Journal of Experimental Social Psychology 47 (2011) 1096–1106
Contents lists available at ScienceDirect
Journal of Experimental Social Psychology
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 / j e s p
People avoid situations that enable them to deceive others☆
Shaul Shalvi ⁎, Michel J.J. Handgraaf, Carsten K.W. De Dreu
Department of Psychology, University of Amsterdam, The Netherlands
a r t i c l e
i n f o
Article history:
Received 3 September 2010
Revised 20 April 2011
Available online 30 April 2011
Keywords:
Ultimatum game
Deception
Decision making
Exit option
Dishonesty
Bargaining
a b s t r a c t
Information advantage enables people to benefit themselves by deceiving their counterparts. Using a
modified ultimatum bargaining game with an exit option, we find that people are more likely to avoid settings
enabling them to privately deceive their counterparts than settings which do not enable deception. This
tendency is explained by people's reduced desire to become responsible for the other's outcomes when
deception is possible. Results of three experiments show that people avoid entering a setting that enables
deception by appearing fair while being unfair (Exp. 1–3). Experiment 2 showed that this tendency was
reduced when interaction partners were displayed as competitive rather than cooperative. Experiment 3
showed a stronger tendency to avoid tempting situations that enable private deception than to approach
situations in which one can privately benefit others. We conclude that when navigating through social space,
people avoid situations enabling them to deceive others.
© 2011 Elsevier Inc. All rights reserved.
Since the advent of behavioral game theory over 50 years ago (e.g.,
Deutsch 1985; Schelling 1960; Thibaut & Kelley 1959; von Neumann &
Morgenstern 1947), a great many studies in social psychology,
sociology, and behavioral economics have studied human decision
making in tightly modeled social exchange situations (for reviews see,
e.g., Camerer & Fehr 2006; Komorita & Parks 1995; Pruitt & Kimmel
1977; Weber et al. 2004). Typically, interaction partners are assumed to
be “in the situation” and their tendencies towards or away from
cooperation are being analyzed. Far less research has been targeted at
the questions how and why individuals end up in particular social
exchange situations in the first place — what are the drivers towards or
away from social interaction, and why do people avoid some social
exchange interactions while entering others? These and related
questions are the focus of the present studies. We assume that when
deciding to enter or avoid social exchange situations, people engage
mental operations, projecting and imagining themselves into future
states (Gilbert et al. 1998; McClure et al. 2004). They compare and
evaluate alternative states and possible scenarios (Roese 1997), consider
whether they should take responsibility for the fate and outcomes of
their prospective interdependent other (Handgraaf et al. 2008), and feel
attracted to the positive emotions, and aversive to the negative feelings
such future states may generate (Gilovich & Medvec 1995).
☆ This research was financially supported by grant NWO-400-07-701 to Carsten K.W.
De Dreu.
⁎ Corresponding author at: Department of Psychology, University of Amsterdam,
Roetersstraat 15, 1018 WB, The Netherlands.
E-mail address: [email protected] (S. Shalvi).
0022-1031/$ – see front matter © 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.jesp.2011.04.015
We propose that one fundamental cue that people use when
deciding to enter or avoid social exchange situations, is whether the
situation provides opportunities to serve personal interests in a
private and anonymous manner (i.e., without others being aware of
it). Possessing private information about the value of the goods
provides one bargainer with an advantage over the counterpart. Such
an information advantage allows the better informed person to
secretly serve self-interests at the expense of their counterpart: this
enables one to deceive another person by making offers that appear
fair while in fact being unfair.
Studies on Moral Hypocrisy (Batson et al., 1997; 1999) as well as
neoclassical economic theory would predict that individuals would be
attracted to all those situations in which they can appear fair while
(secretly) serving personal self-interest. For example, individuals are
attuned to the expectations of others. Dana et al. (2006) asked
participants to choose whether and how they wanted to split $10
between themselves and another person, or instead take $9 as an exit
option to avoid playing the game altogether (in which case the
prospective partner would not be informed about the game). Across
two studies, Dana et al. found that about one-third of the participants
exited the $10 dictator game and opted to take $9 instead. Because
choosing the exit option was personally costly (i.e., one could have
played the game and kept the $10 to oneself) these results suggest
that people are concerned with not violating others' expectations
about the offers they might receive, and preferred to pay a small “fine”
for this.
An alternative perspective on the Dana et al. results proceeds on
the basis of the assumption that people avoid situations that create
tension between (privately) serving their self-interest and being both
privately and publicly fair and considerate. Indeed, there is ample
S. Shalvi et al. / Journal of Experimental Social Psychology 47 (2011) 1096–1106
evidence that decision makers take into account their own immediate
and personal gains and losses as well as the outcomes of others (e.g.,
De Dreu et al. 2000; Handgraaf et al. 2008; Loewenstein et al. 1989;
Messick & Sentis 1979; Van Beest & Van Dijk 2007). As a case in point,
consider research using the ultimatum bargaining game, in which a
proposer offers a division of a commodity (e.g., money), and a
responder can accept or reject the proposed division. If the responder
accepts, the commodity is divided as proposed; if the responder
rejects, both parties receive nothing (Güth et al. 1982). Studies
consistently show that people propose distributions that approximate
fairness (i.e., a 50/50 split of the resource) and tend to reject
distributions that substantially deviate from fairness (e.g., Handgraaf
et al. 2003a,b; Oosterbeek et al. 2004). Clearly, proposers take into
account responders' outcomes, and responders not only consider
what they would get but also the outcome to be received by the
proposer. Whether grounded in reputation concerns or in genuine
concern for others, people value fairness and a self-image of being a
morally appropriate, and socially considerate human being (Haidt
2007; Tabibnia et al. 2008; Tangney et al. 2007).
Recent work further indicates that people are to some extent lieaversive even when lying advances their personal interests (Mazar
et al. 2008; Mead et al. 2009; Shalvi et al. 2011; Shalvi, Dana,
Handgraaf & De Dreu, 2011; Steinel & De Dreu 2004). Even under
complete anonymity, when reputation concerns and external punishment are excluded, people restrict the extent to which they lie
(Mazar et al. 2008), and lie only to the extent that reasonable selfjustifications are available (Shalvi et al. 2011). Along similar lines,
people are less likely to deceive another person to benefit themselves
the higher the negative consequences for the partner's outcome are
(Gneezy 2005). This suggests that people balance the psychological
cost of being dishonest against the financial gain the lie produces (see
also Lundquist et al. 2009). Clearly people are sensitive to other's
outcomes and such other-concerns provide input into people's social
decision making even under conditions of complete anonymity.
While clearly showing that other-concerns may lead people away
from pursuing their self-interest, the question remains how such otherconcerns influence people's likelihood to engage in social interactions
and exchange in the first place. To answer this question, we employed
an asymmetric information version of the ultimatum bargaining game
(Kagel et al. 1996). In this version of the game, participants as proposers
could exploit their information advantage about the value of the goods
they possessed (chips to be converted to money) and propose offers that
appeared fair to their partner while being actually unfair in monetary
terms. We gave participants the opportunity to play the game or to exit
at a personal cost (similar to Dana et al. 2006). Accordingly, we test
whether in settings in which partners lack knowledge about the value of
the goods, people avoid or enter interactions that enable them to exploit
the other by appearing but not being fair.
As argued above, people may be relatively aversive of, rather than
attracted to, situations that enable them to exploit their information
advantage and deceive others into accepting offers that seem fair
while in fact they are self-serving to the proposer. Providing further
support to this reasoning, work on moral dilemmas found that when
people are confronted with a morally challenging situation they are
rather averse from taking action that would make them responsible
for others' outcomes (Haidt 2007). For example, people consider
saving the lives of five other people by killing one person differently as
a function of whether or not they have to actively take an action that
would make them feel directly responsible for the outcome. This is
why people are more likely to switch a lever and direct a trolley into a
track that would kill one person and prevent the death of five other
people, but are less likely to push this one person off a bridge in order
to save the lives of the five others. Pushing the person off the bridge
makes people feel responsible for this one person's (ill) fate (see
Greene & Haidt 2002 for a review), and this shows up in general
negative effect, and feelings of guilt and shame in particular (e.g.,
1097
Baumeister et al. 2007; Ritov & Baron 1995; Van Kleef et al. 2010;
Zeelenberg et al. 1998). From this work it follows that one particular
reason tempting social exchange situations may be avoided is because
the responsibility over another's fate is aversive.
Taken together, people's social decisions are influenced substantially by considerations of fairness and other-concern. People
experience psychological costs when lying and deceiving others
and, especially in morally tempting situations, they seek to avoid
being or becoming responsible for others' outcomes. Accordingly, we
predicted (H1) that compared to situations that do not enable people
to deceive their partner, people avoid situations that enable them to
exploit their information advantage and deceive their partner to
accept an offer that may seem but is not fair. Second, we predicted
(H2) that settings that enable people to exploit their information
advantage reduces people's willingness to become responsible for
their partner's outcome compared to settings which do not enable for
such exploitation. Third, we predicted (H3) that people's decision to
avoid situations enabling them to exploit other's information
disadvantage is mediated by their reduced willingness to be(come)
responsible for their partner's outcomes. These hypotheses were
tested in Experiments 1–3; In Experiment 3 we also examined
the extent to which moral emotions like guilt and shame relate to the
aversion of becoming responsible for other's outcomes, and the
decision to enter or avoid particular social exchange situations.
If our reasoning holds, and people's reduced willingness to become
responsible for the other's outcome is driving avoidance of situations
that enable deception, one key moderator that may influence people's
willingness to become responsible for another person's outcome are
cues about who this other person is. In bargaining settings, one
relevant cue is whether this individual is a cooperative or a
competitive person. People generally feel more positively towards
cooperative than competitive people (Vonk 1998), trust cooperative
people more (Kelley & Stahelski 1970), and are more honest and less
deceptive towards cooperative rather than competitive others (Rubin
et al. 1994; Steinel & De Dreu 2004). Accordingly, facing a cooperative
partner should make entering a tempting social setting particularly
aversive. That is, when one considers engaging in a tempting social
interaction – one that allow this person to deceive the counterpart to
benefit the self – learning that this other person is a cooperative
person may make it particularly aversive to take responsibility for this
person's outcome by engaging in interaction. In contrast, facing a
competitive partner should make entering the tempting social setting
less aversive because people feel less positively about competitive
others, and deceptive strategies are easier to legitimize (Steinel & De
Dreu 2004). Put differently, balancing the desire to profit (financially)
with the desire to behave honestly when deciding whether to engage
in interaction, a competitive counterpart may evoke less aversion of
becoming responsible for this person's outcome, even in tempting
settings. Accordingly, Experiment 2 tested our final prediction (H4)
that people avoid entering a tempting setting more than a less
tempting setting especially when the counterpart is cooperative;
when the counterpart is competitive, the tendency to avoid entering a
tempting setting will be attenuated.
Experiment 1
To test our hypotheses we modified ultimatum bargaining games
to create a low and a high temptation game (LT and HT, respectively).
We asked participants to enter the situation to make an offer or,
alternatively, to take an exit value (see Dana et al. 2006). Participants,
as proposers, had to allocate 30 chips. However, only the proposers
were aware of the chip value, the responders were not. In the LT
condition, each chip was worth 2 monetary units (MU; 1 MU = 0.10
€ ≈ US$ 0.14) to the proposer and 2 MUs to the responder; in the HT
condition however, each chip was worth 2 MUs to the proposer and 1
MU to the responder (Kagel et al. 1996). Under LT, a 15:15 split thus
1098
S. Shalvi et al. / Journal of Experimental Social Psychology 47 (2011) 1096–1106
implies an equal monetary outcome for proposer and responder
(2 × 15 = 30 MUs to both), but under HT, an unequal outcome
emerges (i.e., 15 × 2 MU = 30 MUs to proposer and 15 × 1 MUs = 15
MUs to responder). Since a 15:15 split is generally taken as fair and
likely to be accepted by responders (Kagel et al., 1996), HT proposers
have the option of serving self-interest by appearing (but not being)
fair. Compared to LT, the HT situation is thus equally risky, more fitting
to maximize self-interest, yet is also more morally challenging. 1
One of the dominant reference points in an UBG offer is the fair split
of the chips. To explore whether the exit option's value influences
entrance decisions, we manipulated the exit value that was proposed
to participants to be one chip higher (vs. lower) than the equal chip
split (15). Obviously, the exit's value would make a difference in
extreme cases; a very low exit option will lead more people to enter
any of the games, while a very high exit option will lead most people
not to. However, in Experiment 1 we focused on a rather subtle
manipulation of the exit option, assessing if people's entrance decision
will be influenced by whether the exit enables them (or not) to secure
what is commonly perceived as a fair (50%–50%) offer that is likely to
be accepted, without engaging in the interaction. This subtle
manipulation was included in this initial study chiefly to calibrate
the exit option we offered participants as an alternative to entering the
game and making an offer.
Additionally, in Experiment 1 we took some measures to address
several alternative explanations that might arise because of some
differences between the HT and LT games. First, it is more difficult to
calculate what a fair offer is in the HT game than in the LT situation
(e.g., a 20:10 split in HT vs. a 15:15 split in LT). We measured the time
taken to make the entrance decision: If calculation complexity plays a
role, we should find that entrance decisions are made slower in the HT
rather than the LT game. Second, people may decide to avoid the HT
game more than the LT game because the former confronts them with
a (distributionally) unjust setting more than the LT situation. We
measured the extent to which participants felt that the situation
enabled them to make a fair offer. If people avoid the HT game because
they feel that it limits their ability to make a fair offer, we should find
differences on this measure between the two settings.
Task and procedure
Participants were seated in front of computers in separate rooms and
cubicles, preventing them from seeing or communicating with each
other. Because doors were kept closed throughout the session
participants did not know how many people were present in the lab
at any given moment. Participants were asked to read instructions
explaining the rules of the UBG. They read that they were assigned to the
role of (potential) proposer, and that they would get the option to divide
30 chips, each worth 2 MU to them and, depending on experimental
condition, 1 MU or 2 MU to the other person (in the HT vs. LT,
respectively). They were told that if the responder accepted their offer,
the chips would be divided as proposed. If the responder rejected the
offer, neither participant would receive any money. Participants were
further told that if they decided to make an offer, the responder would
learn the total number of chips divided (i.e., 30) but would not learn the
value each chip held for either proposer or responder. We informed all
participants that every fifth dyad in the experiment would be paid
according to their actual decisions.3 They were also told that they would
remain anonymous to each other at all times.
Hereafter, we gave participants the possibility to avoid the UBG by
telling them that, instead of making an offer, they could also take a
fixed amount of chips (14 vs. 16, depending on condition). If they
choose not to make the offer, the responder would receive no
information about the game, meaning that these others would never
learn of the possibility of a division of chips taking place. The
computer then prompted the participant make an entrance decision
and, thereafter, to respond to a short questionnaire. This completed
the experiment. Participants were debriefed and paid.
Dependent variables
The main dependent variable was whether participants choose to
enter the interaction. To assess calculation complexity, we measured the
time people needed to make their entrance decision. To assess the
mediator, the aversion of becoming responsible of the other's outcomes
we asked participants to what extent they were willing to “be responsible
for the other person's outcome” (1 = not at all to 7 = very much). Finally,
participants responded to several items assessing their understanding of
the decision situation, and whether they felt that the situation enabled
them to make a fair offer (1 = not at all to 7 = very much).
Method
Results
Design and participants
We used a 2 (temptation level: high vs. low) 2 × 2 (exit option: 16
vs. 14 chips) between subjects factorial design. Ninety-one students
(75% females) participated as part of a course requirement. They
received €7 for participation, and were randomly assigned to an
experimental condition (n = 22–26 per condition). Gender had no
effects in this experiment (or in any of the other studies) and is not
addressed any further.
1
Note that these two games are identical on the level of risk one takes when making a
given offer. Strategic risk-related decisions are based on (1) the profit the offer yields
relative to the available exit value, and (2) the likelihood of the offer being accepted. Both
LT and HT games were intentionally equal on these two factors. First, proposers' chip value
was held constant at 2 MUs in both games. Thus, the profit a given (accepted) offer yields
to the proposer is equal in both games. Second, because responders were unaware of chip
values at all times, a fact known to proposers, the proposers' assessment of the likelihood
of an offer being accepted is equal across conditions.
2
To verify that indeed the HT situation evokes higher levels of temptation
compared with the LT situation we introduced to a difference sample of 60 students
one of the two games. We asked them about the extent to which they felt that the
situation described created a moral challenge (e.g., “I felt that my morality was tested”,
“deciding whether to make an offer or not was morally challenging”, “the decision is
morally difficult”; 10 items, α = .89). Confirming that our manipulation was
successful, participant who evaluated the HT situation found it more morally
challenging (M = 3.66, SD = 1.04) that those evaluating the LT situation (M = 3.06,
SD = 1.12), t (58) = 2.10, p b .05.
Manipulation checks
Participants indicated that if an offer was accepted the chips would
be divided as offered, and that if an offer was rejected both
participants would receive nothing. They also indicated the value of
the chips to themselves and the responders, and that these values
would not be known to their partner. We excluded from all reported
analyses those 13 participants who failed to answer any of these
questions correctly. However, including these participants did not
change the results or main conclusions.
Entrance behavior
A logistic regression was used to determine whether the proportion of participants entering the bargaining situation varied as a
function of experimental condition. Results supported Hypothesis 1:
The decision to make an offer (dummy coded to hold 1 for entering
and 0 for avoiding) was affected by temptation condition, B = −1.58,
3
In order to pay participants according to their actual decisions, we asked all
participants at the end of the experiment to indicate the lowest amount of chips they
would accept as responders. For participants who chose to make an offer we compared
their actual offer to the minimal acceptable amount set by the participant that was
seated in their cubicle in the previous session. Participants who chose not to divide
were paid according to the number of chips offered for not dividing (14 or 16 chips,
depending on condition). Those participants, who were selected to receive pay, were
paid an additional €3.5 on average.
S. Shalvi et al. / Journal of Experimental Social Psychology 47 (2011) 1096–1106
Wald's χ 2 (1, N = 78) = 3.88, p b .05. Regardless of the exit option (14
or 16 chips) the percentage of participants who entered the social
interaction was higher in the LT condition (86.5%), than in the HT
condition (58.5%). This pattern was present both when 14 chips were
offered as an exit option, χ 2 (1) = 3.67, p = .055 (marginal), and when
16 chips were offered as an exit option, χ 2 (1) = 4.16, p b .05. The
effects for exit value (B = .78, Wald's χ 2 (1, N = 78), ns.) and the
interaction between exit value and temptation level (B = .26, Wald's
χ 2 (1, N = 78), ns.) were not significant.
Outcome responsibility
To test whether aversion of becoming responsible for the other's
outcomes mediates the effects of temptation on the entrance
decisions, a series of regression models were estimated using the
procedures in Baron and Kenny (1986). We standardized the logistic
regression coefficients to enable mediation analysis with binary
dependent variables (MacKinnon & Dwyer 1993). We submitted to a
regression analysis the temptation level, exit value and the interaction
between them as predictors of the willingness to become responsible
for the other's outcome. Supporting Hypothesis 2, we found only an
effect for temptation predicting aversion of becoming responsible
(β = −.71, t (77) = −3.35, p b .01). Participants in the HT condition
reported lower desire to be in responsible for the other's outcomes
(M = 3.90, SD = 1.51) than those in the LT condition (M = 4.94,
SD = 1.20). A logistic regression analysis predicting the decision to
enter from both responsibility and temptation further showed that
while responsibility influenced the decision to enter (B = .79,
SE = .32; Wald's χ² (1, N = 78) = 6.23, p b .025) the effect of temptation on entrance rates was significantly reduced (B = −1.06, SE = .61;
Wald's χ² (1, N = 78) = 2.96, p = .09), Sobel Z = − 1.60, p = .05. In
other words, in support of Hypothesis 3, the aversion of being
responsible for the other's outcome fully mediated the relation
between temptation level and the entrance decision (see Fig. 1). 4
Alternative explanations
One possible interpretation to our results is that proposers avoided
making an offer in the HT more than the LT setting because in the HT
setting it is more difficult to calculate what a fair offer is (i.e.,
calculating a split of chips in 2:1 vs. 2:2 ratios). This alternative
explanation would suggest that people would require more time to
take the decision of whether to make an offer in the HT compared to
the LT settings as they must engage more mental capacity in figuring
out the situation. However, a 2 (temptation: high vs. low) × 2 (exit
value: 14 vs. 16 chips) ANOVA with (log-transformed) decision time
revealed no significant effects for temptation, F (1, 77) = 3.06, ns., exit
value, F (1, 77) = .09, ns., or their interaction, F (1, 77) = 1.67, ns. On
average, participants spent 6.01 s (SD = 5.91) on deciding whether to
move into the bargaining game. 5
Another possibility is that the HT situation leads participants to
feel more restricted in their ability to make a fair offer. This idea
suggests that people avoid social situations that make them feel
restricted in their ability to make a fair offer, due to the unjust payoff
4
We note that the focus of our investigation is on studying a situational factor, the
level of temptation embedded in the structure of interdependency, and how it
activates the willingness to take responsibility over other's fate in turn influencing the
decision to entering interdependent interaction. As such, we do not study
interdependent outcome preferences from an individual difference perspective (e.g.,
Social Value Orientation; De Dreu & Boles, 1998; De Dreu & Van Lange, 1995; Kuhlman
& Marshello, 1975; Van Dijk et al., 2004; Van Lange, 1999) but rather from a situational
perspective. It might be argued that especially individuals with a cooperative
orientation are aversive of tempting situations, and thus in all three experiments we
measured social value orientation (using the measurement and procedures described
in De Dreu & Boles, 1998). In none of the experiments we found any effects of social
value orientation, alone or in interaction with other experimental variables, suggesting
that social value orientation does not differentially affect entrance behavior.
5
Reaction time for taking the decision of whether to enter or not was additionally
measured in Experiments 2 and 3 and never produced any significant effects.
1099
structure. Again, results did not support this idea. A 2 (temptation:
high vs. low) × 2 (exit value: 14 vs. 16 chips) ANOVA with the ability
to make a fair offer as dependent variable revealed no significant
effects for temptation (MHT = 5.13 vs. MLT = 5.62), F (1, 77) = 1.97, ns.,
exit value (Mlow = 5.27 vs. Mhigh = 5.48), F (1, 77) = .38, ns., or their
interaction, F (1, 77) = 1.72, ns. Thus, participants did not feel that the
HT game limited their ability to make a fair offer more than the LT
game; in both conditions they understood that making a fair offer was
possible. We additionally ran all mediation analyses detailed above
while controlling for the extent to which participants felt that they
could have made a fair offer. This did not change the results.
Discussion and introduction to Experiment 2
Experiment 1 supported Hypothesis 1 that people avoid tempting
social situations even when doing so comes at a cost to oneself. This
general tendency was independent of the exit option manipulation,
and was not driven by calculation complexity or one's feeling of being
able to make a fair offer. Results also supported the outcome
responsibility (H2) and the mediation (H3) predictions that, when
confronted with tempting situations, people are less willing to
become responsible for the other's outcomes, and therefore decide
to avoid the interaction altogether.
In Experiment 2, we provided proposers with information about
the other person's cooperativeness or competitiveness and tested
Hypothesis 4 that facing a cooperative rather than competitive
partner makes tempting situations particularly aversive and promotes
avoidance of such situations. We reasoned that since people's reduced
willingness to become responsible for the other's outcome drives
avoidance of situations that enable deception, cues about who this
other person is should moderate the relation between temptation and
avoidance. In bargaining settings the other's level of cooperativeness
vs. competitiveness is highly relevant cue — people are generally less
positive towards competitive people (Vonk, 1998), and are less
aversive to the idea of deceiving them (Steinel & De Dreu, 2004). We
thus provided participants with information about the potential
counterpart's personality and assessed whether indeed, facing a
competitive counterpart attenuates the tendency to avoid the socially
tempting situations.
Method
Design and participants
We used a 2 (temptation level: high vs. low) × 2 (responder's
putative personality: cooperative vs. competitive) between-subjects
design with 76 undergraduate students (71% females) as participants.
Participants received €7 for participation, and were randomly
assigned to conditions (n = 19 per cell).
Procedure and task
The procedure and task were identical to Experiment 1 except that
we manipulated the responder's putative personality to be cooperative or competitive (Steinel & De Dreu, 2004; Van Kleef, De Dreu &
Manstead, 2006). At the beginning of the experiment, participants
completed a 20 item Personality Test allegedly measuring collaboration skills. The items addressed the extent to which one tends to
cooperate in daily life (e.g., “In the bus, I vacate my seat for older
people”; “I enjoy working with other people”; “Love and respect are
more important than status and money”; “Winning is everything”; “I
like situations in which it is me against someone else”). Participants
were asked to indicate their agreement on 5-point Likert scales (1 =
strongly disagree to 5 = strongly agree).
Upon completion of the test, participants received the same instructions as in Experiment 1 (here we offered all participants 16 chips as
exit option). Shortly before the actual decision making, participants were
told that proposers would receive feedback about their responder's test
1100
S. Shalvi et al. / Journal of Experimental Social Psychology 47 (2011) 1096–1106
100
Tempting Setting
-1.06(-1.47**)
90
.79**
Enter
Social Exchange
Situation
Entrance (%)
-.71**
Willingness to
become Responsible
for Other’s Outcome
Game
80
HT
LT
70
60
50
Fig. 1. Becoming responsible for the other's outcomes as a mediator of the setting's
temptation level (game) on the decision to enter social exchange situations (Exp. 1).
** p b .01. Note: the direct effect of the setting's temptation level on the decision to enter
is in parentheses.
Cooperative
Competitive
Responder
Responder
Fig. 2. Entrance (percentage) as a function of the setting's temptation level (game) and
responder's orientation (Exp. 2).
results, whereas the responders would not (Steinel & De Dreu, 2004). In
the cooperative partner condition, participants received a feedback sheet
that suggested a rather cooperative profile for their partner. In the
competitive partner condition, they received a feedback sheet suggesting
a rather competitive profile. Specifically, on the partner's feedback sheet,
participants saw all the answers the partner gave and a total score
indicating a cooperative or a competitive orientation.
Dependent variables
In addition to the measures from Experiment 1, we checked the
adequacy of the manipulation of other's personality by asking
participants, at the very end of the experiment, to indicate the extent
to which their partner was competitive or cooperative (1 = very
competitive, to 7 = very cooperative).
Results
Manipulation checks
Using the comprehension checks from Experiment 1, 14 participants who failed to correctly answer any of these questions were
excluded from all reported analyses those. Again, however, including
them did not change the reported results.
The manipulation check for partner's personality was analyzed in a
2 (temptation: high versus low) × 2 (responder's orientation: cooperative vs. competitive) ANOVA. This analysis only showed a main
effect for other's personality: Participants in the cooperative partner
condition rated their partner as more cooperative and less competitive than those in the competitive partner condition, M = 6.48 versus
M = 1.93, F (1, 61) = 289.25, p b .001. Neither the effect for temptation
level nor the interaction term were significant, both F (1, 61) b 1, ns.
Entrance behavior
We used a logistic regression with temptation, responder's
orientation and their interaction as independent variables, and
entrance decision as criterion. Neither the main effect for temptation
level, B = − 1.34, Wald's χ 2 (1, N = 62), ns., nor responder's
personality, B = 1.27, Wald's χ 2 (1, N = 62), ns., were significant. As
expected, however, supporting Hypothesis 4, the interaction term
between temptation and responder's personality was significant, B =
−2.62, Wald's χ 2 (1, N = 62), p b .05. Fig. 2 shows that with a
cooperative partner, fewer participants entered into the HT than the
LT game (64.3% versus 86.7%). But with a competitive partner, more
participants chose to enter in the HT than LT game (87.5% versus
64.7%).
Discussion and introduction to Experiment 3
Experiment 2 showed that participants entered the LT game more
than the HT game, but only when confronted with a cooperative
responder. Indeed, as predicted, confronted with a cooperative
counterpart, people avoided engaging in tempting interactions with
the outcome responsibility they entail. This corroborates our
suspicion that, when navigating through social space, people avoid
those settings that enable (and tempt) them to violate standards of
fairness, morality, and social considerateness. Only when the other
appears competitive, it becomes less averse to become responsible for
the other's outcomes and indeed the tendency to avoid tempting
settings was attenuated. In fact, participants were no longer avoiding
the tempting setting but were rather attracted to them. This suggest
that some cues may legitimize entering tempting (and potentially self
benefiting) situations. When people balance between their competing
desires to profit (financially) and maintain being and feeling honest,
cues that may liberate them from the need to be honest may actually
steer them towards, rather than away from, tempting settings.
An interesting result in Experiment 2 was that the proportion of
people entering interaction with a cooperative person resembled the
data obtained in Experiment 1 in which no information about the other
person was provided. This suggests that participants in our experiments
either considered the other as cooperative by default or, and perhaps
more likely, that the manipulation of interacting with a competitive
other is stronger than the manipulation of interacting with a cooperative
other. The latter possibility is more likely as past research suggests that
information about other's competitiveness had more impact on people's
impressions (De Bruin & Van Lange, 1999) as well as their behavioral
choices (De Dreu & Van Kleef, 2004; De Dreu, Yzerbyt & Leyens, 1995)
than information suggesting cooperativeness. Experiment 3 was
designed to achieve three goals. First, in addition to the aversion of
being responsible for the other's outcomes, the level of temptation
present in the situation evokes moral emotions that may mediate
decision making (De Hooge, Zeelenberg, & Breugelmans, 2007;
Schweitzer & Gibson, 2008). Specifically, when individuals consider
entering high (rather than low) temptation situations they may also
anticipate greater feelings of guilt and shame. This in turn may lead
them to avoid such social situations. Accordingly, in Experiment 3 we
assessed anticipated negative moral emotions and expected the HT
game to evoke anticipated moral emotions more than the LT game. We
explored whether such moral emotions mediate entrance decisions
over and above the impact of outcome responsibility avoidance.
Thus far, we compared high with low temptation situations, and
uncovered that people are averse of becoming responsible for the
other's outcomes and thus of entering HT situations. What is unclear,
however, is whether avoidance of the HT situation is driven by the
motivation to do the other no harm (e.g., Baron, 1995) or, alternatively,
that the approach of the LT situation is motivated by the ability to
benefit the other. Accordingly, we created a third game called
Generous. In Generous, the responder's chip value was higher (3 MU)
S. Shalvi et al. / Journal of Experimental Social Psychology 47 (2011) 1096–1106
than the proposer's (2 MU). As a result, a fair split of the chips (15:15)
would lead the responder to walk away with more (15 × 3 MU = 45
MU) than the proposer (2× 15= 30), thus enabling the proposer to
appear fair and to be (privately) generous. If entrance behavior is
driven primarily by a desire to benefit others, people might be eager to
avoid the HT game and to enter the Generous game even more than the
LT game. However, if entrance decisions are geared primarily towards
avoiding being unfair and immoral, we would expect a stronger
tendency to avoid the HT game than to enter both Generous and LT. We
expected the latter to be the case, as research has shown that people are
more reluctant to hurt others than eager to benefit others (Baron, 1995;
Leliveld, Van Dijk, & Van Beest, 2008; Van Beest, Van Dijk, De Dreu &
Wilke, 2005).
A final and more exploratory objective in Experiment 3 was to
assess the offers that are made by proposers who choose to enter the
game. Interpreting offer behavior in voluntary settings requires a base
line condition in which proposers are forced to enter (as is common
practice in game-experimental research). We thus manipulated
entrance to be either voluntary (as in Experiments 1 and 2) or forced.
Past research on forced entrance conditions found that in asymmetric
UBG conditions, in which only proposers knew the chips' value,
proposers tended to be strategic and make offers that appeared fair in
terms of chips, but were self-servingly unfair in terms of money
(Boles, Corson & Murninghan, 2000; Pillutla & Murnighan, 1995;
Pillutla & Murnighan, 2003; Van Dijk, et al., 2004; Van Dijk & Vermunt,
2000). We explored whether the strategic behavior observed under
forced conditions replicates when people voluntarily enter the game.
Method
Design and participants
We used a 3 (temptation level: high vs. low vs. generous) × 2 (entry
conditions: voluntary vs. forced) between-subjects factorial design with
132 students (75% females) as participants. Participants received €7 for
participation, and were randomly assigned to conditions (n = 25–31 per
voluntary condition; n = 15–16 per forced condition).
Procedure and task
The procedure and task were identical to those of Experiment 1
with 16 chips as exit option. The only difference in the current
experiment was that unlike the proposers in the voluntary entrance
condition, those in the forced entrance condition did not receive the
option of not making an offer. After reading the instructions of the
game, they were instructed to provide an offer to the responder. In
both conditions, participants then answered a short questionnaire,
and were debriefed and dismissed.
Dependent variables
The main dependent variable was again the decision of whether to
make an offer or not. Additionally, we asked the participants in the
voluntary entrance condition who chose to make an offer, as well as
those in the forced entrance condition, to indicate the number of chips
(between 0 and 30) they offered the responder. In the post-task
questionnaire we again assessed participants' willingness to become
responsible for their counterpart's outcomes (3 items, α = .86) and
added a series of questions on anticipated moral emotions. Specifically, participants were asked indicate to what extent they anticipated
that their decision to make an offer would make them feel guilty,
ashamed, embarrassed, and comfortable (reverse coded), (1 = very
little; to 5 = very much). A factor analysis revealed that all four items
loaded on a single factor explaining 45.4% of the variance, EV = 1.816,
all variable loadings N.60, and ratings were averaged into one index of
anticipated moral emotions (Cronbach's α = .62). Finally, as in
Experiment 1, we asked participants for the extent to which they
felt that “it was possible to make a fair offer”.
1101
Results
Manipulation checks
Analyses of the comprehension checks also used in Experiments 1
and 2 revealed that 21 participants failed to correctly answer any of
these questions. These were excluded from all reported analyses.
Including them did not change the results.
Entrance behavior
Our hypotheses pertaining to entrance behavior could only be
tested in the voluntary conditions. Using a logistic regression we
tested whether individuals avoided the HT game more than the other
games by contrast coding temptation level (.667 for HT and −.333 for
LT and G). Supporting Hypothesis 1, the percentage of participants
who chose to enter the situation was lower in the HT game (54.2%)
than in the LT and G games combined (77.2%), B = −1.08, Wald's χ 2
(1, N = 68), p = .05. People entered the HT game less than the LT game
(54.2% versus 81.8%, p b .025); entrance to the HT game was
somewhat lower than the G game (54.2% versus, 72.7% p b .10). This
pattern is consistent with the idea that people are primarily motivated
by avoiding situations that might tempt them to wrongdoing, and
inconsistent with the idea that people enter the LT game more
because it enables them to be fair (indeed, people entered the G game
as much as the LT game, B = −.93, Wald's χ 2 (1, N = 68), ns.).
Mediation analysis: being responsible or moral emotions?
To test whether aversion of being responsible and/or anticipating
moral emotions mediate the effects of temptation on entrance decisions,
a series of regression models were estimated (Baron and Kenny, 1986)
while standardizing the logistic regression coefficients as required when
assessing mediation with binary dependent variables (Mackinnon &
Dwyer, 1993). The predictor variable in the models was temptation level
(contrasting the HT game with the other games), the mediators were
aversion of being responsible and anticipated moral emotions (mediators were not correlated with one another, r (68)= −.05, ns.), and
entrance decision was the dependent variable. Replicating the results of
Experiment 1 and further supporting Hypothesis 2, a regression analysis
revealed that temptation predicts aversion to being responsible (β =
−.75, t (67) = −2.26, p b .05). Participants in the HT game reported
lower willingness to be responsible for the other's outcome (MHT=
4.04, SDHT = 1.49) than those in the other games (MLT = 4.83,
SDLT= 1.06; MG = 4.76, SDG = 1.37). Similarly, temptation level predicted anticipated moral emotions, (β = .20, t (67) = 2.25, p b .05).
Participants in the HT game anticipated moral emotions to a greater
extent (MHT= 2.32, SDHT = .48) than those in the other games (MLT =
1.91, SDLT = .40; MG = 1.92, SDG = .39).
As shown in Fig. 3, a logistic regression predicting entrance
decisions from temptation, responsibility aversion, and anticipated
moral emotions, showed that responsibility aversion influenced
entrance decisions (B = .72, SE = .24; Wald's χ² (1, N = 68) = 9.05,
p b .01) and the effect of temptation on entrance decision was reduced
to non-significance (B = −.72, SE = .64; Wald's χ² (1, N = 68) = 1.25,
ns.), directional Sobel Z = − 1.80, p b .05. Anticipated moral emotions
did not exert any additional effect (B = .29, SE = .81; Wald's χ²
(1, N = 68) = .12, ns.). These results provide additional support for
Hypothesis 3, suggesting that reduced desire for outcome responsibility mediates the relation between temptation level and the
entrance decision and render an explanation in terms of anticipated
moral emotions less likely.
Offers
We analyzed the chips offered by the proposers who volunteered
into the game and chose to make an offer as well as those made by
proposers in the forced play condition. We used an ANOVA with
3 (temptation: high vs. low vs. generous) × 2 (entry conditions:
1102
S. Shalvi et al. / Journal of Experimental Social Psychology 47 (2011) 1096–1106
-.75*
Willingness to
become Responsible
for Other’s Outcome
-.72 (-1.06*)
Tempting Setting
.20*
.72**
Enter
Social Exchange
Situation
Moral
.29
Emotions
Fig. 3. Becoming responsible for the other's outcomes as a mediator of the setting's
temptation level (game) on the decision to enter social exchange situations (Exp. 3).
* p = .05, ** p b .01. Note: the direct effect of the setting's temptation level on the
decision to enter is in parentheses.
voluntary vs. forced) as the independent variables and number of
chips offered as the dependent variable.
Results show that neither the main effect for game nor the main
effect for entry condition were significant, both F's (1, 89) b 1.30, ns.
However, the interaction term was significant (F (1, 89) = 3.03, p b .05,
η 2 = .07). As can be seen in Fig. 4, replicating previous findings (Boles,
et al., 2000; Pillutla & Murnighan, 1995; Van Dijk, et al., 2004; Van Dijk
& Vermunt, 2000), temptation had no effect in the forced-play
conditions, F (1, 89) = .29, ns. Proposers made similar offers (which
carry similar likelihood of acceptance) in HT (M = 12.60, SD = 4.12),
LT (M = 13.38, SD = 2.80), and G (M = 13.50, SD = 5.03) games. In the
voluntary entrance condition, however, proposers offered more chips
to the responder in HT (M = 15.85, SD = 4.06), than in LT (M = 12.17,
SD = 3.38) and G (M = 13.19, SD = 1.72), F (2, 89) = 3.98, p b .05. A
Dunnett's post hoc analysis revealed that within the voluntary
entrance condition, the amount of chips offered by the proposers in
HT was higher than the amount offered by proposers in LT (D = 3.86,
p b .01) and G (D = 2.66, p = .05). The offers made in G did not differ
from those made in LT. We discuss these exploratory findings and
their limitations in the discussion below.
We further recoded the offers to four categories: lower than 14
chips, equal to 15 chips (seem fair), range between 16 and 19 chips,
equal to 20 (fair in monetary terms in the HT condition; no offers were
higher than 20 in our sample). We used these categories as they
allowed us to further assess whether entering to the HT situation
leads people to make more offers that are actually fair and not only
18
17
Game
16
Chips offered
seem fair. Indeed, as can be seen in Table 1, while in the LT and G
conditions, the distributions of the voluntary and forced conditions
did not differ than one another (LT: χ² (3, N = 34) = 1.72, ns.; G: χ²
(2, N = 28) = 4.6, ns.) in the HT condition the distributions of offers
did, χ² (3, N = 28) = 8.49, p b .05. Specifically, proposers that volunteered into the HT condition were more likely (31%) to make offers
that were fair in monetary terms (i.e., offer 20 chips) than offers that
seemed fair (i.e., offer 15 chips; 15%), while proposers who were
forced into the HT condition were more likely to make offers that
seemed fair (47%) than offers that were actually fair (0%).
Distributional unjustness as an alternative explanation
As in Experiment 1, a One-Way ANOVA with (temptation: HT vs.
LT vs. G) predicting the ability to make a fair offer as dependent
variable revealed no significant effects for temptation (MHT = 5.17 vs.
MLT = 5.63 vs. MG = 5.14), F (2, 67) = .79, ns. This means that
participants did not perceive the HT situation to limit their ability to
make a fair offer compared to the LT or G situations. As in Experiment
1, controlling for this variable in all reported analyses did not change
any of the reported results. This further corroborates the evidence so
far suggesting that entrance was driven by outcome responsibility
considerations and not by proposers' perceived ability to make a fair
offer.
Discussion
Experiment 3 provided new support for the hypothesis that people
avoid tempting situations more than less tempting ones. It further
showed that people are not more attracted to situations that enable
them to be (secretly and anonymously) generous. Together, these
findings suggest that people avoid tempting situations because of a
desire not to act immorally and socially inconsiderate more than
because of a desire to approach situations in which they can
demonstrate generosity. Preventing doing harm seems more important than doing well to others (the do-no-harm principle; Baron,
1995).
In addition to this new support, Experiment 3 showed that while
tempting situations trigger anticipated moral emotions to a greater
extent than less tempting situations, these moral emotions played no
role in driving entrance decisions. Anticipating moral emotions thus
appears to be a by-product rather than a key driver of the decision to
create interdependence. In Experiment 3, we additionally explored
offer behavior and found that those who volunteered to enter the high
temptation game made more generous offers than those volunteering
in the low temptation or generous games (or those in the forced
entrance conditions). To clarify, when referring to generosity in this
context, we mean that proposers kept fewer chips to themselves in
one condition more than the other, thus decreasing own outcome to
benefit the other who was unaware of the differential value of the
chips. Keeping in mind the selection bias inherent to this secondary
analysis, this suggests that those who volunteered into the high
temptation game were not the small subset of shrewd and selfish
HT
15
LT
14
G
13
12
Table 1
Percentage of proposers making an offer as a function the setting's temptation level
(game) and exit option (Exp. 3).
Game
Entrance
11
HT
10
Voluntary offer
Forced offer
Fig. 4. Chips offered as a function of the setting's temptation level (game) and entry
decision (Exp. 3).
LT
G
Voluntary
Forced
Voluntary
Forced
Voluntary
Forced
Offers
14 or less (%)
15 (%)
16–19 (%)
20 or higher (%)
30.8
46.7
72.2
56.3
62.5
41.7
15.4
46.7
22.2
31.3
37.5
33.3
23.1
6.7
5.6
6.3
0
25.0
30.8
0
0
6.3
0
0
S. Shalvi et al. / Journal of Experimental Social Psychology 47 (2011) 1096–1106
individuals who exploit every possible opportunity to profit on the
expense of others.
Finally, the results of Experiment 3 further ruled out two
alternative interpretations to our results. As in Experiment 1, we
measured and ruled out the possibility that the HT situation might
lead participants to feel that it reduces their ability to make a fair offer
compared with the low temptation and generous settings. Put
differently, this explanation suggests that people avoid social
situations which make them feel that their ability to make a fair
offer is limited. Participants in both Experiments 1 and 3 did not
report experiencing such a feeling and controlling for the extent to
which participants felt that they were able (or not) to make a fair offer
never affected in any way the mediation analyses we reported.
Additionally, the idea that people avoid the HT game more than
the LT game because in the former it is more difficult to calculate what
is fair or unfair, does not seem to drive any of the reported effects. This
possibility is rendered unlikely by three sets of findings. First,
differences in calculative complexity (that might potentially evoke
feelings of ambivalence or uncertainty) would have produced
differences in the time needed to take the avoidance versus entrance
decision. We did, however, never find this. Second, in the generous
game it is as difficult to calculate a fair offer as in the HT game
(splitting chips in 2:3 and 2:1 ratios). A calculative complexity
account thus implies that both the G and the HT games should be
avoided equally and more than the LT game. Again, this is not what we
found — the HT game was avoided more than both the G and the LT
game. Third, in Experiment 2 participants entered rather than avoided
the HT game more than the LT game when the counterpart was
depicted as a competitive (but not cooperative) person. This pattern is
in line with our prediction (H4) that people's entrance decisions into
more or less tempting situation would vary as a function of lowering
the ethical threshold required in the situation (as is the case when
interacting with a competitive person). It is hard to imagine that
complexity considerations were driving these effects as there is no
reason to assume that when interacting with a competitive person
people would be suddenly attracted (rather than averse) to complex
settings.
Together, results suggest that people do not avoid the HT situation
because they feel that it limits their ability to make a fair offer or
because of complexity considerations of any sort, but rather because
they do not wish to take active responsibility for the other's outcomes.
General discussion
Humans are social animals — they interact with one another and
cooperate to mutual benefit. Interacting with other individuals
secures protection, fulfills basic affiliation needs, and provides
opportunity to advance both individually and collectively (Aronson,
2008; Baumeister & Leary, 1995; Williams, 2001). Yet social interactions are also conflict-prone — individuals compete over scarce
resources, they derogate each other's views and perspectives, and lash
out when self-worth is compromised by others' disrespectful behavior
(De Dreu, 2010; Miller, 2001). Accordingly, it stands to reason that
people gravitate towards those social interactions that have the
potential to fulfill basic needs for security, affiliation, and exploration.
Equally reasonable is to assume that people avoid those social
interactions that may create conflict and exploitation, and that may
undermine rather than enhance one's feelings of self-worth. However,
this latter possibility remained largely unexplored, as we know little
about reasons people have to move away from, or towards, social
situations.
Here we found that people avoid social situations that provide the
opportunity to privately and anonymously put one's partner in a
disadvantageous position while still appearing fair and moral. Second,
we found that these tendencies to avoid tempting situations are
driven by people's decreased willingness to take responsibility for the
1103
other's outcomes. While people clearly care about satisfying others'
expectations (Dana et al. 2006), our results suggest that people seek
to avoid situations in which they may deceive another person more
than situations that do not enable such deception, even when the
other's expectations are exactly the same (due to this person's lack of
knowledge about the value of the goods). The risk of deceiving the
other and thus compromising one's self-image of being a socially
considerate, morally appropriate person may cause people to decide
to avoid the tempting situation altogether. Indeed, when contemplating the decision to enter or avoid, participants anticipated more
guilt and shame when facing tempting situations. Third, and finally,
we found that people have a stronger aversion of tempting situations
than of situations that enable them to be generous — avoiding hurting
others appears more important than seeking to benefit others.
Theoretical implications
Social psychology is about people living in groups, and interacting
with others. A lot is known about how people behave in certain social
situations. Far less theorizing and research focuses on what type of social
situations people approach or avoid (see Small, Gelfand, Babcock &
Gettman, 2007). However, a number of studies addressed parts of this
broad issue. Interdependency theory (Thibaut & Kelley, 1959; Kelley &
Thibaut, 1978; Rusbult & Van Lange, 2003; also see Lewin, 1947)
proposes that people leave social situations when the social interaction
yields less utility than the “comparison level of alternatives.” That is,
when people expect to achieve greater utility in other (social) situations
they become motivated to leave their current relationship — they may
quit their job when more interesting employers become available, or
end their relationship when a more attractive partner can be secured
(see also Dawes et al., 1977). Similarly, negotiators become tougher or
end communications when the negotiation appears to result in
outcomes below the individual's “best alternative to negotiated
agreement” (BATNA, see e.g., Bazerman, Curhan, Moore, & Valley, 2000).
To these relatively disconnected areas of research, the present
series of experiments contributes two insights. First, this earlier work
considered people's motives to leave social situations and did not
consider what drives people towards or away from entering particular
social settings. Our experiments uncovered that people are relatively
averse of social situations that enable them to privately and
anonymously cheat their partner into a disadvantageous position.
Although less strongly, our experiments also suggest that people are
not particularly attracted to situations that allow them to be generous.
In short, our work shows that people navigate through social space
with a particular eye towards avoiding undesirable (tempting)
settings (see related work on self-control, Trope & Fishbach, 2000;
Fishbach & Trope, 2005).
Second, previous work focused on selfish motives as driving
people's decision to enter or leave social situations. It showed that
people leave social situations when they can improve their personal
outcomes elsewhere (Thibaut & Kelley, 1959) or avoid social
situations that entail high risk of being exploited by others (Dawes
et al., 1977). Our studies go one step further and show that people do
not only leave the latter type of aversive situations, but may even
avoid entering them altogether. In addition, avoiding or entering
social situations is determined by the extent to which people are
willing to take responsibility over other's fate and outcomes. Across
three studies we found that in such tempting situations, people are
less likely to engage in social interaction due to the outcome
responsibility it entails. Because of their concerns for the other's
fate, people gravitate towards those social interactions that do not
tempt them into acting immorally and unfairly, even under conditions
of complete anonymity.
Experiment 3 showed that people avoid tempting situations more
than that they are attracted to situations in which they may (secretly)
benefit the other. This finding is in line with the idea that people are
1104
S. Shalvi et al. / Journal of Experimental Social Psychology 47 (2011) 1096–1106
more sensitive to (privately or publicly) harming others than
benefiting them (do-no-harm principle; Baron, 1995). Recent work
also supports this notion in the context of social decision making,
including Ultimatum Bargaining (Leliveld et al., 2008) and coalition
formation (Van Beest, Van Dijk, De Dreu & Wilke, 2005). For example,
Leliveld et al. found that framing the offer in terms of taking money
from the responder leads proposers to make more generous offers
compared to the offers that are made when offers are framed as giving
to the responder. It thus seems that when deciding if and how to
interact with others, people are more sensitive to not harming others
than to benefiting them. However, future research, especially in the
context of maneuvering in and out of social situations, is needed to
substantiate this proposition.
A final albeit more descriptive contribution the current studies
make relates to the differences in offers between voluntary entrance
and forced entrance conditions in Experiment 3. Past research
using experimental games predominantly used forced entrance
conditions — participants volunteer for the experiment but are brought
into the game and asked to make an offer, to respond to their partner's
offer, to decide between a cooperative or non-cooperative decision
alternative, and so on. Such work indicates that when people are given
the opportunity to privately and anonymously exploit their partner and
deceive him or her into a disadvantageous position, they are quite likely
to do so (e.g., Boles, et al., 2000; Pillutla & Murnighan, 1995; Van Dijk
et al., 2004; Van Dijk & Vermunt, 2000). Replicating these findings,
participants in the forced condition of Experiment 3 indeed acted in a
strategic manner, offering similar amounts of chips across the different
games. Compared to such forced entrance conditions, and keeping in
mind the inherent self-selection issue in the voluntary conditions, we
found that individuals who volunteer into tempting situations were
more honest and fair. This finding may cast new light on the idea that
when push comes to shove, individuals act selfishly in their own
interests and ignore or even derogate those of others. Although our
findings do not dismiss that people can be egoistic and dishonest, they
warn against overestimating the prevalence of immoral, unfair and selfinterested behavior. For settings into which people volunteer, such as
marriages, labor exchanges, or economic trade negotiations, the rate of
deception and competition may be lower than previously concluded
from forced-entry type studies. We suggest that new work should
engage in a more systematic analysis of negotiation and trade tendencies
when participants volunteered into the situation.
Study limitations
In all three experiments reported here we assessed people's
likelihood to engage in social interaction with another person who
they may deceive (or not). We found repeated evidence that people
seek to avoid those settings in which it was easy to serve one's selfinterest by making an offer that seemed fair while in fact being unfair.
Based on moral psychological findings, we further tested the prediction
and gained support for the idea that this pattern of behavior is driven by
people's reduced desire to become responsible for the other's outcomes
in tempting settings. This was found in Experiment 1 and replicated
in Experiment 3. In both experiments however, the data about the
people's willingness to be responsible for the other's outcomes was
collected after the decision to enter or not (Experiment 1) and after the
decision to enter or not and the offer that was made (Experiment 3).
This raises the possibility that participants retrospectively interpreted
their decision to enter or not in terms of their desire to be responsible to
the other's outcomes. While it would have been ideal to counterbalance
the order of presented variables (the decision to enter or not and the
items assessing participants' willingness to be responsible for the
other's outcomes) we believe that the results of Experiment 2 provide
further support for our theory and not to this alternative interpretation.
Specifically, results of Experiment 2 showed that people in tempting
settings avoided a competitive partner less than in a non-tempting
setting. This supported our expectation as past work identified that
people are generally less positive towards competitive people (e.g.,
Steinel & De Dreu, 2004), which suggests that they would be less
inclined to avoid responsibility over these other's outcomes. Theory
supporting people's reduced desire for outcome responsibility in
morally challenging settings together with the set of results obtained
in all three experiments suggest that avoidance of outcome responsibility impacted the entrance decision and not the other way around.
A second issue to consider is the level of anonymity participants in
our experiments experienced as their decisions were recorded for
analysis. It could be argued that in settings in which one's behavior is
not monitored in any way, people would be more likely to enter a
socially tempting setting than our results indicate. Indeed, situations
in which anonymity considerations may dictate behavior could be
studied experimentally by (1) employing paradigms that insure
complete anonymity (i.e., decisions are collected only on an aggregate
level; see Shalvi et al., 2011), (2) in settings in which participants are
not aware that their behavior is being monitored, or (3) in settings
that evoke complete anonymity towards others (experimenter and
counterpart) but individual decisions are recorded (as in the current
studies) a fact which is communicated to participants. While there is
an advantage in analyzing data on an aggregate level at least in getting
as close as possible to creating full anonymity conditions in lab settings,
such settings prevent assessing the mediating process based on
individuals' decisions. Not informing participants that their behavior
is being monitored involved deceiving participants — which in the
current setting seemed unnecessary. As we were interested not only in
identifying people's tendency to avoid tempting settings but also in
clarifying the mechanism for this behavior (avoidance of outcome
responsibility) we opted for the third approach. This allowed studying
the psychological mechanism while avoiding experimental deception.
Concluding thoughts
Others before us noted that people avoid social situations with a
high risk of being exploited by others (e.g., Dawes, McTavish &
Shaklee, 1977; Maner, et al. 2007; Orbell and Dawes, 1993) and thus
prefer social situations with systems that sanction defective behavior
(Gurerk, Irlenbusch & Rockenbach, 2006). These tendencies are
generally taken as reflecting a desire to avoid being exploited by
others. A perhaps less obvious yet equally fitting explanation is that
people avoid these situations because they are at odds with what feels
morally right and socially appropriate. In our studies, participants
were always in the driver seat — it was them who could decide to
serve personal interests by exploiting others or, alternatively, to be
fair or even generous. In our studies, there was no risk of being
exploited and abused by others (it was a single-shot completely
anonymous interaction). Still, participants avoided the driver seat
when the temptation to deceive others increased. Past findings
together with the current work thus reveals that avoiding or taking
responsibility over another's fate operates as a situation selection
mechanism that facilitates avoiding social situations that may lead
one into temptation.
References
Aronson, E. (2008). The Social Animal (10th ed.). New York: Worth/FreemanBaron.
Baron, J. (1995). Blind justice: Fairness to groups and the do-no-harm principle. Journal
of Behavioral Decision Making, 8, 71–83.
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in
social psychological research: Conceptual, strategic, and statistical considerations.
Journal of Personality and Social Psychology, 51, 1173–1182.
Batson, C. D., Kobrynowicz, D., Dinnerstein, J. L., Kampf, H. C., & Wilson, A. D. (1997). In a
very different voice: Unmasking moral hypocrisy. Journal of Personality and Social
Psychology, 72, 1335–1348.
Batson, C. D., Thompson, E. R., Seuferling, G., Whitney, H., & Strongman, J. (1999). Moral
hypocrisy: Appearing moral to oneself without being so. Journal of Personality and
Social Psychology, 77, 525–537.
S. Shalvi et al. / Journal of Experimental Social Psychology 47 (2011) 1096–1106
Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for interpersonal
attachments as a fundamental human motivation. Psychological Bulletin, 117,
497–529.
Baumeister, R. F., Vohs, K. D., Dewall, N. C., & Liging, Z. (2007). How emotion shapes
behavior: feedback, anticipation, and reflection, rather than direct causation.
Personality and Social Psychology Review, 11, 167–203.
Bazerman, M. H., Curhan, J. R., Moore, D. A., & Valley, K. L. (2000). Negotiation. Annual
Review of Psychology, 51, 279–314.
Boles, T. L., Corson, R. T. A., & Murninghan, J. K. (2000). Deception and retribution in
repeated ultimatum bargaining. Organizational Behavior and Human Decision
Processes, 83, 235–259.
Camerer, C. F., & Fehr, E. (2006). When does “Economic Man” dominate social behavior?
Science, 311, 47–52.
Dana, J., Cain, D. M., & Dawes, R. M. (2006). What you don't know won't hurt me: Costly
(but quiet) exit in dictator games. Organizational Behavior and Human Decision
Processes, 100, 193–201.
Dawes, R., McTavish, J., & Shaklee, H. (1977). Behavior, communication, and
assumptions about other people's behavior in a commons dilemma situation.
Journal of Personality and Social Psychology, 35, 1–11.
De Bruin, E. N. M., & Van Lange, P. A. M. (1999). Impression formation and cooperative
behavior. European Journal of Social Psychology, 29, 305–328.
De Dreu, C. K. W. (2010). Social conflict: The emergence and consequences of struggle
and negotiation. In S. T. Fiske, D. T. Gilbert, & G. Lindzey (Eds.), Handbook of Social
Psychology (5th edition). 2010. New York: Wiley.
De Dreu, C. K. W., & Boles, T. L. (1998). Share and share alike or winner takes all? The
influence of social value orientations upon choice and recall of negotiation
heuristics. Organizational Behavior and Human Decision Processes, 76, 253–276.
De Dreu, C. K. W., & Van Kleef, G. A. (2004). The influence of power on the information
search, impression formation, and demands in negotiation. Journal of Experimental
Social Psychology, 40, 303–319.
De Dreu, C. K. W., & Van Lange, P. A. M. (1995). The impact of social value orientations
on negotiator cognition and behavior. Personality and Social Psychology Bulletin, 21,
1178–1188.
De Dreu, C. K. W., Weingart, L. R., & Kwon, S. (2000). Influence of social motives on
integrative negotiation: A meta-analytical review and test of two theories. Journal
of Personality and Social Psychology, 78, 889–905.
De Dreu, C. K. W., Yzerbyt, V. Y., & Leyens, J. -P. (1995). Dilution of stereotype-based
cooperation in mixed-motive interdependence. Journal of Experimental Social
Psychology, 31, 575–593.
De Hooge, I. E., Zeelenberg, M., & Breugelmans, S. M. (2007). Moral sentiments and
cooperation: Differential influences of shame and guilt. Cognition and Emotion, 21,
1025–1042.
Deutsch, M. (1985). The resolution of conflict: Constructive and destructive processes. New
Haven, CT: Yale University Press.
Fishbach, A., & Trope, Y. (2005). The substitutability of external control and self-control.
Journal of Experimental Social Psychology, 41, 256–270.
Gilbert, D. T., Pinel, E. C., Wilson, T. D., Blumberg, S. J., & Wheatley, T. P. (1998). Immune
neglect: A source of durability bias in affective forecasting. Journal of Personality and
Social Psychology, 75, 617–638.
Gilovich, T., & Medvec, V. H. (1995). The experience of regret: What, when, and why.
Psychological Review, 102, 379–395.
Gneezy, U. (2005). Deception: The role of consequences. The American Economic Review,
95, 384–394.
Greene, J., & Haidt, J. (2002). How (and where) does moral judgment work? Trends in
Cognitive Sciences, 6, 517–523.
Gurerk, O., Irlenbusch, B., & Rockenbach, B. (2006). The competitive advantage of
sanctioning institutions. Science, 312, 108–111.
Güth, W., Schmittberger, R., & Schwarze, B. (1982). An experimental analysis of ultimatum
bargaining. Journal of Economic Behavior and Organization, 3(4), 367–388.
Haidt, J. (2007). The new synthesis in moral psychology. Science, 316, 998–1002.
Handgraaf, M. J. J., Van Dijk, E., & De Cremer, D. (2003a). Social utility inultimatum
bargaining. Social Justice Research, 16, 263–283.
Handgraaf, M. J. J., Van Dijk, E., Vermunt, R. C., Wilke, H. A., & De Dreu, C. K. W. (2008).
Less power or powerless? Egocentric empathy gaps and the irony of having little
versus no power in social decision making. Journal of Personality and Social
Psychology, 95, 1136–1149.
Handgraaf, M. J. J., van Dijk, E., Wilke, H. A. M., & Vermunt, R. (2003b). The salience of a
recipient's alternatives: Inter- and intrapersonal comparison in ultimatum games.
Organizational Behavior and Human Decision Processes, 90, 165–177.
Kagel, J. H., Kim, C., & Moser, D. (1996). Fairness in ultimatum games with
asymmetric information and asymmetric payoffs. Games and Economic Behavior,
13(1), 100–110.
Kelley, H. H., & Stahelski, A. J. (1970). Social interaction basis of cooperators' and
competitors' beliefs about others. Journal of Personality and Social Psychology, 16,
66–91.
Kelley, H. H., & Thibaut, J. (1978). Interpersonal relations: A theory of interdependence.
New York: Wiley.
Komorita, S. S., & Parks, C. D. (1995). Interpersonal relations: Mixed-motive interaction.
Annual Review of Psychology, 46, 183–207.
Kuhlman, D. M., & Marshello, A. F. (1975). Individual differences in game motivation as
moderators of preprogrammed strategy effects in prisoner's dilemma. Journal of
Personality and Social Psychology, 32, 922–931.
Leliveld, M. C., Van Dijk, E., & Van Beest, I. (2008). Initial ownership in bargaining:
Introducing the giving, splitting, and taking ultimatum bargaining game.
Personality and Social Psychology Bulletin, 34, 1214–1225.
Lewin, K. (1947). Frontiers in group dynamics. Human Relations, 1, 5–41.
1105
Loewenstein, G. F., Thompson, L., & Bazerman, M. (1989). Social utility and decision
making in interpersonal contexts. Journal of Personality and Social Psychology, 57,
426–441.
Lundquist, T., Ellingson, T., & Johannesson, M. (2009). The aversion to lying. Journal of
Economic Behavior and Organization, 70, 81–92.
MacKinnon, D. P., & Dwyer, J. H. (1993). Estimating mediated effects in prevention
studies. Evaluation Review, 17, 144–158.
Maner, J. K., DeWall, C. N., Baumeister, R. F., & Schaller, M. (2007). Does social exclusion
motivate interpersonal reconnection? Resolving the “Porcupine Problem”. Journal
of Personality and Social Psychology, 92, 42–55.
Mazar, N., Amir, O., & Ariely, D. (2008). The dishonesty of honest people: A theory of
self-concept maintenance. Journal of Marketing Research, 45, 633–644.
McClure, S. M., Laibson, D. I., Loewenstein, G., & Cohen, J. D. (2004). Separate neural
systems value immediate and delayed monetary rewards. Science, 306, 503–507.
Mead, N. L., Baumeister, R. F., Gino, F., Schweitzer, M. E., & Ariely, D. (2009). Too tired to
tell the truth: Self-control resource depletion and dishonesty. Journal of
Experimental Social Psychology, 45, 594–597.
Messick, D. M., & Sentis, K. P. (1979). Fairness and preference. Journal of Experimental
Social Psychology, 15, 418–434.
Miller, G. (2001). The Mating Mind: How Sexual Choice Shaped the Evolution of Human
Nature. London: Vintage.
Oosterbeek, H., Sloof, R., & van de Kuilen, G. (2004). Cultural differences in ultimatum
game experiments: Evidence from a meta-analysis. Experimental Economics, 7,
171–188.
Orbell, J. M., & Dawes, R. M. (1993). Social welfare, cooperators' advantage, and the
option of not playing the game. American Sociological Review, 58, 787–800.
Pillutla, M. M., & Murnighan, J. K. (1995). Being fair or appearing fair: Strategic behavior
in ultimatum bargaining. Academy of Management Journal, 38, 1408–1426.
Pillutla, M. M., & Murnighan, J. K. (2003). Fairness in bargaining. Social Justice Research,
16, 241–262.
Pruitt, D. G., & Kimmel, M. J. (1977). 20 years of experimental gaming — Critique,
synthesis, and suggestions for future. Annual Review of Psychology, 28, 363–392.
Ritov, I., & Baron, J. (1995). Outcome knowledge, regret, and omission bias.
Organizational Behavior and Human Decision Processes, 64, 119–127.
Roese, N. J. (1997). Counterfactual thinking. Psychological Bulletin, 121, 133–148.
Rubin, J. Z., Pruitt, D. G., & Kim, S. H. (1994). Social conflict: Escalation, stalemate, and
settlement (2nd ed.). New York: McGraw-Hill.
Rusbult, C. E., & Van Lange, P. A. M. (2003). Interdependence, interaction, and
relationships. Annual Review of Psychology, 54, 351–375.
Schelling, T. C. (1960). The strategy of conflict. MA: Harvard University Press Cambridge.
Schweitzer, M. E., & Gibson, D. (2008). Fairness, feelings, and ethical decision making:
Consequences of violating community standards of fairness. Journal of Business
Ethics, 77, 287–301.
Shalvi, S., Handgraaf, M. J. J., & De Dreu, C. K. W. (2011). Ethical maneuvering: Why
people avoid both major and minor lies. British Journal of Management, 22, s16–s27.
Shalvi, S., Dana, J., Handgraaf, M. J. J., & De Dreu, C. K. W. (2011). Justified ethicality:
Observing desired counterfactuals modifies ethical perceptions and behavior.
Organizational Behavior and Human Decision Processes, 115, 181–190.
Small, D. A., Gelfand, M., Babcock, L., & Gettman, H. (2007). Who goes to the bargaining
table? Understanding gender variation in the initiation of negotiations. Journal of
Personality and Social Psychology, 93, 13.
Steinel, W., & De Dreu, C. K. W. (2004). Social motives and strategic misrepresentation
in social decision making. Journal of Personality and Social Psychology, 86,
419–434.
Tabibnia, G., Satpute, A. B., & Lieberman, M. D. (2008). The sunny side of fairness:
Preference for fairness activates reward circuitry (and disregarding unfairness
activates self-control circuitry). Psychological Science, 19, 339–347.
Tangney, J. P., Stuewig, J., & Mashek, D. J. (2007). Moral emotions and moral behavior.
Annual Review of Psychology, 58, 345–372.
Thibaut, J. W., & Kelley, H. H. (1959). The Social Psychology of Groups. New York: John
Wiley & Sons.
Trope, Y., & Fishbach, A. (2000). Counteractive self-control in overcoming temptation.
Journal of Personality and Social Psychology, 79, 493–506.
Van Beest, I., & Van Dijk, E. (2007). Self-interest and fairness in coalition formation: A
social utility approach to understanding partner selection and payoff allocations in
groups. European Review of Social Psychology, 18, 132–174.
Van Beest, I., Van Dijk, E., De Dreu, C. K. W., & Wilke, H. A. M. (2005). Do-no-harm in
coalition formation: Why losses inhibit exclusion and promote fairness cognitions.
Journal of Experimental Social Psychology, 41, 609–617.
Van Dijk, E., De Cremer, D., & Handgraaf, M. J. J. (2004). Social value orientations and the
strategic use of fairness in ultimatum bargaining. Journal of Experimental Social
Psychology, 40, 697–707.
Van Dijk, E., & Vermunt, R. (2000). Strategy and fairness in social decision making:
Sometimes it pays to be powerless. Journal of Experimental Social Psychology, 36,
1–25.
Van Kleef, G. A., De Dreu, C. K. W., & Manstead, A. S. R. (2006). Supplication and
appeasement in conflict and negotiation: The interpersonal effects of disappointment,
worry, guilt, and regret. Journal of Personality and Social Psychology, 87, 510–528.
Van Kleef, G. A., De Dreu, C. K. W., & Manstead, A. S. R. (2010). An interpersonal
approach to emotion in social decision making: The emotions as social information
model. Advances in Experimental Social Psychology, 42, 45–96.
Van Lange, P. A. M. (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.
von Neumann, J., & Morgenstern, O. (1947). The Theory of Games and Economic Behavior.
Princeton: Princeton University Press.
1106
S. Shalvi et al. / Journal of Experimental Social Psychology 47 (2011) 1096–1106
Vonk, R. (1998). Effects of cooperative and competitive outcome dependency on
attention and impression preferences. Journal of Experimental Social Psychology, 34,
265–288.
Weber, J. M., Kopelman, S., & Messick, D. M. (2004). A conceptual review of decision
making in social dilemmas. Personality and Social Psychology Review, 8, 281–307.
Williams, K. D. (2001). Ostracism: The power of silence. New York: Guilford Publications.
Zeelenberg, M., van Dijk, W. W., van der Pligt, J., Manstead, A. S. R., van Empelen, P., &
Reinderman, D. (1998). Emotional reactions to the outcomes of decisions: The role
of counterfactual thought in the experience of regret and disappointment.
Organizational Behavior and Human Decision Processes, 75, 117–141.