected that when playing the Ultima-
interactive context (Camerer, 2003). More specically, these tasks tum Game in the role of responder to nd a relationship between
show how people's decision-making behavior consistently deviates degree of schizotypal symptoms and acceptance rates of unfair offers,
from the economically self-interested behavior predicted by standard as difculties in understanding bargaining behavior might result in
Economic Theory. Instead, a growing body of research using these higher acceptance rates in people with more schizotypal symptoms.
tasks has shown that emotional and social aspects of the environment In case the above-mentioned main correlations proved to be
are also crucial in decision-making. Take, for instance, the Ultimatum signicant we performed follow-up analyses to test whether specic
Game (Guth et al., 1982) in which two players must agree on the subscales were associated with game play behavior. For these follow-
division of a sum of money provided by the experimenter. Player 1, up analyses, we predicted a positive correlation between ratings on
the proposer, makes a proposal as to how the money should be split, positive and disorganized schizotypy and proposing higher offers in
and Player 2, the responder, has the opportunity to either accept or the Ultimatum Game. This was based on the idea that for successful
reject this proposal. If it is accepted, the money gets divided as bargaining it is essential to understand the goals and intentions of
proposed, but if the responder rejects this offer, neither receives any others and these so-called Theory of Mind capacities have been
money, but in both cases the game is over. This is a one-shot game, in associated specically with positive schizotypy and disorganization in
which players only play once with each other. Contrary to economic the past (Marjoram et al., 2006; Pickup, 2006; Sprong et al., 2007). For
theory, many low, unfair, offers are rejected, demonstrating that responder behavior, we predicted a positive correlation with social
motivations other than nancial self-interest come into play when emotional abnormalities as measured by the interpersonal factor of
engaged in social interactions. the Schizotypal Personality Questionnaire-Brief version (SPQ-B) and
These rejections of unfair offers have been shown to be related to increased acceptance rates in the Ultimatum Game.
feelings of anger (Pillutla and Murnighan, 1996), to neural activation
in areas associated with disgust, i.e. insula (Sanfey et al., 2003), and 2. Methods
psychophysiological responses of emotional arousal (Van 't Wout
et al., 2006). Furthermore, the insula activation and emotional arousal 2.1. Participants

reported above was specic for unfair offers made by human partners,
1691 undergraduate students from the University of Arizona completed the
as compared to non-human, computer partners, which highlights the Schizotypal Personality Questionnaire-Brief version (SPQ-B, Raine and Benishay, 1995).
social nature of decision-making in this game. Therefore, the Age range: 1849, mean 18.9 (S.D. 1.8). Their scores covered the complete range on the
administration of tasks from Game Theory to participants that are scale, i.e. from 0 to 22, with a mean score of 8.76 (S.D. 6.42). From these 1691 students,
we invited 69 individuals from across the full SPQ-B distribution to participate in a
thought to show socialemotional disturbances could give more
follow-up study. Mean age was 18.2 years (S.D. 0.8) with a range from 18 to 20 years
insight in the decits in social interactions. and a male:female ratio of 25:44.
In one example of this approach, Agay et al. (2008) recently Before the administration of the tasks, subjects again lled out the SPQ-B. Mean
showed that patients with schizophrenia did not fully exploit the SPQ-B score was 8.12 (S.D. 5.43). There was high agreement between the two test
typical strategy of offering a slightly unfair offer to the other player. scores (r = 0.85, P b 0.0001), but we observed a slight regression to the mean for the
highest scores. The study was conducted in compliance of the Declaration of Helsinki
For example, many players when dividing a $10 pot will make offers
and local ethics committee approval, and all participants provided written informed
of $3 or $4 to their partners, as these slightly unfair offers are usually consent.
accepted (Camerer, 2003). The behavior observed in patients with
schizophrenia on the other hand, was suggestive of a decit in 2.2. Measures
understanding social bargaining strategies.
However, to draw rm conclusions about decits in social 2.2.1. Schizotypal Personality Questionnaire-Brief version (SPQ-B)
bargaining strategies it is important to rule out other reasons for To be able to screen large groups of individuals on schizotypal traits, the
Schizotypal Personality Questionnaire-Brief version (Raine and Benishay, 1995) was
giving high offers, such as experiencing positive feelings of altruism,
developed. This is a short 22-item questionnaire with a dichotomous response format
being inequity averse, or implementing a social rule of giving fair (yes or no) based on the Schizotypal Personality Questionnaire (Raine, 1991). Higher
shares. Additionally, it is important to examine whether these decits scores on the SPQ-B are indicative of more schizotypal symptoms. The primary
only occur in a social setting or whether bargaining is also different in advantage of the SPQ-B is that it still includes the main three factor structure of the
non-social situations. These issues can be addressed by comparing longer schizotypal personality questionnaire, despite being short enough to allow easy
screening of large groups. These three main factors are the Cognitive-Perceptual Factor,
behavior on the Ultimatum Game with a game in which responders referring to delusional ideation, unusual perceptual experiences and covering the
have no opportunity to reject unfair offers, and by introducing non- positive schizotypal symptoms; the Interpersonal Factor, including social anxiety and
social interactions such as computer partners. More specically, in the other negative schizotypal symptoms; and the Disorganized Factor that includes odd or
Ultimatum Game proposers understand that responders can reject unusual speech and behavior.
unfair offers, thereby punishing them. To prevent offers from being
rejected, proposers offer higher amounts of money. However, when a 2.2.2. Ultimatum Game
responder cannot punish a proposer for an unfair offer, proposers can In the Ultimatum Game (Guth et al., 1982) a proposer (player 1) proposes the
division of a sum of money to a responder (player 2), with the money (in this case $10)
take advantage of the situation by offering lower, i.e. more unfair provided by the experimenter. The responder can decide to accept this offer or not, in
offers. This latter game is called the Dictator Game (Kahneman et al., either case ending the game. If the responder accepts the offer, the amount of money is
1986). Thus, disturbances in bargaining strategies might become split as agreed. However, if the responder rejects it, neither player receives anything.
94 M. van 't Wout, A.G. Sanfey / Psychiatry Research 185 (2011) 9296

Before playing, participants were instructed as to the nature and rules of the Ultimatum With respect to the schizotypal continuum, there was a signicant
Game.
positive correlation between schizotypal scores and the amount of
Participants played two sets of games, one in the role of responder and one as
proposer (order of games was counterbalanced). Each set consisted of multiple, single- money offered to partners, Spearman r = 0.36, P = 0.003 for human
shot rounds with different partners. Partners could be a human or a computer. This was partners; Spearman r = 0.25, P = 0.04 for computer partners. To test
done to investigate whether differences in bargaining behavior were specic for social whether people higher on the schizotypy continuum are also
interactions, instead of a general difference in bargaining style. Participants did not proposing more fair ($5) shares, we correlated the total number of
meet their partners but participants saw a picture of the partner that they played with
for that round. In the task instructions it was emphasized that the participant's partners
$5 offers with amount of schizotypal symptoms which was positively
in the games played the game independently of each other, with no collusion. correlated, Spearman r = 0.39, P = 0.0009. To provide support that the
Participants were told that they would be paid according to their choices in the games, above results are not due to proposing hyperfair offers only (amount
i.e. 10% of the total amount of money that was earned in both sets of Ultimatum Games. of hyperfair offers correlated signicantly with amount of schizotypal
Participants played 29 rounds, 20 times with another person (10 males, 10 females)
symptoms, Spearman r = 0.29, P = 0.03), we re-analyzed the data and
and 9 times with the computer, each time dividing $10.
When playing as a responder 9 out of the 29 rounds involved a fair split, i.e. a 50% removed all hyperfair offers. The remaining correlation was signi-
split of the $10 (three times each against a computer, a male, and a female partner) and cant for human partners, Spearman r = 0.27, P = 0.03, but not for
20 rounds involved unfair divisions; these involved six times $3 out of $10 (twice from computer partners, Spearman r = 0.19, P = 0.11.
a computer, male and female partner), six times $2 out of $10 (twice from a computer, Follow-up correlations between specic SPQ-B subscales and the
male and female partner). The remaining eight unfair offers were $1 out of $10 (twice
for computer partner, three times a male and three times a female partner). This set of
amount of money proposed to partners showed that higher scores on
offers was used as we were mostly interested in unfair offers, because previous the cognitive-perceptual subscale correlated with the amount of
research (Sanfey et al., 2003; Van 't Wout et al., 2006; Koenigs and Tranel, 2007) has money proposed, Spearman r = 0.34, P = 0.005 for human partners;
shown larger rejection rates for these offers. The different types of offers (fair and unfair Spearman r = 0.28, P = 0.02 for computer partners. Scores on the
splits, human and computer partners) were displayed in a random order.
disorganized subscale of the SPQ-B also correlated positively with
When playing as a proposer, were free to offer whatever amount they wanted to
each partner. Participants played again 29 rounds: nine against a computer, ten against offer amount in the game, but only for human partners, Spearman
a male and ten against a female partner. Participants rst saw a picture of their partner, r = 0.299, P = 0.015. See Table 1 for correlations between SPQ-B
after which they could make their proposal of how they wanted to split the $10. ratings and money proposed to human partners. However given that
Rejection rates were programmed according to realistic data in which fair offers ($5) or there were signicant correlations between SPQ-B subscales (all
offers more than $5 were always accepted, but acceptance rates decreased as proposed
P 0.001), we performed partial correlations to test whether
offers became more unfair (in 80% of cases $3 and $4 out of $10 were accepted and 40%
of $1 and $2 out of $10 were accepted). subscales were correlated with task performance independent of
the other subscales. This resulted in reduced correlations between
2.2.3. Dictator Game scores on the cognitive-perceptual subscale and money proposed to
The Dictator Game (Kahneman et al., 1986) is similar to the Ultimatum Game in human partners, r = 0.24, P = 0.055 and computer partners, r = 0.22,
that the rst player proposes a division of a sum of money (again $10) to a responder.
P = 0.08. The correlations between scores on the disorganized
However, in the Dictator Game the responder must accept the proposed offer, thus
ending the game and leaving both players with the money as suggested by the subscale and money proposed to human or computer partners in no
proposer. As there is no decision by the responder, we only examined proposer longer signicant (all P 0.3).
behavior in this game. Participants played nine rounds in total and made offers to six
human partners (three males, three females) and three computers. Similar to the
procedure in the Ultimatum Game, participants saw a picture of the partner they played
3.2. Ultimatum Game responder behavior
with on each round. Again, before playing, participants were instructed as to the nature
and rules of the Dictator Game. Participants were told that they would be paid
according to their choices in the games. In this analysis we report data from 65 instead of 69 participants.
Two participants were excluded from the analysis as they rejected
2.3. Procedure and statistical analyses every offer that was presented, including fair offers, and two people
did not complete the task due to technical difculties. Fair offers ($5)
The order of the three sets of games played by each participant (Dictator Game
Proposer, Ultimatum Game Proposer and Ultimatum Game Responder) was random- were almost always accepted (99% of cases), and, as is normally
ized across participants. The data were analyzed using correlation analyses between observed, acceptance rates decreased as offers decreased (for
schizotypy scores and behavior on the different games (percentage of the sum that was $3:79.7%, $2:36.7%, and $1:19.9%). In general, unfair offers from
offered when playing as proposers, percentage of rejections of unfair offers when humans were rejected more often (59.4%) than those from computers
playing as a responder). Given the nature of our data (non-normal distribution and
binary responses: accept or reject) we used non-parametric Spearman correlations. In
(51.5%), Wilcoxon Signed Rank test = 2.90, P = 0.004.
addition, we performed non-parametric correlations between task performance and Schizotypy scores correlated positively with higher acceptance
the specic subscales (interpersonal, disorganized and cognitive-perceptual) of the rates of the most unfair offers ($9:$1) from human partners,
SPQ-B. All tests are two-tailed. Spearman r = 0.27, P = 0.03. This suggests that individuals higher on
the schizophrenia continuum reject these unfair offers to a lesser
3. Results extent. Follow-up correlations with particular subscales resulted in
signicant positive correlations between acceptance rates of these
3.1. Ultimatum Game proposer behavior very unfair offers and the interpersonal subscale as well as the
disorganized subscale, Spearman r = 0.33, P = 0.008 and Spearman
Overall, participants offered an average of $3.80 of the $10 pot (S. r = 0.27, P = 0.03 respectively. See Table 1 for correlations between
D. 8.1) to their partners, which is within the range typically seen for SPQ-B ratings and acceptance rate $1 offers made by human partners.
offers in this game (35%40% of total money amount; Camerer, 2003).
There was a signicant difference in the amount of money offered to a Table 1
human as opposed to a computer partner, Wilcoxon Signed Rank Correlations (Spearman r and P-values) between SPQ-B ratings and performance on
test = 3.52, P = 0.0004. The average proposal to human partners each task and scores for human partners only.

was $3.90 (S.D. 8.5, range $0$10), with on average $3.50 (S.D. 10.2, Ultimatum Game Ultimatum Game Dictator Game
range $0$8) offered to computer partners. The range of offers shows proposer behavior acceptance $1 proposer behavior
that occasionally more than half, i.e. $5 of the $10 was offered. These Cognitive- r = 0.34, P = 0.005 r = 0.09, P = 0.46 r = 0.13, P = 0.31
so-called hyperfair offers, although not necessary abnormal, are rare perceptual
in Ultimatum and Dictator game studies, and in the present Interpersonal r = 0.21, P = 0.098 r = 0.33, P = 0.008 r = 0.01, P = 0.92
Disorganization r = 0.29, P = 0.015 r = 0.27, P = 0.03 r = 0.02, P = 0.90
experiment were exclusively made by participants with higher
Total SPQ-B r = 0.36, P = 0.003 r = 0.27, P = 0.03 r = 0.05, P = 0.68
schizotypal scores.
 M. van 't Wout, A.G. Sanfey / Psychiatry Research 185 (2011) 9296 95

However when using partial correlations these correlations are non- inequity aversion will be reected in the proposing of higher offers
signicant (P 0.14). (Fehr and Schmidt, 1999; Bolton and Ockenfels, 2000). In order to
further test whether schizotypal symptoms were indeed associated
3.3. Dictator Game behavior with aberrant bargaining we examined performance on the Dictator
Game, a task similar to the Ultimatum Game with the difference being
On average, participants offered $2.80 (S.D. $0.14) of their $10 to that partners were not able to reject any offers. Therefore, if
their partners in the Dictator Game. To human partners, participants participants propose generous or fair offers in both the Ultimatum
gave on average $3.30 (S.D. $0.15, range $0$8), whereas the average as well as the Dictator Game, we can conclude that fairness is a strong
offer to computer partners was $1.80 (S.D. $0.15, range $0$5). motivator behind proposer decision-making in these games (Kahne-
Compared to the Ultimatum Game trials, proposals were signicantly man et al., 1986; Forsythe et al., 1994; Scheres and Sanfey, 2006). If on
lower to both human partners, Wilcoxon Signed Rank test = 3.12, the other hand, participants offer less in the Dictator Game as
P = 0.002 and computer partners, Wilcoxon Signed Rank test = compared to the Ultimatum Game, we could assume that these higher
6.39, P b 0.0001. But similar to the results in the Ultimatum Game, Ultimatum Game offers are strategic in nature, and are made because
the amount of money offered to a human partner was signicantly of anticipation that unfair offers will be rejected (Camerer, 2003).
higher than the amount offered to a computer partner, Wilcoxon Our results showed no signicant correlation between schizotypy
Signed Rank test = 6.64, P b 0.0001. On ve occasions participants and the amount of money offered in the Dictator Game. Additionally,
offered more than 50% to a human partner, again all by people who given that the amount of money proposed to partners in the Dictator
scored relatively higher on the SPQ-B. Game was considerably lower as compared to the amount proposed in
However, there was no signicant correlation between total the Ultimatum Game, the increased offers proposed in the Ultimatum
schizotypy score and the amount of money offered to human partners, Game are likely due to being afraid that unfair offers could be rejected,
Spearman r = 0.05, P = 0.68, and computer partners, Spearman r = as opposed to an inequity aversion account. Our data therefore
0.05, P = 0.69. There were also no signicant correlations between support Agay et al's (2008) conclusion that schizophrenia phenomena
any of the SPQ-B subscales and performance on the Dictator Game. can be associated with poorer bargaining strategies even in a one-shot
See Table 1 for correlations between SPQ-B ratings and money interaction. Being able to understand what range of offers others
proposed to human partners. would be willing to consider is crucial for good bargaining. Typically,
slightly unfair offers, that is, offers around 35%40% of the total, are
4. Discussion accepted and considered reasonable by others (Camerer, 2003).
Individuals with a greater degree of schizotypal symptoms appear less
In this study we examined interactive decision-making behavior sensitive to this norm and as a result display abnormalities in
using classic economic game theory tasks in individuals with varying bargaining.
levels of schizotypal symptoms. As expected, our results show that the The observed similarity in proposer behavior on the Ultimatum
people with more schizotypal symptoms proposed more money to Game between non-clinical undergraduate students that score higher
their partners in the Ultimatum Game. In addition, higher levels of on schizotypy and patients with schizophrenia as reported by Agay
schizotypal symptoms were related to more often proposing both a et al. (2008) is of clinical importance. Not only do our results
fair distribution and hyperfair offers. These data echo previous strengthen the previous ndings in patients since the observed
ndings reported by Agay et al. (2008), who found that patients behavioral pattern in the Ultimatum Game cannot be due to
with a clinical diagnosis of schizophrenia were less strategic and more confounding factors such as severe symptomatology, medication use
often proposed hyperfair offers to their partners compared to non- or hospitalization in schizophrenia. The strikingly similar aberrations
schizophrenia controls in a multi-round Ultimatum Game. in interactive decision-making that we observed in healthy under-
Furthermore, we examined non-social interactions and, contrary graduates also reect at least in part a vulnerability to schizophrenia.
to our predictions, found that proposing higher offers in the Hence, abnormal strategic interactive decision-making appears to be
Ultimatum Game was not specic for human interactions, as associated with vulnerability for schizophrenia and not with being ill.
participants higher on schizotypy offered greater amounts to all Furthermore, our ndings support the idea of a continuity between
partner types (human and computers). Although this correlation did symptoms characteristic of schizophrenia and schizophrenia-related
not remain signicant after we removed hyperfair offers. We further phenomena in the normal population, which is important for the
investigated whether specic schizotypal symptoms were related to development for adequate preventive intervention methods.
proposer behavior and observed that proposing higher offers in the With respect to the responder role in the Ultimatum Game, that is,
Ultimatum Game specically correlated with having more cognitive- when participants were in the position to either reject or accept
perceptual and disorganized schizotypal symptoms, but not with proposed offers, we observed that individuals with more schizotypal
interpersonal schizotypal symptoms. The correlation between more symptoms accepted highly unfair offers made by human partners
cognitive-perceptual symptoms and proposing higher offers to a more often. This is a surprising nding because one could hypothesize
human partner remained signicant independent of the correlations that people who propose higher offers themselves might also expect
with the other subscales. This suggests that proposing higher offers is higher offers, and thus reject these very unfair offers more often to
related to positive schizotypal symptoms, and disorganized symp- signal how one wants to be treated (Camerer, 2003; Fehr and
toms to some extent. Given that an adequate Theory of Mind Camerer, 2007). The decreased rejection rate of very unfair offers in
(understanding the goals and intentions of others) is crucial for participants with more schizotypal symptoms could be interpreted as
successful bargaining, our data appear consistent with previous supporting the idea that individuals with higher schizotypy scores
studies reporting a relationship between mentalizing and Theory of signal their social boundaries to a lesser extent. Alternatively, the
Mind capacities and positive symptoms and disorganization in both increased acceptance rates could reect reduced emotional responses,
clinical patients as well as high-risk individuals (Marjoram et al., i.e. blunted reactions, from higher schizotypal scoring participants
2006; Pickup, 2006; Sprong et al., 2007). when confronted with such offers. This interpretation is further
However from the Ultimatum Game data alone it is difcult to supported by the nding that specically severity of interpersonal
conclude that participants with more schizotypal symptoms have schizotypy, part of the negative schizotypal symptom dimension and
aberrant bargaining strategies. For instance, one reason for proposing which includes social withdrawal, social anxiety and blunted affect,
higher offers is that individuals scoring higher on schizotypy care was related to the increase in acceptance rates. However these
more about fairness than people with less schizotypal symptoms. This explanations of the data are more speculative and should be
96 M. van 't Wout, A.G. Sanfey / Psychiatry Research 185 (2011) 9296

interpreted as such, especially given that after controlling for Corrigan, P.W., Green, M.F., 1993. Schizophrenic patients' sensitivity to social cues: the
role of abstraction. The American Journal of Psychiatry 150, 589594.
disorganization and cognitive-perceptual symptoms the correlation Edwards, J., Jackson, H.J., Pattison, P.E., 2002. Emotion recognition via facial expression
between interpersonal schizotypy did not remain signicant. and affective prosody in schizophrenia: a methodological review. Clinical
All participants played multiple games and, although the order of Psychology Review 22, 789832.
Fehr, E., Camerer, C.F., 2007. Social neuroeconomics: the neural circuitry of social
the games was fully randomized to reduce systematic effects of order preferences. Trends in Cognitive Sciences 11, 419427.
on a particular task, the administration of multiple games might have Fehr, E., Schmidt, K.M., 1999. A theory of fairness, competition, and cooperation. The
inuenced the validity of the games and behavior accordingly. Quarterly Journal of Economics 114, 817868.
Forsythe, R., Horowitz, J., Savin, N.E., Sefton, M., 1994. Replicability, fairness and pay in
For instance people typically offer on average somewhere between experiments with simple bargaining games. Games and Economic Behavior 6,
35% and 40% of the total money amount (Camerer, 2003) and this was 347369.
also true for our participants. Therefore when playing as responders Green, M.F., Olivier, B., Crawley, J.N., Penn, D.L., Silverstein, S., 2005. Social cognition in
schizophrenia: recommendations from the measurement and treatment research
our participants received a relative large amount of unfair offers (6
to improve cognition in schizophrenia new approaches conference. Schizophrenia
times 20% and 8 times 10% of total). This might have surprised Bulletin 31, 882887.
participants who were rst asked to propose offers and as a result Greig, T.C., Bryson, G.J., Bell, M.D., 2004. Theory of mind performance in schizophrenia:
they might have rejected more unfair offers when they played as diagnostic, symptom, and neuropsychological correlates. The Journal of Nervous
and Mental Disease 192, 1218.
responders. Moreover, participants did not meet their partners, but Gottesman, I.I., Gould, T.D., 2003. The endophenotype concept in psychiatry: etymology
saw a picture of their partners instead which might have made the and strategic intentions. The American Journal of Psychiatry 160, 636645.
games less realistic. Nevertheless studies (Van 't Wout et al., 2005; Guth, W., Schmittberger, R., Schwarze, B., 1982. An experimental analysis of ultimatum
bargaining. Journal of Economic Behavior and Organization 3, 367388.
2006; Harle and Sanfey, 2007) in which subjects did not meet their Hans, S.L., Marcus, J., Henson, L., Auerbach, J.G., Mirsky, A.F., 1992. Interpersonal
partners reported similar behavior compared to those where behavior of children at risk for schizophrenia. Psychiatry 55, 314335.
participants rst met their partners. Future research might want to Harle, K.M., Sanfey, A.G., 2007. Incidental sadness biases social economic decisions in
the Ultimatum Game. Emotion 7, 876881.
use a design in which participants meet their partners while keeping Henry, J.D., Bailey, P.E., Rendell, P.G., 2008. Empathy, social functioning and schizotypy.
play behavior of partners under control as was done by Sanfey et al. Psychiatry Research 160, 1522.
(2003). Lastly, we tested only undergraduate students and this Jahshan, C.S., Sergi, M.J., 2007. Theory of mind, neurocognition, and functional status in
schizotypy. Schizophrenia Research 89, 278286.
reduces the degree of generalizability of our ndings and did not Kahneman, D., Knetsch, J., Thaler, R., 1986. Fairness and the assumptions of economics.
conrm severity of schizotypal symptoms. The Journal of Business 59, S285S300.
Future research might want to focus on factors that are known to Koenigs, M., Tranel, D., 2007. Irrational economic decision-making after ventromedial
prefrontal damage: evidence from the Ultimatum Game. The Journal of Neurosci-
inuence social interactive decision-making and test these in patients
ence 27, 951956.
with schizophrenia or people high on the schizophrenia continuum. Kohler, C.G., Brennan, A.R., 2004. Recognition of facial emotions in schizophrenia.
For instance, we recently have showed that the effortless processing Current Opinion in Psychiatry 17, 81.
of social cues inuences behavior in social interactions (Van 't Wout Langdon, R., Coltheart, M., 1999. Mentalising, schizotypy, and schizophrenia. Cognition
71, 4371.
and Sanfey, 2008) and we are currently testing whether patients with Lenzenweger, M.F., 1994. Psychometric high-risk paradigm, perceptual aberrations,
schizophrenia based their decisions to a lesser degree on such social and schizotypy: an update. Schizophrenia Bulletin 20, 121135.
cues. In addition, other measures of mentalizing or Theory of Mind, Marjoram, D., Miller, P., McIntosh, A.M., Cunningham Owens, D.G., Johnstone, E.C.,
Lawrie, S., 2006. A neuropsychological investigation into Theory of Mind and
executive functioning, planning and (social) reward processing in enhanced risk of schizophrenia. Psychiatry Research 144, 2937.
relation to aberrations in (social) bargaining. Penn, D.L., Ritchie, M., Francis, J., Combs, D., Martin, J., 2002. Social perception in
The main advantage of our experimental set-up was to address one schizophrenia: the role of context. Psychiatry Research 109, 149159.
Pickup, G., 2006. Theory of mind and its relation to schizotypy. Cognitive Neuropsy-
of the limitations mentioned by Agay et al. (2008) as they could not chiatry 2, 117192.
control the unfairness of the offers proposed and which made the low Pillutla, M.M., Murnighan, J.K., 1996. Unfairness, anger, and spite: emotional rejections of
rejection rates of patients difcult to interpret. Our observation of ultimatum offers. Organizational Behavior and Human Decision Processes 68, 208224.
Poreh, A.M., Whitman, R.D., Weber, M., Ross, T., 1994. Facial recognition in
unusual bargaining behavior comparable to recent ndings of Agay hypothetically schizotypic college students. The role of generalized poor perfor-
et al. (2008) in clinical patients with schizophrenia suggests that mance. Journal of Nervous and Mental Disorder 182, 503507.
aberrant interactive decision-making behavior could point to vulner- Raine, A., 1991. The SPQ: a scale for the assessment of schizotypal personality based on
DSM-III-R criteria. Schizophrenia Bulletin 17, 555564.
ability for serious mental illness. To our best knowledge, this is the
Raine, A., Benishay, D., 1995. The SPQ-B: a brief screening instrument for schizotypal
rst study that examines consequential interactive decision-making personality disorder. Journal of Personality Disorders 9, 346355.
behavior in psychosis-prone individuals. Sanfey, A.G., Rilling, J.K., Aronson, J.A., Nystrom, L.E., Cohen, J.D., 2003. The neural basis
of economic decision-making in the Ultimatum Game. Science 300, 17551758.
Scheres, A., Sanfey, A.G., 2006. Individual differences in decision making: drive and
Acknowledgments reward responsiveness affect strategic bargaining in economic games. Behavioral
This research was supported by a Rubicon grant from the Netherlands Organisation and Brain Functions 2, 35.
for Scientic Research (MvtW), and R03 MH077058-02 from the National Institute of Sprong, M., Schothorst, P., Vos, E., Hox, J., Van Engeland, H., 2007. Theory of mind in
Mental Health (AGS). schizophrenia: meta-analysis. The British Journal of Psychiatry 191, 513.
Toomey, R., Schuldberg, D., 1995. Recognition and judgment of facial stimuli in
schizotypal subjects. Journal of Communication Disorders 28, 193203.
References Toomey, R., Schuldberg, D., Corrigan, P., Green, M.F., 2002. Nonverbal social perception
and symptomatology in schizophrenia. Schizophrenia Research 53, 8391.
A.P.A., 2000. Diagnostic and Statistical Manual of Mental Disorders IV-TR, 4th ed. Van 't Wout, M., Aleman, A., Kessels, R.P.C., Cahn, W., de Haan, E.H.F., Kahn, R.S., 2007.
Revised. American Psychiatric Association Press, Washington, DC. Exploring the nature of facial affect processing decits in schizophrenia. Psychiatry
Agay, N., Kron, S., Carmel, Z., Mendlovic, S., Levkovitz, Y., 2008. Ultimatum bargaining Research 150, 227235.
behavior of people affected by schizophrenia. Psychiatry Research 157, 3946. Van 't Wout, M., Aleman, A., Kessels, R.P.C., Laroi, F., Kahn, R.S., 2004. Emotional processing
Baas, D., Van 't Wout, M., Aleman, A., Kahn, R.S., 2008. Social judgment in clinically in a non-clinical psychosis-prone sample. Schizophrenia Research 68, 271281.
stable patients with schizophrenia and healthy relatives: behavioural evidence of Van 't Wout, M., Kahn, R.S., Sanfey, A.G., Aleman, A., 2006. Affective state and decision-
social brain dysfunction. Psychological Medicine 38, 747754. making in the Ultimatum Game. Experimental Brain Research 169, 564568.
Baum, K.M., Walker, E.F., 1995. Childhood behavioral precursors of adult symptom Van 't Wout, M., Kahn, R.S., Sanfey, A.G., Aleman, A., 2005. Repetitive transcranial
dimensions in schizophrenia. Schizophrenia Research 16, 111120. magnetic stimulation over the right dorsolateral prefrontal cortex affects strategic
Bolton, G.E., Ockenfels, A., 2000. ERC: a theory of equity reciprocity and competition. decision-making. NeuroReport 16, 18491852.
The American Economic Review 90, 166193. Van 't Wout, M., Sanfey, A.G., 2008. Friend or foe: the effect of implicit trustworthiness
Camerer, C.F., 2003. Behavioral Game Theory. Russell Sage Foundation, New York. judgments in social decision-making. Cognition 108, 796803.
Corcoran, R., Mercer, G., Frith, C.D., 1995. Schizophrenia, symptomatology and social Vollema, M.G., Sitskoorn, M.M., Appels, M.C.M., Kahn, R., 2002. Does the Schizotypal
inference: investigating theory of mind in people with schizophrenia. Schizo- Personality Questionnaire reect the biological-genetic vulnerability to schizo-
phrenia Research 17, 513. phrenia? Schizophrenia Research 54, 3945.
Corrigan, P.W., Addis, I.B., 1995. The effects of cognitive complexity on a social Waldeck, T.L., Miller, L.S., 2000. Social skills decits in schizotypal personality disorder.
sequencing task in schizophrenia. Schizophrenia Research 16, 137144. Psychiatry Research 93, 237246.