Download Understanding Sanctions in Social Dilemmas: A Biopsychological

Understanding Sanctions in Social Dilemmas:
A Biopsychological Perspective
on Costly Punishment in Public Goods Situations
Dissertation
zur Erlangung des Doktorgrades Dr.phil.
der Fakultät für Ingenieurwissenschaften und Informatik
der Universität Ulm
vorgelegt von
Stefan Pfattheicher
aus Malsch (Kreis Karlsruhe)
2014
Amtierende Dekanin: Prof. Dr. Tina Seufert
Gutachter: Prof. Dr. Johannes Keller
Gutachter: Prof. Dr. Thomas Kessler
Gutachter: Prof. Dr. Arnd Florack
Tag der mündlichen Qualifikation: 06. Juni 2014
Contents
Preface ....................................................................................................................................... v
1 Introduction ........................................................................................................................... 1
1.1 The importance of investigating social dilemma situations ............................................ 3
1.2 Solving social dilemmas .................................................................................................. 5
1.3 The value of a biopsychological perspective on social dilemmas ................................... 8
1.4 Paradigms from game theory......................................................................................... 11
2 Why and under which conditions individuals costly punish ........................................... 16
2.1 Different perspectives on how to investigate costly punishment .................................. 16
2.2 Costly punishment from a biopsychological perspective .............................................. 17
2.2.1 Fairness and justice.............................................................................................. 17
2.2.2 Impulsivity........................................................................................................... 21
2.2.3 Anticipated reward .............................................................................................. 25
2.2.4 Reputation............................................................................................................ 28
2.3 Summary........................................................................................................................ 29
3 Empirical papers ................................................................................................................. 31
3.1 Towards a biopsychological understanding of costly punishment: The role of basal
cortisol (Pfattheicher & Keller, in press, Appendix A) ................................................ 31
3.1.1 Discussion............................................................................................................ 34
3.2 The eye cue manipulation as subtle induction of a sense of being watched:
Empirical validation and theoretical integration (Pfattheicher & Keller, 2014,
Appendix B) ................................................................................................................. 37
3.2.1 Discussion............................................................................................................ 40
3.2.2 Methodological remarks ...................................................................................... 43
3.3 Punishing for the sake of reputation? Effects of reputational concerns on costly
punishment (Pfattheicher, Landhäußer, & Keller, 2014, Appendix C) ........................ 45
3.3.1 Discussion............................................................................................................ 50
3.3.2 Further remarks ................................................................................................... 52
3.3.3 Discussing costly signaling theory ...................................................................... 53
3.3.4 Explaining actor-observer discrepancies ............................................................. 56
3.4 Vigilant self-regulation and costly punishment in public goods situations
(Pfattheicher & Keller, 2013, Appendix D) ................................................................. 61
3.4.1 Theories of motivation and regulatory focus theory ........................................... 62
3.4.2 The empirical paper ............................................................................................. 67
3.4.3 Integrating reputational concerns and oughts ...................................................... 69
3.4.4 Additional theoretical remarks ............................................................................ 70
3.4.5 Integrating findings on cortisol and prevention focus ......................................... 73
3.5 Individual differences in antisocial punishment in public goods situations: The
interplay of cortisol with testosterone and dominance (Pfattheicher, Landhäußer, &
Keller, in press, Appendix E) ....................................................................................... 74
3.5.1 Relations to other work ....................................................................................... 77
3.5.2 Proposals for future research ............................................................................... 78
4 General discussion ............................................................................................................... 81
4.1 Summarizing the findings .............................................................................................. 82
4.2 Discussion of the applied paradigm .............................................................................. 84
4.3 Discussion of the biopsychological perspective ............................................................ 85
4.4 The subjective construction of social dilemmas ............................................................ 87
4.5 Implications for behavior in “real life”.......................................................................... 89
4.6 The role of altruism in costly punishment ..................................................................... 92
References ............................................................................................................................... 96
Zusammenfassung ................................................................................................................ 136
Appendix ............................................................................................................................... 137
PREFACE v
Preface
In 2006 I participated in a seminar on “Theories of Political Decisions” held at my
former university (at Mannheim). At that time I was interested in learning about how
people in our society make decisions when they vote, particularly because I was
convinced that many of these decisions were not founded on reasonable considerations.
In short, the seminar did not answer my questions. Instead, I found myself in a seminar
on game theory in the political decision context. I was immediately fascinated by the
possibility of using simple game theory paradigms to model human decisions and social
situations. Almost two years later the story continued when I joined a seminar on “Social
Interaction”. What bothered me in the political seminar on game theory was that the
ultimate focus was on human behavior from a rational choice perspective with basal
human needs, motivations, and social situations being neglected. Happily, these issues
were addressed in the seminar on “Social Interaction”. In fact, this seminar was a tasty
antithesis to the rational choice perspective primarily taught at my university (cf.
Kroneberg & Kalter, 2012).
When it came to deciding where to write my thesis for my diploma I recalled the
inspiring seminar on “Social Interaction” and contacted my former lecturer. In a nutshell,
this begun my work with Johannes Keller on social dilemma situations and the notion of
costly punishment. While writing my diploma thesis, I was very happy when Johannes
offered me the possibility to become a PhD student at his department and to continue
working on this topic (I remember that I possessed a promotion-focus that day).
Becoming a doctoral candidate was appealing given the possibility it offered to gain a
deeper understanding of human behavior. For instance, I found it interesting how
humans’ behavior can be shaped by simple social cues (cf. Pfattheicher & Keller, 2014,
Appendix B) as a function of basal motivational orientations (Keller & Pfattheicher,
2011, Appendix F). However, the understanding of why humans behave as they do was
not easily served. Specifically, the understanding, crystallized in this dissertation, of why
individuals engage in costly punishment in public goods situations is the result of a long
PREFACE vi
(and still ongoing) process. My thinking began with the conceptualization of costly
punishment as a socially desirable behavior, mainly deliberatively executed to uphold
social norms and moral principles (Batson, Ahmad, Powell, & Stocks, 2008; Batson,
Ahmad, & Tsang, 2002; Kurzban, DeScioloi, O’Brien, 2007). I ended up at a very
different point in what constitutes this thesis.
I would not be where I am now without the help and encouragement I received from my
colleagues in academia and my friends. I am deeply grateful to Johannes for being such
an outstanding supervisor. You not only established a pleasant working atmosphere but
always had time for me and were willing to listen and to support me. Thanks to you, I
believe I now recognize the importance of a theory and what it means to conduct basic
research. Although you may support another football team, our discussions about
football (and other important issues) also contributed to my well-being during recent
years. Thank you for the, literally, open door you offered.
I am also deeply grateful to Toshi who has been an, almost, indefatigable and
indestructible companion in our office. Thank you for the fun times we shared, although
sometimes delicious food and a few drinks were necessary to make this happen. Thank
you, Claudia and Svenja, for being there in our self-help group (and beyond). Thanks to
all the research assistants who worked with me during the past couple of years,
particularly Cia, Friedrich, and Silvi. Thank you, Simon, for our inspiring talks about
existence. Thank you, Philipp, for our deep friendship and for introducing me to honeymustard salad dressing. Thank you, Cathrin, for your kindness and baking skills. Thank
you, Sebastian, for sharing my ridiculous sense of humor. Thank you, Flo and Thomas,
for sharing your devotedness to pasta and lively discussions. Thank you, Mum and Dad,
for offering me the chance to do what I have. Thank you, Bro, for being there, always.
Finally, and most importantly: Thank you, Katharina, for your love.
INTRODUCTION 1
1 Introduction
Imagine a single, lonely angler who knows from experience that she has a good chance
of catching a considerable amount of fish at a small lake. Having caught two fish, the
angler faces the possibility of catching a third, a fourth, and a fifth one. A rational angler
whose benefits exceed the costs of catching more than two fish would go for a third fish
(Esser, 2000). Indeed, individuals often strive to maximize their own benefit which fits
the basic idea of an orthodox rational actor referred to as homo economicus (Becker,
1976).1 Given perfect information about the costs and benefits of several behavioral
options, rational actors choose the option that yields the highest benefit.
The situation involving the angler is a very specific situation because it is not a social
situation in which the actions of several individuals relate to and influence each other.
The behavior of the angler is moreover not a social action. An “action is ‘social’ if the
acting individual takes account of the behavior of others and is thereby oriented in its
course” (Weber, 1922/1972, p. 1). In the present ideal-typical example, the rational
angler simply follows her own preferences (i.e., maximizing her benefit through catching
fish) and she does not take (and does not need to take) the expectations or behaviors of
other individuals, which may influence her preferences, into account. In social situations,
however, individuals affect other individuals’ affect, cognitions, and behaviors, and an
actor’s actions are, in turn, also affected by the affect, cognitions, and behaviors of other
individuals.
There are social situations in which the benefits to the individuals involved are
remarkably construed. To follow the example above, let us now consider not only a
single angler but three anglers sitting at the lake. Now the benefit of the angler is not
only determined by her own actions but also by the actions of the other anglers. For
instance, the other anglers could have more effective lures than the first angler which
may influence the first angler’s success. That is to say, the success of an angler also
depends on the actions of other anglers, which reflects the notion of mutual
1
When I refer to rational actors in the present work I mean actors that maximize material benefits; I do not
include other possible benefits, for instance psychological benefits such as satisfaction.
INTRODUCTION 2
interdependence. A second aspect of this social situation is worth noting. As mentioned
above, a rational angler would benefit from catching a lot of fish because catching more
fish results in more food and the possibility of gaining money through selling the fish at
a local market. However, catching a lot of fish affects the fish stocks in that there are less
fish for all of the anglers who angle at the lake. To emphasize this point, every rational
angler benefits from catching a lot of fish. However, if every angler behaves in this way,
the fish stocks are damaged so that the collective (i.e., all anglers) catch less fish. This
reflects the notion of conflicting interests between the individual and the collective. The
kind of social situation just described is termed a social dilemma situation, and the
dilemma is reflected in the divergence of individual and collective interests (Dawes,
1980).
If several individuals are involved in a social dilemma situation, two general types of
situation can be distinguished by means of three criteria.2 The first type involves a
common good and fits the example above. There are typically several anglers catching
fish at a lake which is home to a limited amount of fish. The anglers can take fish out of
the common good (i.e., catch a fish), they typically cannot exclude individuals who
would also like to catch fish, and the benefits are subtractable because for every fish
caught there is one less fish available for other anglers. So, taking, non-excludability,
and subtractable benefits are the criteria for a common good.
The second type, a social dilemma situation involving a public good, is characterized by
the aspect of giving. Let us assume that anglers have to keep the lake and its
surroundings clean, and that the anglers have to put (i.e., to give) a lot of effort into
cleaning. Individuals cannot be excluded from using the public good (i.e., the lake and its
surroundings). Using the public good is, moreover, non-rival because every angler can
angle and use the lake no matter how the other anglers behave. So, giving, non-
2
The focus of the present dissertation is social dilemma situations involving more than two individuals.
Therefore, social dilemma situations involving two individuals such as the “Prisoner’s Dilemma”
(Diekmann, 2009) are neglected.
INTRODUCTION 3
excludability, and non-rivalry are the criteria for a public good. Public goods are the
focus of the present dissertation so only these are considered in detail below.
In public goods situations, the benefit-maximizing strategy (i.e., not helping to keep the
lake clean) of several rational actors results in a deficit Nash equilibrium (Nash, 1950). A
Nash equilibrium is the state that results from the converging strategies of several, in this
case, rational actors in which no actor benefits from changing their behavior (i.e., the
equilibrium). In the example above, if every single angler pursues the benefitmaximizing strategy the result would be a ruined lake. An angler would gain almost no
benefit from changing their behavior, that is, cleaning a ruined lake completely on their
own. This situation is deficient in the sense that there actually is a collective optimum,
that is, mutual cooperation (i.e., all anglers participate in cleaning the lake and its
surroundings).
These considerations emphasize why Hardin (1968) speaks of a “tragedy” when he
describes social dilemma situations. Behaving selfishly is always the most beneficial
strategy from a rational choice perspective. This results in a Nash equilibrium in which
no one has any incentive to change their behavior. Thus, social dilemma situations are
intractable, strong situations in which individuals benefit from acting against the
collective (Esser, 2000).
1.1 The importance of investigating social dilemma situations
Why is it important to be concerned about social dilemma situations? In the example
above only very few individuals (anglers) take part in the dilemma situation and are
potentially affected by uncooperative behavior. If social dilemma situations occurred
rarely and affected only a very few individuals, investigating how to establish
cooperation would surely be essential for these individuals but not much beyond that.
However, this is undeniably not the case (van Lange, De Cremer, van Dijk, & van Vugt,
2007; van Lange, Joireman, Parks, & van Dijk, 2013). There is no individual on earth
that is not affected by social dilemma situations. This is true because far-reaching social
dilemma situations that affect all individuals do exist, for instance earth’s climate
(Hansen et al., 2013; van Vugt, Griskevicius, & Schultz, 2014). Specifically, a rational
INTRODUCTION 4
individual is best off if s/he does not devote resources (e.g., time, money) to protecting
the environment. However, the collective suffers from environment-damaging behavior
(i.e., uncooperative behavior) because the environment-damaging behavior of several
individual leads to climate change (e.g., Irwin & Berigan, 2013). Another example: the
fact that most individuals on earth are part of a nation state also creates numerous
dilemma situations (Ostrom, 1990). For instance, individuals are better off when evading
taxes but then the collective (i.e., the state) will suffer due to tax evasion because it will
be more difficult to run the healthcare system, the education system and the legislature
effectively. Smaller groups also face social dilemma situations, such as a student study
group in which each individual has incentives not to prepare for a meeting but,
nonetheless, benefits from the other (prepared) students. In sum, social dilemma
situations are present in a huge variety of social situations (van Lange et al., 2007; van
Lange et al., 2013).
However, not only quantity counts when answering the question of why it is important to
be concerned about social dilemma situations. The specific quality of social dilemma
situations also makes them worth investigating (Camerer, 2003). Individuals’ behavior is
more likely to be successfully predicted when benefits can be obtained on the individual
micro-level and the collective macro-level (i.e., no social dilemma). Halevy, Chouc, and
Murninghan (2012), for instance, showed that almost all of the individuals in their
studies behaved cooperatively when the highest personal benefit for them and the others
involved in the situation could be obtained through cooperation. However, because in
social dilemma situations individuals’ interests are in contrast to the interests of the
collective (cooperation is not the Nash equilibrium), it is often difficult to establish
cooperation (see, e.g., Fehr & Gächter, 2002). By definition, cooperation between
entities is a beneficial and valuable state for the collective. Given that selfish interests,
however, are in contrast to the collective’s interest the question of how to establish
cooperation between several individuals is of substantial value and worth investigating.
This proposal is strengthened by manifold empirical work. In her seminal book, Ostrom
(1990) documents the breakdown of cooperation in several real life situations, for
INTRODUCTION 5
instance when observing systems for water irrigation where overusing water (i.e.,
damaging the collective) during a drought may save one’s harvest. Empirical work from
the laboratory points in the same direction. Kelley and Grzelak (1972) were the first to
empirically investigate cooperation between several individuals in public goods
situations (referred to as an N-person prisoners’ dilemma). They observed a decreasing
level of cooperation across multiple repeated interactions (the same pattern was later
observed by Fehr and Gächter, 2002).
To conclude, social dilemma situations are likely to emerge when several individuals
interact to produce benefits and when mutual interdependence is provided in a way in
which individuals’ personal interest is in contrast to the collective’s interest (Esser,
2000). These situations exist in a variety of contexts, that is, social dilemma situations
frequently affect and guide humans’ behavior. Additionally, given the inherent
suboptimal beneficiary structure, social dilemma situations have a special quality that
essentially affects and guides humans’ behavior which is well-documented empirically
(e.g., Balliet, Mulder, & van Lange, 2011; Fehr & Gächter, 2002; Ostrom, 1990; van
Lange, Balliet, Parks, & van Vugt, 2014). These considerations constitute the rationale
for why factors that foster cooperation in social dilemma situations should be
investigated. These factors are outlined in the following section.
1.2 Solving social dilemmas
Kollock (1998) has proposed three general perspectives regarding how social dilemmas
can be solved, that is, how cooperation can be established: a structural perspective, a
strategic perspective, and a motivational (psychological) perspective. The structural
perspective emphasizes changing the structure of a social dilemma situation in order to
make cooperation more beneficial. Prominent instances of this are iterations of
interactions (rather than one-shot interactions) especially when individuals can be
identified (Fox & Guyer, 1978; Jerdee & Rosen, 1974) and when they can communicate
(Caporael, Dawes, Orbell, & van de Kragt, 1989). When these structural conditions are
given, higher levels of cooperation could be observed (cf. Shinada & Yamagishi, 2007).
One can also change the payoff structure to a situation where cooperation is rewarded by
INTRODUCTION 6
a higher personal return (Isaac & Walker, 1988; Kelley & Grzelak, 1972), and where
one’s contribution to a public good has a noticeable effect (e.g., in small groups) and is
crucial to the provision of the public good (Kerr & Kaufman-Gilliland, 1994; note,
however, that the literature on group size produced mixed results; see Shinada &
Yamagishi, 2007).
The strategic perspective emphasizes the notion that actors try to strategically adapt their
own behavior and shape the behavior of others in order to gain personal benefits. The
most prominent and one of the most successful strategies of this is the tit-for-tat strategy
(Axelrod & Hamilton, 1981). The tit-for-tat strategy states that an actor cooperates at
first and then responds to an action by a second actor by replicating the second actor’s
action (also known as direct reciprocity; Trivers, 1971). Thus, if a second actor
cooperates, then the first actor using the tit-for-tat strategy will cooperate as well.
However, if a second actor does not behave cooperatively, the uncooperative behavior is
responded to with equally uncooperative behavior. Another useful strategy is choosing
partners that have behaved cooperatively in the past because these types of individuals
are more likely to behave cooperatively in the future (Nowak & Sigmund, 1998;
Wedekind & Braithwaite, 2002). Halevy et al. (2012) have shown that individuals do
behave in accordance with a strategic approach. For instance, it was found that when
individuals faced a competitive situation in which cooperation provides a higher benefit
than defection if the other defects, individuals are more likely to cooperate the higher the
likelihood was that the other individual would defect. Accordingly, one can argue that
individuals strategically adapt their behavior in order to gain personal benefits.
The core of the structural and strategic perspectives is the fundamental goal to maximize
benefits (Kollock, 1998). However, there is wide agreement that these approaches do not
capture the entire spectrum of possibilities of how to establish cooperation (e.g.,
Camerer, 2003; Kollock, 1998; Shinada & Yamagishi, 2007; van Lange et al., 2007; van
Lange et al., 2013). The third approach proposed by Kollock (1998), the motivational
perspective, captures this point. This perspective emphasizes that individuals not only
take their own benefit into account but also that of the other individuals involved in the
INTRODUCTION 7
situation. Batson and colleagues (Batson et al., 2008; Batson et al., 2002) specified the
motivational perspective and put forward three general non-egoistic motivations to
explain why people act for the public good. These are: collectivism (to increase the
benefit of a group), principlism (to uphold moral principles), and altruism (to increase
the benefits of others). Strengthening these basal motivational orientations should
increase cooperation (Batson et al., 2008; Batson et al., 2002).
In sum, research completed over decades has identified several tools for avoiding freeriding and establishing cooperation (for a compelling review see van Lange et al., 2013).
In her seminal book Governing the Commons (1990), Nobel Prize winner Elinor Ostrom
has systematically offered several principles for how to promote cooperation and prevent
free-riding. First, clearly defined boundary conditions have to be set up in terms of who
is allowed to benefit from the common or public good. Taking the example given at the
beginning of the present work, the anglers could, for instance, unionize and restrict
access to the lake. Second, rules that regulate the use of the common or public goods
should be adapted to local conditions. For example, the anglers could restrict the amount
of fish that is allowed to be caught. Third, Ostrom argued that it is important that
individuals involved in the dilemma situation are able to participate in modifying
operational rules. Fourth, external authorities (such as a federal government or religious
institution) must recognize the rules adopted by the local individuals directly involved in
the dilemma. Fifth, boundary conditions and rules for how the goods are to be used need
to be monitored. Sixth, if a conflict emerges a low-cost local institution must be in place
to resolve the conflict. Seventh, Ostrom proposed that appropriate sanctions for breaking
the rules are necessary in order to prevent uncooperative behavior and to foster
cooperative behavior.
Proposing a sanction system is indeed a worthwhile approach given the remarkable
amount of empirical evidence (cf. the meta-analysis of Balliet et al., 2011) that has
convincingly shown that (a) negatively sanctioning (i.e., punishing) uncooperative others
leads to an increase in cooperative behavior of the punished individuals in future
interactions, and (b) there is a higher cooperation level in situations in which an option to
punish is available as compared with situations in which no such option is available. In
this regard, Gürerk, Irlenbusch, and Rockenbach (2006) showed that individuals also
INTRODUCTION 8
recognize the importance of a sanction system and adapt their behavior accordingly.
They offered participants the choice between a social dilemma situation in which
negative sanctions (i.e., punishing other individuals involved in the social dilemma) were
possible and the same social dilemma situation but without the possibility to negatively
sanction others. Individuals could change their choice after each interaction. At the
beginning, about 60% of the participants chose the social dilemma situation in which
negative sanctions were possible. After thirty interactions, more than 90% had chosen
the situation in which negative sanctions were possible.
It is the central feature of the present work to examine the sanction systems that Ostrom
(1990) and others have proposed (e.g., Olson, 1968; Yamagishi, 1986). This is a
worthwhile approach given that a sanction system is of substantial value when
cooperation needs to be established (Balliet et al., 2011; Fehr & Gächter, 2002; Gürerk et
al., 2006). Specifically, the main theme in the present dissertation is the question of why
and under which conditions individuals engage in the negative sanctioning of individuals
involved in social dilemma situations. That is, it is the aim of the present work to
understand negative sanctions in the context of social dilemma situations.
1.3 The value of a biopsychological perspective on social dilemmas
In the present work, I will propose several explanations for why and under which
conditions individuals engage in negative sanctions and punish others. This approach is
in line with recent theorizing emphasizing that several routes can lead to specific
behavior (cf. Bullock, Green, & Ha, 2010; Fiedler, Schott, & Meiser, 2011; Hayes,
2013). When explaining punishment in public goods situations I take the position that
individuals’ decisions are driven by factors of the person and by factors of the situation
as well as the interaction between these two types of factors (Ross & Nisbett, 1991).
This dissertation focuses on specific factors of the individual, that is, endocrinological
and psychological
factors, which constitutes
a biopsychological
perspective.
Specifically, I propose that one has to take endocrinological and psychological factors as
well as the present situation into account when aiming to comprehensively understand
social behavior. A good example that emphasizes the importance of this perspective is
INTRODUCTION 9
the study by Josephs, Sellers, Newman, and Mehta (2006). These scholars have
documented that in specific social situations the testosterone levels of individuals is
relevant in terms of how they experience emotions. Specifically, when individuals with a
relatively low level of testosterone were placed in a high status position they reported
greater emotional arousal and worse cognitive functioning as compared with when they
were placed in a low status position. The reverse was the case for individuals with a
relatively high level of testosterone. That is to say, it was not the testosterone levels itself
and also not the social situation itself (i.e., holding a low or high status position) that
predicted how the individuals reacted. The interaction of testosterone and the social
situation predicted behavior.
Also in line with a biopsychological perspective, Carré, Putnam, and McCormick (2009)
found that the testosterone levels of men who have won (vs. lost) a previous competition
increased, and this increase was positively related to aggression after a provocation.
However, this relation only holds true for those with a relatively high dispositional level
of dominance. A study by Klinesmith, Kasser, and McAndrew (2006) also impressively
shows the importance of a biopsychological perspective. These authors reported that
interacting with a gun (vs. a toy) increased proactive aggression (i.e., aggression without
a prior provocation) which could be explained by an increase in testosterone when
participants interacted with a gun.
It should have become apparent that when aiming to comprehensively understand social
behavior one has to take endocrinological factors and their interaction with situational
and other personal variables into account. Thus, a biopsychological perspective offers
the possibility to reach a better understanding of social behavior, which is the principal
objective of the present dissertation – to understand negative sanctions in public goods
situations.
Investigating social behavior from a biopsychological perspective has several additional
advantages. First, specific endocrinological factors are related to specific motives and
psychological processes. For instance, testosterone regulates dominant and destructive
behavior in status-relevant situations (Josephs et al., 2006; Mehta, Jones, & Josephs,
INTRODUCTION 10
2008; van den Bos, Golka, Effelsberg, & McClure, 2013). Basal levels of the steroid
hormone cortisol are negatively related to impulsivity (Schmidt, Mussel, & Hewig, 2013,
and Raine et al., 2006) and positively related to behavioral withdrawal (Blair, Peters, &
Granger, 2004; Brown et al., 1996; Montoya, Terburg, Bos, & van Honk, 2012; Nava,
Caldiroli, Premi, & Lucchini, 2006; Tops & Boksem, 2011; van Honk, Harmon-Jones,
Morgan, & Schutter, 2010; Windle, 1994). Thus, by relating behavior in social dilemma
situations to endocrinological factors (e.g., cortisol and testosterone) one can shed light
on underlying motives and processes. This is even more remarkable as individuals
themselves often have difficulty in and a restricted ability to correctly report the motives
and processes underlying their behavior (Nisbett & Wilson, 1977). Second, neglecting
endocrinological factors can lead to either false conclusions or no conclusions. If
Klinesmith et al. (2006) had not assessed testosterone levels they could not have
explained why interacting with a gun increased proactive aggression (i.e., no
conclusion). Furthermore, if Carré et al. (2009) had not assessed changes in testosterone
in their test subjects they could have concluded that the dispositional level of dominance
is not related to reactive aggression (i.e., a false conclusion). Thus, a biopsychological
perspective not only has the potential that conclusions can be drawn but also to prevent
false conclusions or no conclusion. Third, endocrinological factors are unbiased in terms
of socially desirable responding. This is important given that social desirability may play
a role in the domain of prosocial behavior (such as cooperation) and antisocial behavior
(such as the punishment of cooperative individuals, cf. Chapter 3.5). Thus, by assessing
endocrinological (e.g., testosterone) and psychological factors (e.g., a concern for
dominance) in social dilemma situations one can accommodate concerns regarding
socially desirable responding.
In sum, a biopsychological perspective on behavior in social dilemma situations seems to
be a valuable approach: a better understanding can be reached, false or no conclusions
can be avoided, and social desirability concerns can be addressed. Generally, the
framework of the present dissertation follows the perception of other models, such as the
General Aggression Model (Anderson & Bushman, 2002; Bushman & Anderson, 2002;
DeWall, Anderson, & Bushman, 2011), emphasizing not only traits, sex, values, goals,
INTRODUCTION 11
and behavioral scripts of a person that influence aggression but also individuals’
endocrinology.
The following section (Chapter 2) outlines why and under which conditions individuals
punish others, taking the biopsychological perspective just proposed. Before elaborating
on this central goal of the present dissertation, I provide an overview of how negative
sanctions can be investigated in the laboratory.
1.4 Paradigms from game theory
It is possible to model social dilemma situations and negative sanctions in a laboratory
setting, allowing the investigation of cooperation and negative sanctions in a reduced
situation in which external influences can be controlled and causal analytic approaches
applied. A public goods game including the option to negatively sanction other
individuals involved in the dilemma (i.e., to punish others) is often used to investigate
the emergence of cooperation. The public goods game is also extensively applied in the
empirical studies presented below, so it is described in the following paragraph.3
In a typical public goods game, four players constitute one group (cf. Fehr & Gächter,
2002). Each player is endowed with 20 money units (MUs; the MUs are converted to
real money and paid out at the end of a study) and are free to choose how many of these
to keep and how many to contribute to a common group project (i.e., the public good).
Each MU contributed is multiplied by 1.6. Next, each player receives one fourth of the
public good, independent of their contribution. Accordingly, it is always in the material
self-interest of every individual to keep all of their MUs irrespective of how much the
other three subjects contribute: If every group member invested 20 MUs, each subject
would earn 0.4 × 80 = 32 MUs. If one group member engages in free-riding (e.g., s/he
contributes 0 MUs) and the other three group members still invest their 20 MUs, the
free-rider earns 44 MUs (20 MUs already owned plus one fourth of the public good, that
is, 24 MUs) and each of the other three group members earns 24 MUs.
3
To establish transparency, I want to mention that the following paragraph includes sentences taken from
Pfattheicher, Landhäußer, and Keller (in press, Appendix E).
INTRODUCTION 12
Given the equal division of the public good irrespective of an individual’s contribution,
individuals benefit from selfish behavior regardless of the actions made by other
individuals involved in the situation. However, and this again is the tragedy Hardin
(1968) referred to, all individuals gain lower benefits if all behave selfishly than if all
individuals behave cooperatively (Dawes, 1980). That is, if all of the individuals behave
selfishly and contribute 0 MUs they will all end with their original endowment of 20
MUs. In contrast, if all of the individuals behave cooperatively and contribute 20 MUs,
they will all end up with 32 MUs.
Following this stage, each player is given accurate information on the contributions of
the other three players and is then given the option to punish them by investing their own
MUs (between 0 and 10 for each player) which reduces the selected other player’s payoff
by the factor of three (e.g., the investment of 2 MUs decreases the payoff of another by 6
MUs). The investment of own MUs reflects the negative sanctions Ostrom (1990)
referred to and is also known as costly punishment (e.g., Henrich et al., 2006).4
As far back as the 1970s, Caldwell (1976) investigated the effectiveness of a sanction
system in a public goods game. He let participants play repeated public goods games in
groups of five members. The study included three conditions. In the first condition, the
group members were not allowed to communicate. Communication was possible in the
second condition. The third condition was an extension of the second; Caldwell added a
(costless) option to reduce another individual’s payoff. If at least two group members
voted to punish another individual, the income of the other individual was reduced. As
with Fehr and Gächter (2002), Caldwell (1976) was able to document a substantially
higher level of cooperation when a sanction system was available (vs. not available).
In the field of negative sanctions, another game theory paradigm is often used to assess
costly punishment: the ultimatum game (Güth, Schmittberger, & Schwarze, 1982).5 The
4
Fehr and Gächter (2002) termed costly punishment of uncooperative interaction partners altruistic
punishment. This term is discussed in the general discussion of this dissertation.
5
Costly punishment can also be assessed when an individual simply observes uncooperative behavior. For
instance, Kurzban et al. (2007) included an option for observers to costly punish uncooperative actors of a
INTRODUCTION 13
ultimatum game consists of two players, the first player can offer a split of a specific
amount of money and the second player decides whether to accept the split or not. If the
offer is accepted, both players earn the money according to the offer. If the offer is
rejected, both players earn nothing. The rejection is often considered to reflect costly
punishment (e.g., Henrich et al., 2006; Marlowe et al., 2008) because the rejection
reflects punishment “at a cost to […] sanction unfair behavior” (Fehr & Fischbacher,
2003, p. 785).
It is not proposed here that costly punishment in a public goods game and costly
punishment in an ultimatum game reflect exactly the same behavior based on the same
motives. For instance, uncooperative behavior in a public goods game affects the benefit
of several individuals whereas in an ultimatum game only one individual is affected.
Accordingly, punishment in order to increase group cohesion (Horne, 2001) cannot be a
possible motive of costly punishment in the ultimatum game. Also, (interdependent)
collectivistic orientations are likely to emerge in groups (Markus & Kitayama, 1991)
whereas the group context does not exist in ultimatum games (although interdependence
is not precluded in ultimatum games). Additionally, if one is punished in an ultimatum
game, the punishment definitely comes from the second player whereas in a typical
public goods game one does not know from which group member the punishment comes.
So, in a public goods game, costly punishment may be executed in a less (subjectively
perceived) identifiable way (although in both games individuals typically act under
conditions of anonymity). Moreover, regarding costly punishment in a public goods
game, one has to invest resources taken from one’s own endowment to punish whereas
in the ultimatum game one rejects a benefit that was not owned before. Research on the
endowment effect (Kahneman, Knetsch, & Thaler, 1991; Knetsch, 1989) shows that
individuals assign a higher value to things they possess than things they do not possess.
Accordingly, assuming all other things being equal, one could expect that individuals are
trust game and a prisoner’s dilemma game. This kind of punishment is termed third party punishment
(Fehr & Fischbacher, 2004). The present work focuses on punishment that occurs when individuals are
directly affected by uncooperative behavior, so third party punishment is neglected in the remainder of the
text.
INTRODUCTION 14
less likely to punish in a public goods game (vs. in an ultimatum game) where
individuals have to punish using their own endowment. Empirical research on the
endowment effect and costly punishment is, however, lacking. In the end it is a (still
uninvestigated) empirical question whether those who punish in a public goods game
will also punish in an ultimatum game. Despite the differences between the two games,
both modes of punishment reflect an impulsive response to the perceived unfair behavior
of another individual (cf. Chapter 2 “Why and under which conditions individuals
punish”). That is, both modes of punishment have the same core. Therefore, in the
remainder of the current text and when referring to literature regarding costly
punishment, costly punishment in a public goods game will not be distinguished from
costly punishment in an ultimatum game if the decisive factor for why a specific study is
mentioned (e.g., punishment as a response to unfair behavior) is relevant in both games.
In the studies reported on below, the paradigm by Fehr and Gächter (2002) is used to
investigate costly punishment in public goods situations. In this paradigm, individuals
typically behave under conditions of anonymity, that is, individuals’ behavior cannot be
traced back to them. Additionally, one-shot games were applied, that is, individuals
interacted only once with another individual. Thus, strategic considerations, such as
personal reputation building, mutual reciprocity, and shaping other individuals’ behavior
to gain personal benefits, were minimized (cf. Fehr & Gächter, 2002). This represents a
much reduced situation and it is questionable whether these kinds of situations emerge
outside the laboratory raising fundamental external validity concerns (cf. Feinberg,
Cheng, & Willer, 2012; Guala, 2012; Guney & Newell, 2013).
However, the benefits of investigating costly punishment in a reduced situation are
manifold from the perspective of basic research. Specifically, a causal analytic approach
can be applied; moreover, variables can be controlled, for instance when minimizing
strategic considerations of individuals that may interfere or interact with the central
research question. This is particularly important given that we investigated whether
costly punishment is strategically executed in order to gain a valuable reputation (see
below and Pfattheicher, Landhäußer, & Keller, 2014, Appendix C). If costly punishment
INTRODUCTION 15
had already been strategically executed (i.e., if strategic considerations had not been
minimized), the likelihood that we would have been able to have observed an effect in
terms of reputation on top of already strategically executed punishment would have been
fairly low. In other words, it would have been difficult to investigate the central research
question of whether costly punishment is executed to strategically gain a valuable
reputation.
I do not deny that it is worthwhile investigating costly punishment in situations in which
it can be used as a strategy (e.g., when repeated interactions are possible). In fact,
individuals strategically adapt their behavior in accordance with a maximizing benefit
account (Halevy et al., 2012). However, before we add this additional factor that will
likely change the nature of costly punishment we need to understand the basic notion of
costly punishment. These thoughts also fit the aim of using parsimonious paradigms to
investigate basic human behavior (van Lange et al., 2014).
These considerations are also central for the three other papers focusing on costly
punishment presented below. The aim of these articles was to understand the basic
notion of costly punishment by linking it to cortisol and reactive aggression (Pfattheicher
& Keller, in press, Appendix A) and by investigating the basal motivational orientations
that may underlie costly punishment (Pfattheicher & Keller, 2013, Appendix D). The
same applies to the analysis of antisocial punishment (Pfattheicher et al., in press,
Appendix E). The goal here was to understand the basal motives and endocrinological
factors of costly punishment beyond strategic considerations.
In sum, the present dissertation contributes to the understanding of costly punishment in
public goods situations while focusing on the essence of costly punishment. Following
this central aim it was methodologically necessary to apply parsimonious paradigms
such as the public goods game of Fehr and Gächter (2002).
WHY AND UNDER WHICH CONDITIONS INDIVIDUALS COSTLY PUNISH 16
2 Why and under which conditions individuals costly punish
The present dissertation aims to understand costly punishment from a biopsychological
perspective. Accordingly, this chapter deals with the question of why and under which
conditions individuals punish. This chapter will endeavor to build the conceptual
framework for the empirical papers presented below. Let me begin by reviewing
different possible perspectives to investigate costly punishment.
2.1 Different perspectives on how to investigate costly punishment
When solving social dilemma situations one can apply the three general perspectives
proposed by Kollock (1998), that is, the structural perspective, the strategic perspective,
and the motivational (psychological) perspective. These three general perspectives can
also be applied when examining costly punishment. The structural perspective refers to
the possibility of costly punishment changing the structure of a social dilemma situation.
Specifically, by implementing the option to costly punish uncooperative group members
one can increase the cost of uncooperative behavior (because uncooperative behavior is
likely to be punished; see Fehr & Gächter, 2002). As a consequence, cooperation levels
are substantially higher when an option to costly punish is implemented (vs. not
implemented; Fehr & Gächter, 2002) because individuals are motivated to avoid the
costs of being punished (Berg, Dickhaut, & McCabe, 1995). Thus, costly punishment
can represent a structural approach to increasing cooperation.
The strategic perspective refers to the notion that actors try to (strategically) adapt their
own behavior and shape the behavior of others in order to gain personal benefits (Halevy
et al., 2012). Costly punishment can be used as an individual strategy to gain benefits in
several respects, for instance through increased cooperative behavior of a punished
individual in the future, given repeated and mutual interactions. Costly punishment may
also be used as a strategy to acquire a good reputation which may benefit the costly
punisher in future interactions (Barclay, 2006; Nelissen, 2008; but see Chapter 3.3 below
regarding this issue).
The core of the structural and strategic perspective is the fundamental goal to maximize
benefits (Kollock, 1998). It is true that individuals’ behavior can be shaped by structural
WHY AND UNDER WHICH CONDITIONS INDIVIDUALS COSTLY PUNISH 17
approaches (Camerer, 2003), for instance through costly punishment (Fehr & Gächter,
2002). It is also true that individuals can strategically and deliberately shape another
individual’s behavior which is, in principle, possible through costly punishment (Boyd &
Richerson, 1992). However, one has to ask the question whether costly punishment is
actually used as a strategy by individuals. That is, do individuals engage in costly
punishment to maximize benefits (e.g., to gain a valuable reputation)?
Obviously, there is a difference in describing a behavioral tendency as having the
potential to be used as a strategy and the notion that a behavioral tendency is actually
used as a strategy. These considerations refer to the question of why individuals punish
and to the motives and psychological motivations that drive costly punishment (i.e., the
motivational perspective). The next chapter outlines in more detail the underlying
motives and boundary conditions of costly punishment.
2.2 Costly punishment from a biopsychological perspective
2.2.1 Fairness and justice
As outlined above, individuals have incentives for behaving selfishly in public goods
situations (i.e., contributing very little or nothing) which damages the collective benefit
of the other individuals involved in the situation. A low contribution is often considered
unfair (e.g., Crockett, Clark, Lieberman, Tabibnia, & Robbins, 2010; Fehr &
Rockenbach, 2003) or a violation of a fairness norm based on fundamental human justice
concerns (e.g., Crockett, 2009; Fehr & Fischbacher, 2003, 2004). A low contribution
violates distributive justice in that the principle of equal outcomes and equal
contributions is violated which results in the perception of unfairness (Fowler, Johnson,
& Smirnov, 2005; van Dijk & Wilke, 2000). When fundamental moral principles are
violated, moral emotions emerge in individuals (Haidt, 2003). The moral emotion of
anger (but not other basic emotions such as sadness; Srivastava, Espinoza, & Fedorikhin,
2009) is consistently reported as the emotion that drives costly punishment in ultimatum
games (Pillutla & Murnighan, 1996; Rotemberg, 2008; van’t Wout, Kahn, Sanfey, &
Aleman, 2006) and in public goods games (Dawes, Fowler, Johnson, McElreath, &
Smirnov, 2007; Fehr & Gächter, 2002; Small & Loewenstein, 2005).
WHY AND UNDER WHICH CONDITIONS INDIVIDUALS COSTLY PUNISH 18
Haidt (2003) and others (e.g., Nisbett & Cohen, 1996) have pointed out that anger often
elicits the motivation to take revenge. On this basis one can assume that uncooperative
behavior that is considered to be unfair elicits anger which in turn results in the
motivation to take revenge. Evaluating these considerations, one could deduce that given
the striking evidence that costly punishment results from anger about unfair behavior,
costly punishment reflects a desire to seek revenge (e.g., de Quervain et al., 2004;
Gollwitzer & Bushman, 2012; McCullough, 2008). However, anger can also elicit the
motivation to reduce injustice (Haidt, 2003), so it is also possible that costly punishment
reflects a strategy to reinstall justice. The issue of whether costly punishment is driven
by the motivation to take revenge and/or the motivation to reinstall justice (be it
consciously or unconsciously) is still a neglected issue in the field of costly punishment.
What can be said at this point is that uncooperative behavior is typically considered to be
a violation of a fairness norm. This perception elicits anger which in turn results in costly
punishment.
Although it is proposed that the concern for fairness and justice is deeply rooted in
humans (McCullough, 2008; Sober & Wilson, 1998), individuals differ in their construal
of fairness and justice. One central axiom in social psychology captures this point: the
subjective construction of reality (Smith & Mackie, 2007). This notion corresponds to
the Thomas-Theorem “if men define situations as real, they are real in their
consequences” (Thomas & Thomas, 1928) and the notion of “the [subjective] definition
of the situation” (Esser, 2000). Applied to the present context, this means that a
relatively low contribution to the public good is not per se unfair; it must be considered
to be unfair, that is, another’s behavior must be subjectively construed to be an unfair
action. Having subjectively construed an action to be unfair, the subsequent cognitive,
affective, and physiological consequences are real for this individual.
As stated, reality is a subjective construction and individuals differ in its construal. For
instance, research on aggression has shown that individuals differ in their subjective
hostile interpretations of especially ambiguous behavior of others (i.e., the hostile
interpretation bias; Crick & Dodge, 1996; Matthews & Norris, 2002). Likewise, one can
WHY AND UNDER WHICH CONDITIONS INDIVIDUALS COSTLY PUNISH 19
assume that individuals differ in their interpretation of a relatively low contribution with
respect to unfairness (Srivastava et al., 2009). Indeed, research on fairness has shown
that individuals possessing a special sensitivity to unfairness show more anger when an
event occurs that can be considered unfair (Schmitt, Baumert, Fetchenhauer, Gollwitzer,
Rothmund, & Schlösser, 2009; Schmitt, Neumann, & Montada, 1995). These findings
are in line with work by Fetchenhauer and Huang (2004) who documented that
individual differences in sensitivity to unfairness were positively related to costly
punishment in an ultimatum game.
A perspective of individual differences reflecting equity and fairness considerations is
also present in two frequently cited articles in the field of behavioral economics (i.e.,
Bolton & Ockenfels, 2000, and Fehr & Schmidt, 1999). These articles represent
cornerstones in this field of research because they are said to deviate from standard
economic theory which relies on homo economicus. According to Fehr and Schmidt
(1999; Bolton and Ockenfels, 2000, made almost similar assumptions), those individuals
who are especially prone to punish possess a so-called “self-centered inequity aversion”.
It is self-centered in the way that these “people do not care per se about inequity that
exists among people but are only interested in the fairness of their own material payoff
relative to the payoff of others” (Fehr & Schmidt, 1999, p. 819). The Fehr and Schmidt
(1999) model does not actually strongly deviate from standard economic models but
shifts the perspective from maximizing one’s own material benefits (standard economic
model) to maximizing one’s own material benefits and costly punishing those who have
jeopardized one’s own material benefits relative to others (i.e., the principle of equity).
This also holds true for the model proposed by Bolton and Ockenfels (2000).6 One can
conclude that in the field of behavioral economics also individual differences in equity
and fairness considerations are acknowledged to predict costly punishment.
6
Selten (1998) emphasized that it is not the objective deviation that counts but the subjectively perceived
“inequity aversion”. This point was not included in the Fehr and Schmidt (1999) model. It is also unclear
why they restricted their model to “self-centered inequality aversion” given that it is well documented that
individuals also react to inequality when they are not directly affected (e.g., Bendor & Mookherjee, 1990).
WHY AND UNDER WHICH CONDITIONS INDIVIDUALS COSTLY PUNISH 20
Another relevant individual difference is that of right wing authoritarianism (RWA). One
dimension of RWA (i.e., authoritarian aggression) reflects the belief that ingroup norm
deviants have to be punished (Cohrs, 2013; Kessler & Cohrs, 2008). RWA is relevant
here because uncooperative behavior can be considered to be a violation of a fairness
norm by an ingroup member (Crockett, 2009; Fehr & Fischbacher, 2003, 2004). Thus,
one can assume that individuals who possess the belief that ingroup norm deviants have
to be punished (i.e., individuals with a relatively strong disposition for authoritarian
aggression) are more likely to engage in costly punishment as compared with individuals
who possess only a weak attitude of authoritarian aggression (cf. Kessler & Cohrs,
2008).
Another individual difference worth mentioning with regard to norm violations is the
prevention focus. Specifically, prevention-focused individuals possess a vigilant
sensitivity to negative social information, especially regarding norm violations (Higgins,
1997; Keller & Pfattheicher, in press, Appendix G). On this basis one can argue that
prevention-focused individuals should be particularly sensitive to the negative event of
an unfair contribution, which likely reflects a violation of a fairness norm (e.g., Crockett,
2009; Fehr & Fischbacher, 2003, 2004). In one of the empirical papers presented below
(Pfattheicher & Keller, 2013, Appendix D), it is shown that strongly prevention-focused
individuals are indeed more likely to engage in costly punishment as compared with
weakly prevention-focused individuals.
To sum up, there is strong empirical evidence that costly punishment is predominantly
driven by anger on the basis of a fairness norm violation by low contributors. Moreover,
as conceptualized in the approaches of individual differences, one can assume that
individuals differ in their inclination to punish unfair others. The question that is asked in
the following part is whether costly punishment is driven by impulsivity or by a more
deliberate process. This is worth discussing because: a) there are contrary perspectives
regarding this issue, and b) to shed more light on the psychological process of costly
punishment.
WHY AND UNDER WHICH CONDITIONS INDIVIDUALS COSTLY PUNISH 21
2.2.2 Impulsivity
Theories on social cognition typically emphasize two systems of information processing
(e.g., Bless, Fiedler, & Strack, 2004; Gawronski & Bodenhausen, 2006, 2007; Petty,
Cacioppo, & Goldman, 1981; Strack & Deutsch, 2004, 2012). In the first system,
information is processed automatically, unconsciously, and effortlessly in terms of
cognitive resources. In the second system, information is processed in a controlled,
conscious, and effortful way. The preconditions for the second system to operate are the
motivation to deliberatively process and the availability of cognitive capacities to engage
in deliberative processing (Fazio, 1990). For instance, Strack and Deutsch (2004, 2012)
have postulated the existence of an impulsive system and a reflective system which can
operate in parallel. In the impulsive system, behavioral outcomes are said to be driven by
an automatically spreading activation in an associative network, resulting in relatively
rapidly executed behavior after a stimulus occurred. In the reflective system, behavioral
outcomes are said to be driven by a deliberative, resource-consuming decision process,
resulting in relatively slowly executed behavior after a stimulus occurred.
The impact of the impulsive and reflective system with regard to economic choice is an
issue of recent research. Rand, Greene, and Nowak (2012) have shown that the more
individuals contribute to a public good the faster, that is, the more impulsively (Strack &
Deutsch, 2004, 2012) they make their decision. In several subsequent studies Rand et al.
(2012) have documented that prosocial, as compared with antisocial, behavior in
different economic games is rather impulsively than deliberately executed. Generally
speaking, the research by Rand et al. (2012) contributes to the debate over whether
humans are basically selfish actors and whether this selfish “default mode” needs to be
overwritten by the reflective system (e.g., Binmore, 2005; Binmore & Shaked, 2010;
DeWall, Baumeister, Gailliot, & Maner, 2008; Haidt & Kesebir, 2010) or whether
humans can also act as genuine prosocial actors with no need for the reflective system to
control a selfish default mode (Haidt & Kesebir, 2010; Zaki & Mitchell, 2013). Research
by Rand et al. (2012; see also Cappelletti, Goth, & Ploner, 2011; Schulz, Fischbacher,
Thöni, & Utikal, in press) indicates that reducing the impact of the reflective system
(e.g., through ego depletion or stimulating intuitive thinking; Cappelletti et al., 2011;
WHY AND UNDER WHICH CONDITIONS INDIVIDUALS COSTLY PUNISH 22
Rand et al., 2012; Schulz et al., in press) increases prosocial behavior and cooperation. If
the reflective system had overwritten a selfish default mode, selfish behavior should
have been more likely to emerge when the reflective system had been reduced. Actually,
the reverse was the case, which speaks to the notion that humans can act as genuinely
prosocial actors.7
Regarding costly punishment, some scholars have argued that costly punishment is based
on the reflective system, specifically, that costly punishment is deliberatively executed in
order to uphold a fairness norm (Knoch & Fehr, 2007; Knoch, Pascual-Leone, Meyer,
Treyer, & Fehr, 2006;). Others have argued that costly punishment is based on the
impulsive system, specifically, that costly punishment is executed impulsively (Crockett
et al., 2010; Sanfey, Rilling, Aronson, Nystrom, & Cohen, 2003; Tabibnia, Satpute, &
Lieberman, 2008).
Empirical evidence for the notion that costly punishment is based on the reflective
system comes from research on a brain region called the right dorsolateral prefrontal
cortex (rDLPFC) which is related to self-controlled, reflective processes (Miller &
Cohen, 2001). Knoch et al. (2006) have shown that disrupting rDLPFC activity via
transcranial magnetic stimulation reduces costly punishment in an ultimatum game,
suggesting that in the default mode (when the rDLPFC is not disrupted) self-controlled
processes are at work driving costly punishment. However, exactly the opposite results
have been documented by Brüne and colleagues (Brüne, Juckel, & Enzi, 2013; Brüne et
al., 2012). In their studies, disrupting the rDLPFC increased costly punishment in an
ultimatum game and a third-party punishment paradigm. Further neuropsychological
evidence for the notion that costly punishment is based on the impulsive system comes
7
Of note, when integrating research on aggression and ego depletion, that is, ego depletion increases
aggression after a provocation (DeWall, Baumeister, Stillman, Gailliot, 2007; Stucke & Baumeister, 2006),
one has to conclude that humans are not per se prosocial or antisocial but that it is dependent on the
context. Specifically, when provoked, individuals have to override an antisocial tendency (i.e., aggression;
DeWall et al., 2007; Stucke & Baumeister, 2006), but in more affectively neutral contexts there seems to
be a tendency in favor of the assumption that humans can also act as genuinely prosocial actors with no
need for the reflective system to control a selfish default mode (Zaki & Mitchell, 2013).
WHY AND UNDER WHICH CONDITIONS INDIVIDUALS COSTLY PUNISH 23
from Tabibnia et al. (2008). These authors were able to document a positive relation
between activation in the right ventrolateral prefrontal cortex (rVLPFC), a brain region
that is associated with regulatory processes in terms of emotions, and individuals’
acceptance of unfair offers in the ultimatum game.8 This result suggests that refraining
from costly punishment (accepting an unfair offer) for selfish reasons requires
regulatory, controlled processes. To sum up the neuropsychological literature, findings
for the notion that costly punishment is based on the reflective system are rather mixed.
Social psychological research has also dealt with the question of which system costly
punishment may be based on. Wang et al. (2011) have investigated costly punishment in
an ultimatum game in which they varied whether individuals could punish immediately
after an unfair offer occurred versus a time delay between the unfair offer and the
decision to punish. Results based on three studies involving different time delay
conditions (simply counting numbers for 5 minutes, no distraction for 5 or 2.5 minutes)
convincingly showed that costly punishment was substantially reduced when individuals
made the decision to punish after a time delay. Thus, disturbing the possibility of
impulsivity using a time delay (e.g., Papies, Barsalou, & Custers, 2012), that is, taking
away the impulsive component of the punishment decision, reduces the amount of costly
punishment.9 From these findings one can infer that costly punishment is an impulsive
choice.
Moreover, Ben-Shakhar, Bornstein, Hopfensitz, and van Winden (2007) showed in a
8
One has to note, however, that the correlation between the activation of the rVLPFC and individuals’
acceptance of unfair offers was based on only eleven participants and was extremely high (r = .76),
possibly too high to be true (cf. Vul, Harris, Winkielman, & Pashler, 2009).
9
This experimental approach is in line with the notion of showing a process by disrupting it (Jacoby &
Sassenberg, 2011). Specifically, a default condition in which the hypothesized process can function is
compared to a condition in which the process is eliminated. The behavior of interest (or in general ‘the
outcome’) should be reduced when the process that accounts for the outcome is eliminated (cf. Spencer,
Zanna, & Fong, 2005). In the study by Wang et al. (2011), the process (i.e., the impulsive system) is
disrupted by the time delay and a reduction in costly punishment is shown.
WHY AND UNDER WHICH CONDITIONS INDIVIDUALS COSTLY PUNISH 24
power-to-take game10 that costly punishers as compared with non-punishers had a
significantly greater skin conductance level (SCL) before and after engaging in costly
punishment. SCLs are related to arousal (e.g., Critchley, Elliott, Mathias, & Dolan,
2000).11 Thus, these results imply that costly punishment accompanies arousal. Strack
and Deutsch (2004) argued that high arousal interrupts the reflective system making the
impulsive system more dominant. Consequently, the findings of Ben-Shakhar et al.
(2007) bolster the assumption that costly punishment is based on the impulsive system.
Additional evidence that costly punishment reflects an impulsive action stems from
research on ego-depletion (cf. Baumeister, 2002). Research on ego-depletion has shown
that impulsive behavior is more likely to emerge when cognitive resources are depleted
(DeWall et al., 2007; Robinson, Schmeichel, & Inzlicht, 2010; Stucke & Baumeister,
2006). Interestingly, Halali, Bereby-Meyer, and Meiran (in press) showed that egodepletion increased costly punishment in an ultimatum game. In another study, these
authors showed that participants were faster in terms of reaction times in rejecting an
unfair offer than in accepting it. As outlined above, faster reactions are assigned to the
impulsive system (Strack & Deutsch, 2004, 2012).
Taking a biopsychological perspective on costly punishment in terms of impulsivity,
recent research has shown that the basal level of the neurotransmitter serotonin in blood
is negatively related to costly punishment in an ultimatum game (Emanuele, Brondino,
Bertona, Re, & Geroldi, 2008) and, accordingly, that the depletion of tryptophan, which
lowers brain serotonin levels, increases costly punishment (Crockett et al., 2010). This is
particularly noteworthy because the depletion of serotonin was positively related to the
expression of anger and impulsivity in other research (Castrogiovanni, Capone,
10
In a power-to-take game, Player A can take money from the endowment of Player B. Subsequently,
Player B can destroy both assets (both receive nothing which is equivalent to the ultimatum game).
Destroying both assets reflects costly punishment if Player B would still have possessed something.
11
Of note, SCLs are not valence specific; for instance, Khalfa, Isabelle, Jean-Pierre, and Manon (2002)
found that SCLs increase when an individual listens to fearful music but also when they listen to happy
music.
WHY AND UNDER WHICH CONDITIONS INDIVIDUALS COSTLY PUNISH 25
Maremmani, Marazziti, 1994; Kavoussi, Armstead, & Coccaro, 1997; Krakowski, 2003),
supporting the notion that costly punishment is indeed driven by impulsivity.
To sum up the literature, costly punishment can be considered to be predominantly
driven by anger (Dawes et al., 2007; Fehr & Gächter, 2002; Pillutla & Murnighan, 1996;
Rotemberg, 2008; Small & Loewenstein, 2005; van’t Wout et al., 2006) on the basis of a
fairness norm violation by low contributors (Cubitt, Drouvelis, Gächter, & Kabalin,
2011; Fehr & Fischbacher, 2004; Knoch, Gianotti, Baumgartner, & Fehr, 2010). Some
scholars have argued that costly punishment is based on the reflective system (Knoch et
al., 2006; Knoch & Fehr, 2007) whereas others have argued that costly punishment is
based on the impulsive system, specifically, that costly punishment is impulsively
executed (Crockett et al., 2010; Sanfey et al., 2003; Tabibnia et al., 2008). Evaluation of
the literature reviewed above reveals that the evidence largely supports the impulsive
account. This does not mean that every decision to punish is impulsively executed.
Rather, costly punishment seems to be predominately impulsively executed.
2.2.3 Anticipated reward
Some scholars argue that “all vengeful acts are attempts to escape pain” (Frijda, 1994, p.
279). Applied to the context of costly punishment the argument is put forward that
individuals engage in costly punishment because they anticipate reward (Carlsmith,
Wilson, & Gilbert, 2008; de Quervain et al., 2004; Knutson, 2004). That is, after the
negative experience of anger in response to the unfair behavior of an uncooperative
other, individuals assume that they can improve their mood by punishing the
uncooperative other. Of note is the idea that anticipated reward is not in contrast with the
impulsivity approach because the basic motive of an impulsive action can indeed be
anticipated reward. This notion is well in line with Gray’s reinforcement sensitivity
theory (RST; Gray, 1970, 1982) and empirical findings (Torrubia, Avila, Moltó, &
Caseras, 2001) that emphasize reward-seeking through impulsive actions. Self-control
theory (Mischel, 1961, 2012) specifies this relation and postulates that impulsive actions
(based on a lack of self-control) are often executed to seek short-term rather than longterm rewards.
WHY AND UNDER WHICH CONDITIONS INDIVIDUALS COSTLY PUNISH 26
Empirical evidence does suggest that when individuals are treated unfairly they expect to
feel better after engaging in costly punishment (Carlsmith et al., 2008; of note, in this
study participants actually felt worse after engaging in costly punishment). In line with
this finding, de Quervain et al. (2004) found that the contemplation of costly punishment
is related to activation in a brain region typically associated with reward (i.e., the dorsal
striatum; Balleine, Delgado, & Hikosaka, 2007). On basis of these findings, de Quervain
et al. (2004) argue that individuals engage in costly punishment because they anticipate
reward. However, the argument of de Quervain et al. (2004) seems to be a weak one as
the activation of the dorsal striatum is also associated with aversive, unexpected, or
novel stimuli (e.g., Bromberg-Martin, Matsumoto, & Hikosaka, 2010; Delgado, Li,
Schiller, & Phelps, 2008). Based on the logic of de Quervain et al. (2004), one can
equally argue that individuals punish because they anticipate feeling worse (which is
obviously not plausible) or because they anticipate engaging in a novel experience by
means of punishing. What must be said is that the activation of a brain area is often too
unspecific to enable conclusions to be drawn about the motives of behavior. Thus, the
study of de Quervain et al. (2004) actually remains salient regarding the driving forces of
costly punishment (cf. Peacock, 2007), although it is frequently citied to support the idea
of anticipated reward when punishing (e.g., Guala, 2012).
Gollwitzer and Bushman (2012) have questioned that individuals typically punish in
order to improve their mood or to seek reward. They argue that costly punishment
reflects retributive punishment elicited by the violation of social norms and that people
punish unfair others in order to give the uncooperative others what they deserve, but not
merely because they want to improve their mood. Gollwitzer and Bushman (2012) have
tested the assumption that individuals strategically engage in costly punishment in order
to feel better. They implemented two conditions and used costly punishment in a public
goods game as the dependent variable. In the first condition, participants were told that
they could not improve their mood in the study (because their mood was frozen by
means of a pill, in reality a placebo). In the second condition, participants were not given
a pill. Gollwitzer and Bushman (2012) first showed that anger about uncooperative
WHY AND UNDER WHICH CONDITIONS INDIVIDUALS COSTLY PUNISH 27
behavior did not differ between the conditions.12 If participants improve their mood by
means of costly punishment, costly punishment would be more likely to be shown in the
condition in which the participants could actually improve their mood, that is, the
condition without mood freezing. However, Gollwitzer and Bushman (2012) reported
that costly punishment did not differ across the two conditions (although statistical
power was sufficiently high).13 Remarkably, this finding was able to be conceptually
replicated twice by the authors (see Study 2 and 3 in Gollwitzer & Bushman, 2012).
By evaluating the (although sparse) empirical findings on anticipated reward one can
come to the conclusion that individuals believe that engaging in costly punishment
improves their mood (Carlsmith et al., 2008). However, as Gollwitzer and Bushman
(2012) showed, participants did not strategically engage in costly punishment in order to
improve their mood. One has to note that merely agreeing with the notion that one can
improve one’s mood by means of costly punishment (like the participants in Carlsmith et
al., 2008) is different from anticipated reward being the actual motive. Nevertheless,
Carlsmith et al. (2008) stated that “people use punishment to strategically repair their
negative mood” (p. 1323, emphasis added). However, this was not tested by these
authors; they did not show, for instance, an association between the belief in anticipated
reward and costly punishment. Only Gollwitzer and Bushman (2012) directly tested
whether individuals strategically engage in costly punishment to improve their mood.
Therefore, the conclusion of Gollwitzer and Bushman (2012) should be preferred, that is,
individuals typically do not strategically costly punish to improve their mood.
The study by Gollwitzer and Bushman (2012) speaks against the “strategic perspective”
of costly punishment, specifically, that costly punishment is part of a benefit-maximizing
12
Indeed, one could criticize the “pill-study” because if the pill actually had frozen one’s mood, anger
should not be elicited. However, given that Gollwitzer and Bushman (2012) replicated the main finding
twice with different methods, their conclusion seems valid.
13
This still represents a problem because with typical statistical methods (as used by Gollwitzer &
Bushman, 2012) one cannot accept but only reject a null-hypothesis. Gollwitzer and Bushman (2012)
could have tested against a small effect, as recently proposed by Simonsohn (2013). That is, to reject the
null-hypothesis that there is a small effect leading to the conclusion that there is a theoretically negligible
smaller-than-a-small-effect or a null-effect.
WHY AND UNDER WHICH CONDITIONS INDIVIDUALS COSTLY PUNISH 28
strategy (i.e., the psychological benefit of improved mood). Building on the findings of
Gollwitzer and Bushman (2012), it seems unlikely that costly punishment is typically
part of a (mood repairing) strategy. Still, this finding does not contradict the notion that
costly punishment is strategically used in another sense, for instance to gain a valuable
reputation. This notion is discussed in the following section.
2.2.4 Reputation
In their seminal study, Fehr and Gächter (2002) wrote: “To rule out that repetition
created cooperation or punishment through [...] reputation, the group composition
changed from period to period such that no subject ever met another subject more than
once” (p. 137, emphasis added). Several years later, Gächter and Herrmann (2009) wrote
that individuals “cooperate and punish others even if there are no gains from future
cooperation or any other reputational gains” (p. 791, emphasis added). The question that
arises from these statements is: do individuals strategically punish to gain a good
reputation? The statements of Fehr and Gächter (2002) as well as Gächter and Herrmann
(2009) do suggest this. Also, the considerations of several other scholars point in this
direction (Barclay, 2006; Dos Santos, Rankin, & Wedekind, 2011; Hagen &
Hammerstein, 2006; Hilbe & Traulsen, 2012; Kurzban et al., 2007; Nelissen, 2008;
Piazza & Bering, 2008; Shaw & Santos, 2012; Sylwester, Herrmann, & Bryson, 2013;
Tennie, 2012). If individuals strategically punish to gain a good reputation, they should
be more likely to engage in costly punishment when reputational concerns are activated.
In the empirical paper presented below (Pfattheicher et al., 2014, Appendix C) the
opposite is shown: costly punishment is actually reduced under conditions of
reputational concerns. Reconsidering the strategic perspective on costly punishment
(costly punishment as part of a benefit-maximizing strategy), the findings speak against
the notion that costly punishment is strategically used to gain benefits, in this case
reputational benefits. The issue of reputation and costly punishment is outlined in detail
below during the discussion of the empirical paper (Pfattheicher et al., 2014, Appendix
C).
WHY AND UNDER WHICH CONDITIONS INDIVIDUALS COSTLY PUNISH 29
2.3 Summary
The previous chapter has dealt with the question of why and under which conditions
individuals punish. One can conclude that costly punishment can be considered to be
driven by anger on the basis of a fairness norm violation by low contributors. It reflects
behavior that seems to result from the impulsive system rather than the reflective system,
that is, costly punishment can be considered to be an impulsive act. Furthermore,
research on anticipated reward and reputation suggests that costly punishment is neither
strategically executed in order to improve one’s mood nor in order to gain a valuable
reputation. The findings, therefore, question the benefit-maximizing strategic
perspective.
Although research on costly punishment has produced a large amount of articles in
recent years (to date, Fehr and Gächter’s seminal article of 2002 has been cited more
than 2,700 times according to Google Scholar), considerable research gaps still exist.
The empirical papers I present in the following sections all address several neglects in
the field of costly punishment. One research gap relates to the endocrinology of costly
punishment. Specifically, in the first paper on costly punishment it is argued that costly
punishment reflects, at least to some degree, reactive aggression. We then built on recent
findings documenting that basal cortisol was negatively related to reactive aggression
(e.g., Böhnke, Bertsch, Kruk, & Naumann, 2010). In line with these findings, it was
possible to show a negative relation between cortisol and costly punishment, a relation
that had not been previously documented. The first paper not only aimed to address a
research gap by investigating the relation of basal cortisol and costly punishment, it was
also written to explicitly link costly punishment to reactive aggression. That is, the first
manuscript also makes a point on the conceptual level.
The second paper on costly punishment deals with the strategic perspective on costly
punishment. As outline above, several scholars consider costly punishment to be
strategically executed to gain reputational benefits, although this notion was barely
tested. As such, the second paper on costly punishment constitutes an important
contribution in that it empirically – and not only on a theoretical level like so many
others before – addresses the relation of reputation and costly punishment.
WHY AND UNDER WHICH CONDITIONS INDIVIDUALS COSTLY PUNISH 30
The third paper on costly punishment attends to another gap in research on costly
punishment, specifically, linking costly punishment to individual differences. As far as I
can judge following an extensive literature review, the paper presented below reflects the
first piece of research that considers the relation of individual differences and costly
punishment in public goods situations empirically. Specifically, it is shown that
dispositional prevention focus (Higgins, 1997) but not dispositional behavioral inhibition
(Carver & White, 1994) predicts costly punishment. That is, the aim was to investigate
costly punishment (a) from the perspective of individual differences and (b) from the
perspective of basal motivational orientations and to contribute in that way to a better
understanding of costly punishment.
The fourth paper on costly punishment deals with the “dark side” of costly punishment.
In this paper the biopsychology of antisocial punishment, that is, the punishment of
cooperative individuals was investigated. Little is currently known about this behavior
that reliably emerges in public goods games when an option to punish is available
(Dreber & Rand, 2012; Herrmann, Thöni, & Gächter, 2008). The paper proposes that,
when scoring off cooperative group members, dominance motives are at work. In line
with these considerations, it is shown that the dominance related hormone testosterone as
well as psychological dominance motives predict antisocial punishment under the
condition of a relatively low cortisol level (reflecting the so-called dual-hormonehypothesis, see Carré & Mehta, 2011; Mehta & Josephs, 2010). Of note, the present
paper reflects the first empirical work that aims to shed light on the motives and
physiology of antisocial punishment and, therefore, also attempts to fill a considerable
research gap.
In sum, the papers presented in the following section paint a biopsychological picture of
costly punishment. This is the aim of the present dissertation: to understand costly
punishment from a biopsychological perspective. The empirical papers presented below
all address specific gaps in the field of costly punishment. As a whole they shed light on
underlying biopsychological factors and boundary conditions of costly punishment in
public goods situations.
EMPIRICAL PAPERS 31
3 Empirical papers
In this chapter I present and discuss five empirical papers that are the result of my time
as a doctoral candidate (a sixth, a seventh, and an eighth paper do not exactly fit into the
theme of the present dissertation but are attached in Appendix F, G, and H).
3.1 Towards a biopsychological understanding of costly punishment: The role of
basal cortisol (Pfattheicher & Keller, in press, Appendix A)
This paper is built on recent findings that have documented an inverse relation of basal
cortisol and aggression after a provocation (i.e., reactive aggression; Böhnke, Bertsch,
Kruk, & Naumann, 2010; Böhnke, Bertsch, Kruk, Richter, & Naumann, 2010; Feilhauer,
Cima, Korebrits, & Nicholson, 2013; Poustka et al., 2010).14 Reactive aggression can be
defined as an angry, impulsive response to perceived provocation or interpersonal
frustration (Crick & Dodge, 1996; van Goozen, Fairchild, Snoek, & Harold, 2007).15 As
outlined above, anger seems to be the emotion that drives costly punishment, and it
seems to be an impulsive response to uncooperative behavior and the violation of a
fairness norm, that is, a perceived provocation or an interpersonal frustration. Therefore,
given that costly punishment corresponds to the main criteria of reactive aggression, one
can reasonably assume that costly punishment reflects, at least to some degree, reactive
aggression. Indeed, Ernst Fehr16 himself called costly punishment “retaliation” (Fehr &
Fischbacher, 2004) which is used as a synonym for reactive aggression (Hubbard,
Dodge, Cillessen, Coie, & Schwartz 2001).
14
Of note, Alink et al. (2008) found no significant inverse relation between basal cortisol and aggression in
a meta-analysis for adolescents (it was only found for young children). However, Alink et al. (2008) did
not differentiate between specific forms of aggression (e.g., reactive or proactive aggression, overt or
covert aggression, social ostracism, blind rage etc.). Thus, they do not consider the issue regarding the
specific relations of specific forms of aggression and basal cortisol, which would be, however, quite
important given that specific forms of aggression are differently correlated with basal cortisol (see e.g.,
Poustka et al., 2010).
15
A clear distinction between reactive and proactive aggression in terms of costly punishment is made in
the empirical paper (Pfattheicher & Keller, in press, Appendix A).
16
The first author of the seminal article Fehr and Gächter (2002).
EMPIRICAL PAPERS 32
On this basis we assumed that costly punishment should be related to biopsychological
factors similar to reactive aggression, in this case, the steroid hormone cortisol. As
reported in Pfattheicher and Keller (in press, Appendix A), basal cortisol was indeed
significantly negatively related to costly punishment.17 Remarkably, research on reactive
aggression has produced similar results to the research on costly punishment. As with
costly punishment, reactive aggression is increased when cognitive resources are
depleted (DeWall et al., 2007; Stucke & Baumeister, 2006; Robinson et al., 2010), it is
mainly driven by anger in response to a provocation (Crick & Dodge, 1996; Denson,
Pedersen, Ronquillo, & Nandy, 2009; Wilkowski & Robinson, 2008), and it reflects an
impulsive action (Crick & Dodge, 1996; Denson, von Hippel, Kemp, & Teo, 2010;
Raine et al., 2006; Richetin & Richardson, 2008).
Why, apart from corresponding empirical findings, should cortisol be inversely related to
costly punishment? Van Goozen and colleagues (van Goozen, Fairchild, & Harold, 2008;
van Goozen et al., 2007; see also Alink et al., 2008; Bushman & Huesmann, 2010;
Eysenck, 1970) postulate a theory of underarousal. They emphasize that under-aroused
individuals possess a pronounced tendency to engage in aggressive and impulsive acts.
That is to say, under-aroused individuals seek stimulation that may be possible through
behaving aggressively. Van Goozen and colleagues (van Goozen et al., 2008; van
Goozen et al., 2007) have predicted and reported corresponding literature that underaroused individuals (e.g., individuals with a low level of cortisol) are more likely to
engage in aggressive and impulsive acts as compared with normally aroused individuals
(see also Schmidt et al., 2013, and Raine et al., 2006). Accordingly, under-aroused
individuals seek stimulation that may be gained by punishing uncooperative free-riders.
Thus, one can assume that individuals with low cortisol levels tend to engage in
impulsive actions. Given that costly punishment can be considered to be an impulsive
17
In this chapter I focus on basal endogenous cortisol while acknowledging that situationally elevated
cortisol levels, individual differences in cortisol reactivity to stressors, and cortisol awakening responses
may also contribute to a better understanding of aggression. However, heterogeneous results are reported
in these vast fields of research (cf. van Goozen, 2005) which is beyond the scope of the present
dissertation.
EMPIRICAL PAPERS 33
act, it seems reasonable to assume that costly punishment is particularly shown by
individuals with low cortisol levels.18
In contrast, high levels of cortisol have been linked to behavioral withdrawal (Blair et al.,
2004; Brown et al., 1996; Montoya et al., 2012; Nava et al., 2006; Tops & Boksem,
2011; van Honk et al., 2010; Windle, 1994). In this regard, the omission of costly
punishment can be regarded as behavioral withdrawal (cf. Pfattheicher & Keller, 2013,
Appendix D) which can lead to the deduction that individuals with high cortisol levels
tend to engage in behavioral withdrawal and do not costly punish.
Another approach to the question of why endogenous cortisol and costly punishment are
inversely related relates to the concept of vigilance. Research by van Honk and
colleagues (1998, 2000) has documented that low cortisol levels are positively related to
vigilant responses to negative social information. We have shown (Pfattheicher & Keller,
2013, Appendix D) that individuals with a strong prevention focus, that is, individuals
possessing a vigilant sensitivity to negative social information, were more likely to
engage in costly punishment compared with weakly prevention-focused individuals.
Integrating these lines of research, we argue that individuals with relatively low levels of
cortisol possess a special vigilance to negative social information. Building on the notion
that the uncooperative behavior of another group member constitutes negative social
information, that is, the violation of a fairness norm, it seems plausible to assume that
individuals that are relatively low (vs. high) in basal cortisol are more likely to engage in
costly punishment due to their special vigilance regarding negative social information.19
I have proposed a multiple mediator perspective here (impulsivity, behavioral
withdrawal, vigilance), which is in line with recent theorizing that emphasizes several
routes to behavior (Bullock et al., 2010; Fiedler et al., 2011; Hayes, 2013). This is worth
mentioning because proposing several mediators is not common in social psychological
18
To establish transparency, I want to mention that this paragraph includes sentences taken from
Pfattheicher and Keller (in press, Appendix A).
19
To establish transparency, I want to mention that this paragraph includes sentences taken from
Pfattheicher and Keller (in press, Appendix A).
EMPIRICAL PAPERS 34
literature (although it is not explicitly denied). For instance, when researchers use the
Baron and Kenny (1986) approach and report that a direct path between an independent
variable X and a dependent variable Y is reduced to zero after controlling for a third
variable M, they often speak of “full mediation” (Rucker, Preacher, Tormala, & Petty,
2011). Full mediation implies that a researcher has fully explained the relation of X and
Y. However, “full mediation” simply means that the shared variance between X and Y
can be statistically explained by M. However, M can be correlated with other mediators
(Z) which are then indirectly involved in the statistical mediation of M regarding X and
Y (Fiedler et al., 2011). Then the relation of X and Y is not only explained by M but by
M and Z. It is also possible that a third variable (say, W) suppresses part of the effect of
X on Y, so controlling for W actually increases the direct effect (MacKinnon, Krull, &
Lockwood, 2000) which is then not fully mediated anymore by M, although neither M
nor X and Y have changed (Rucker et al., 2011). Basically, a multiple mediator
perspective is a worthwhile approach given that several routes can lead to specific
behavior (Bullock et al., 2010; Fiedler et al., 2011; Hayes, 2013). This perspective was
also applied in terms of the relation of basal cortisol and costly punishment.
3.1.1 Discussion
In our study (Pfattheicher & Keller, in press, Appendix A) we found that basal cortisol
was negatively related to costly punishment. This relation has not been documented
before and, therefore, successfully addresses a gap in the field of costly punishment. The
work is not only important regarding costly punishment and cortisol but also regarding
costly punishment and reactive aggression. The connection between costly punishment
and reactive aggression has been made before (Crockett et al., 2010; Osumi et al., 2012;
Stirrat & Perrett, 2010) but this empirical paper (Pfattheicher & Keller, in press,
Appendix A) represents the first work, as far as I can judge after an extensive literature
review, that explicitly builds on findings of reactive aggression and shows corresponding
results. Thus, the presented paper does not only relate costly punishment to a prominent
and frequently investigated endocrinological factor (cortisol) but also emphasizes and
explicitly states that costly punishment should be investigated from the perspective of
reactive aggression. This claim is supported by van Lange et al. (2013) who stated “[…]
EMPIRICAL PAPERS 35
it is surprising that aggression has received so little attention in social dilemmas, because
– unless research suggest otherwise – aggression seems an important orientation in social
dilemmas, albeit one that seems activate primarily by the behavior of others” (p. 75).
After having established a correlational relation between two concepts of interest (here:
cortisol and costly punishment), social psychologists often experimentally manipulate X
to investigate X’s causal influence on Y. From this consideration it seems an obvious
next step to manipulate cortisol and to analyze whether this manipulation affects costly
punishment. Indeed, an enormous amount of literature investigates the effects of
exogenous cortisol (e.g., using an oral dose of cortisol) on a variety of behavioral and
cognitive outcomes (Dickerson & Kemeny, 2004). Exogenous cortisol actually increases
endogenous cortisol levels compared with placebo groups and was even found to
increase reactive aggression in women (e.g., Böhnke, Bertsch, Kruk, Richter, &
Naumann, 2010). However, after exogenous cortisol is given, a negative feedback loop
can be observed, that is, an actually down-regulation of the HPA axis resulting in less
endogenous cortisol (Calogero, Gallucci, Gold, & Chrousos, 1988; Kruk, Halasz, Meelis,
& Haller, 2004). Additionally, hormones crosstalk on several levels (Viau, 2002), for
instance it has been observed that exogenous cortisol lowers endogenous testosterone
levels (Cumming, Quigley, & Yen, 1983). Basically, manipulating cortisol levels
through exogenous cortisol does not seem to be a valuable approach when investigating
the causal relation of endogenous cortisol and costly punishment given that exogenous
cortisol seems to affect several biopsychological variables, not only the level of cortisol
per se.
One may consider manipulating endogenous cortisol not via exogenous cortisol but via
psychological stress. For instance, the Trier Social Stress Test (TSST; Kirschbaum,
Pirke, & Hellhammer, 1993) or the newer TSST for groups (TSST-G; von Dawans,
Kirschbaum, & Heinrichs, 2011) are paradigms that are frequently used to manipulate
stress and to increase cortisol levels (although there is typically a non-negligible amount
of cortisol non-responders; see e.g., Kuhlmann & Wolf, 2006). The experimental
condition of the TSST typically includes a speech task and mathematical task carried out
EMPIRICAL PAPERS 36
in front of a jury. However, the TSST has several confounds (e.g., evaluative social
presence only in the experimental condition, or activating concepts such as “authorities”,
and mood effects; Kirschbaum & Hellhammer, 1994). The confound problem also
applies to other manipulations of stress (e.g., the cold pressure task, Smith, Edwards, &
McCann, 2007). Moreover, following the emergence of a stressor, the length of time
after which specific behavior is investigated is relevant. This is important because
cortisol activates cognitive resources to cope with a stressor, which, however, takes some
time (cf. Vinkers et al., 2013). Having more cognitive resources may strengthen the
deliberative system relative to the impulsive system which may lead to reduced
punishment. That is, it seems to depend on the time costly punishment is investigated
following the emergence of a stressor.
Thus, using the TSST and other methods to manipulate cortisol levels via social stress
seems highly problematic because one does not know exactly why specific effects are
found. To conclude, a clean experimental investigation of the causal relation of cortisol
and costly punishment is almost impossible to accomplish. The present work remains
silent regarding the causal nature of the basal cortisol-costly punishment relation. It is
possible that future research will not solve this problem.
In critically discussing the correlation between costly punishment and cortisol, it could
be the case that there is, in fact, no direct causal relation. That is to say, it is not cortisol
itself that influences costly punishment but other constructs that are associated with
cortisol (e.g., vigilance in individuals with relatively low cortisol levels). From this
perspective one can directly investigate the mediators that explain the relation of basal
cortisol and costly punishment. I have proposed that the impulsive system could be
relevant here, as relatively low basal cortisol levels as well as costly punishment are
related to impulsivity (see above). Specifically, one could manipulate or eliminate the
mediator (cf. Jacoby & Sassenberg, 2011; Spencer et al., 2005), that is, reduce the
influence of the impulsive system. Applying the approach of Wang et al. (2011), one can
include a time delay before giving individuals the option to costly punish. In this way the
impulsive component in costly punishment can be reduced (Papies et al., 2012; Wang et
EMPIRICAL PAPERS 37
al., 2011). If impulsivity is responsible for the cortisol-costly punishment relation,
cortisol should not be a predictor when the impulsive component in costly punishment is
reduced. Applying the opposite research strategy, one can activate the impulsive system,
for instance through ego-depletion (Halali et al., in press) or through time pressure
(Sutter, Kocher, & Strauß, 2003). This should increase costly punishment, especially in
individuals who are sensitive in terms of impulsivity (i.e., individuals with relatively low
cortisol levels).
The same approach could be applied with respect to another proposed mediator,
vigilance in relatively low cortisol individuals. For instance, one can decrease the
vigilant system (e.g., through an induction of mindfulness, e.g., Arch & Craske, 2006;
Grier et al., 2003) and should find decreased costly punishment, especially in individuals
with a strong vigilant system (i.e., individuals with a relatively low level of cortisol).
To conclude, the presented work (Pfattheicher & Keller, in press, Appendix A)
represents a valuable step in the investigation of the endocrinological factors that
underlie costly punishment. Our work not only aimed to address a research gap by
investigating the relation between basal cortisol and costly punishment, we also
explicitly linked costly punishment to reactive aggression and made a point on the
conceptual level. As such, our work contributes to a better understanding of costly
punishment. This goal was also pursued in the papers presented in the following section.
3.2 The eye cue manipulation as subtle induction of a sense of being watched:
Empirical validation and theoretical integration (Pfattheicher & Keller, 2014,
Appendix B)
As outlined above, some scholars have suggested that individuals strategically costly
punish in order to gain a valuable reputation (e.g., Barclay, 2006; Dos Santos et al.,
2011; Hagen & Hammerstein, 2006; Hilbe & Traulsen, 2012; Kurzban et al. 2007;
Nelissen, 2008; Piazza & Bering, 2008; Shaw & Santons, 2012; Sylwester et al., 2013;
Tennie, 2012). We tested this assumption in several studies as presented below
(Pfattheicher et al., 2014, Appendix C). In one of these studies, we activated reputational
concerns via a subtle cue of being watched (i.e., stylized watching eyes). We used this
EMPIRICAL PAPERS 38
cue of being watched because its subtle nature provides several methodological
advantages in comparison with classic manipulations of reputational concerns and social
presence manipulations such as an actual audience or a one-way mirror. These
manipulations always include confounds such as sex, attractiveness or group
membership of observers and may additionally activate other constructs such as private
self-awareness (cf. Pfattheicher & Keller, 2014, Appendix B). In the paper presented
below we validated this subtle cue of being watched.
Starting in 2005 with the seminal article by Haley and Fessler, subtle cues of being
watched have become increasingly popular (for meta-analyses, see Nettle et al., 2013,
and Sparks & Barclay, 2013). All eye cue studies published thus far are displayed in
Table 1. Papers presenting empirical studies involving subtle cues of being watched
typically report the influence of watching eyes on prosocial and antisocial behavior
(Nettle et al., 2013; Sparks & Barclay, 2013). Specifically, individuals’ prosocial
behavior increases and individuals’ antisocial behavior decreases when subtle cues of
being watched are presented.
Research on watching eyes has proposed that a sense of being seen is induced by subtle
cues of being watched (e.g., Izuma, 2012; Nettle et al., 2013). However, as yet no
empirical evidence has tested this assumption. Accordingly, there is a substantial amount
of published articles that have built on the not empirically tested assumption that subtle
cues of being watched induce an increased sense of being seen. This strongly calls for a
validation study.
We (Pfattheicher & Keller, 2014; Appendix B) tested the validity of a subtle cue of being
watched building on the spotlight effect (Gilovich, Kruger, & Medvec, 2002; Gilovich,
Medvec, & Savitsky, 2000). The spotlight effect reflects individuals’ tendency “to
believe that the social spotlight shines more brightly on them than it really does”
(Gilovich et al., 2000, p. 211). Given that the spotlight effect reflects an increased sense
of being seen, we applied this paradigm. In a study involving two independent samples,
we asked participants to imagine that they were wearing a T-shirt displaying a wellknown public figure (the pop singer Florian Silbereisen or the terrorist Osama bin
EMPIRICAL PAPERS 39
Table 1. Overview of published studies involving subtle cues of being watched
Manuscript
Eyes Cue Condition
Control Condition
Criterion
Bateson et al. (2006)
Photos of human eyes
Photos of flowers
Honest payments
Bateson et al. (2013)
“signs featuring a large pair
No Eyes
Littering behavior
Photo of flowers
Disapproval of moral
of staring eyes” (p. 2)
Bourrat et al. (2011)
Photo of neutral human eyes
transgressions
Burnham & Hare
Actual present robot with
No present robot
Contributions in repeated public
(2007)
large eyes
Carbon & Hesslinger
Photo of angry human eyes
Photo of flowers
Pro- and antisocial scenarios
Photo of neutral human eyes
Photo of flowers
Charitable donations in a
goods games
(2011)
Ekström (2011)
supermarket
Ernest-Jones et al.
Photos of human eyes
Photos of flowers
Littering behavior in a university
(2010)
Fehr & Schneider
cafeteria
Ornate eyes and nose
Neutral graphic
Trust game
Francey &
Photos of neutral human
Photos of flowers
Littering behavior at a bus stop
Bergmuller (2012)
eyes
Haley & Fessler
Ornate eyes and nose
PC desktop background
Dictator game
(2010)
(2005)
Keller & Pfattheicher
with laboratory emblem
Neutral-looking Eyes
Blank (no eyes)
Proenvironmental donations
Angry-looking eyes
White desktop
Dictator game
(2011)
Mifune et al. (2010)
background
Nettle et al. (2012)
Female and male eyes
Emblem of the institute
Dictator game
(2 conditions)
Nettle et al. (2013)
Photo of angry human eyes
No eyes
Bicycle theft
Oda et al. (2011)
Ornate eyes and nose
Blank (no eyes)
Single dictator game
Panagopoulos (2014,
Photos of human eyes
American flag, palm tree
Voting behavior
in press)
Powell et al. (2012)
(2 control conditions)
Neutral-looking eyes
Three geometric stars
Charitable donations in a
supermarket
Raihani & Bshary
Photo of neutral human eyes
(2012)
Photo of flowers, black
Dictator game
square, blank (3 control
conditions)
Rigdon et al. (2009)
Three dots resembling
eyes:
Sparks & Barclay
. .
.
Three dots not
Dictator game
.
resembling eyes: . .
Ornate eyes and nose
Landscape image
Dictator game
Ornate eyes and nose
Butterfly
Dictator game
(2013)
Tane & Takezawa
(2011)
EMPIRICAL PAPERS 40
Laden) and to estimate how many people would notice their T-shirt. Results revealed
that under conditions of watching eyes (vs. no watching eyes) participants gave higher
estimations regarding the number of people who would notice the T-shirt (i.e., an
increased sense of being seen). The findings speak to the notion that subtle cues of being
watched do induce an increased sense of being seen.20
3.2.1 Discussion
The question remains of how exactly subtle cues of being watched operate. For instance,
it is possible that subtle cues of being watched increase arousal which in turn influences
a sense of being seen, because arousal is associated with evaluative social presence
(Dickerson & Kemeny, 2004). Indeed, evaluative social presence was even used to
induce arousal, that is, the psychological state of stress (Kirschbaum et al., 1993). To test
the assumption that subtle cues of being watched operate via arousal, one could vary the
presence of a subtle cue of being watched while tracking individuals arousal on-line for
instance via electroencephalography (Klimesch, 1999).
From a motivational perspective, subtle cues of being watched can induce the motivation
to influence others’ perception of oneself (i.e., impression management; Leary, 1995;
Leary & Kowalski, 1990). In other words, subtle cues of being watched activate
individuals’ motivation to present themselves in a positive way, that is, to gain a valuable
reputation (Barclay, 2013; Oda, Niwa, Honma, & Hiraishi, 2011). This can be achieved
by means of socially desirable actions such as prosocial behavior. These considerations
are well in line with the empirical findings that show that subtle cues of being watched
increase prosocial behavior (Nettle et al., 2013; Sparks & Barclay, 2013).21 These
notions also fit with the concepts of public self-awareness (Gervais & Norenzayan, 2012;
Mifune, Hashimoto, & Yamagishi, 2010) and reputational concerns (Emler, 1990),
which reflect a state in which one’s attention is directed to how one appears in the eyes
20
To establish transparency, I want to mention that this paragraph includes sentences taken from
Pfattheicher and Keller (2014, Appendix B).
21
These considerations are further implemented in costly signaling theory (Grafen, 1990; Zahavi, 1975)
which is discussed in the next chapter (3.3). They also correspond to the notion of the ought-self
implemented in regulatory focus theory which is discussed in Chapter 3.4.
EMPIRICAL PAPERS 41
of others (Fenigstein, Scheier, & Busss, 1975; Fenigstein & Vanable, 1992). Given that
humans often present themselves in a valuable way when others are present and are
motivated to gain a valuable reputation (Griskevicius, Tybur, & Den Bergh, 2010; van
Vugt, Roberts, & Hardy, 2007), it seems reasonable to assume that subtle cues of being
watched activate the reputation system. That is, subtle cues of being watched increase
individuals’ motivation to present themselves in a socially desirable, reputationincreasing way (Nettle et al., 2013; but see the discussion below regarding this point).
Recently, Bateson et al. (2013) attended to the question of whether subtle cues of being
watched universally increase prosocial behavior, or whether they induce conformity to
social norms. Often, these two notions fall together given the social norm to behave
prosocially (e.g., Nettle et al., 2013). Bateson et al. (2013) suggested disentangling the
two notions and proposed testing subtle cues of being watched in a situation where there
is a social norm to behave antisocially. They tried to implement this by investigating
subtle cues of being watched in a situation where the descriptive norm (i.e., what others
do) is to litter the environment, specifically, when the ground was littered with leaflets
(Cialdini, Reno, & Kallgren, 1990). Bateson et al. (2013) showed that subtle cues of
being watched increased prosocial behavior (not littering the environment); but they
found no effect for a pronounced descriptive norm under watching eyes. From this
finding one can conclude that subtle cues of being watched do not, at least in this case,
induce conformity to descriptive social norms. The findings of Bateson et al.’s (2013)
study make sense from the perspective of the focus theory of normative conduct
(Cialdini et al., 1990). This theory proposes that individuals align their behavior to
present social norms in a given situation (i.e., descriptive and/or injunctive norms). From
this theoretical perspective, it is possible that Bateson et al. (2013) did not fully
disentangle prosocial behavior and social norms because individuals still may focus on
the present injunctive norm (what one ought to do, that is, not littering). Thus, the
injunctive norm was not eliminated in the study, and individuals may still have focused
on the injunctive norm, especially when being watched. Focusing on the injunctive norm
when being watched may, therefore, run against the effect suggested by Bateson et al.
(2013).
EMPIRICAL PAPERS 42
From my perspective, it is likely that individuals align their behavior to norms,
especially to injunctive norms, when a subtle cue of being watched is present. This seems
reasonable because only injunctive norms and not descriptive norms include the notion
of social disapproval (i.e., negative sanctions or punishment by others) when the
injunctive norm is not fulfilled (Cialdini et al., 1990), and social disapproval is only
possible when others are present, that is, under conditions of subtle cues of being
watched. Accordingly, it seems reasonable to assume that injunctive norms are likely to
be pronounced under subtle cues of being watched given that the violation of injunctive
norms are particularly disapproved of when noticed by present others.
Turning to the more fine-tuned cognitive level, Barclay (2013) suggested that subtle cues
of being watched operate on an unconscious rather than on a conscious level. This
assumption has implications for measuring the actual psychological state that is
produced by subtle cues of being watched. To clarify this issue, one could ask
participants explicit questions about their current state, for instance via items from the
situational public self-awareness scale (Govern & Marsch, 2001; a sample item reads
“Right now, I am concerned about the way I present myself”). However, if subtle cues of
being watched operate on an unconscious level, participants may not be able to give an
explicit account of their current inner state (Nisbett & Wilson, 1977). These
considerations highlight that one needs to assess current psychological states using
indirect methods. For instance, one could let individuals present themselves in a more
positive, reputation-increasing way which should be pronounced when a subtle cue of
being watched is presented. Specifically, one could use a social desirability scale (e.g.,
Crowne & Marlowe, 1960) and experimentally vary the presence of a subtle cue of being
watched. If subtle cues of being watched lead to increased impression management, one
should find increased socially desirable responding when a subtle cue of being watched
is presented (vs. not presented). This approach could contribute knowledge to the field of
subtle cues of being watched in that it would test not only whether subtle cues of being
watched increase prosocial and decrease antisocial behavior, but also that reputational
concerns are actually induced by subtle cues of being watched.
EMPIRICAL PAPERS 43
These considerations lead to a critical point in our studies in which subtle cues of being
watched were applied (this also holds true for the study reported in the following chapter
in which we investigated the effect of subtle cues of being watched on costly
punishment). Specifically, the validation study does not provide direct evidence that the
psychological state of reputational concerns is activated by subtle cues of being watched.
We were able to show that subtle cues of being watched lead to an increased sense of
being seen. It is reasonable to assume that reputational concerns are activated in turn,
because individuals often present themselves in a valuable way and are motivated to gain
a valuable reputation when others are present (Griskevicius et al., 2010; van Vugt et al.,
2007; see also costly signaling theory in Chapter 3.3.3). However, I want to
acknowledge that the validation study provides no direct evidence for the assumption
that subtle cues of being watched render reputational concerns salient whereas it
provides good evidence that subtle cues of being watched lead to an increased sense of
being seen, on which basis one can reasonably assume that reputational concerns are
activated.
3.2.2 Methodological remarks
It is important to mention that in research on watching eyes the question of how to
adequately operationalize the control group arises. Some researchers have argued for
using non-social symbols such as pictures of flowers as a control condition (e.g.,
Bateson, Nettle, & Roberts, 2006). This approach, however, has some significant
drawbacks. For instance, Raihani and Bshary (2012) show that pictures of flowers
increase prosocial behavior as compared with a more neutral condition in which no cues
or symbols were presented. Based on these results, I argue that a condition in which no
cues or symbols could activate certain concepts is the most valid control condition. This
approach was universally applied in the studies presented in this dissertation.22
Another methodological issue is the question of what the stylized eyes should look like.
It is reasonable to assume that angry eyes (e.g., used by Haley and Fessler, 2005) lead to
22
To establish transparency, I want to mention that this paragraph includes sentences taken from
Pfattheicher and Keller (2014, Appendix B).
EMPIRICAL PAPERS 44
a different cognitive state as compared with friendly eyes (e.g., used by Burnham and
Hare, 2007). Specifically, one can assume that individuals engage in prosocial behavior
when being watched by angry eyes in order to avoid negative sanctions (e.g., social
disapproval or punishment) whereas one can assume that individuals engage in prosocial
behavior when being watched by friendly eyes in order to gain positive sanctions (e.g.,
social approval or reward). In all studies presented here we used neutral-looking eye in
order to deal with this concern.
Moreover, one could ask the question of under which conditions subtle cues of being
watched are able to operate. Powell, Roberts, and Nettle (2012) offer a precondition.
They showed that a subtle cue of being watched increased donations to a charity
organization in a supermarket only if the supermarket was poorly crowded. It makes
sense that when social presence is already given subtle cues of being watched cannot, on
top of this, increase social presence.
Another idea is that social presence (and subtle cues of being watched) lead to prosocial
behavior only if accountability for one’s behavior is given (cf. van Bommel, van
Prooijen, Elffers, & van Lange, in press). Social presence is indeed given in classic
studies on the bystander effect which show that the more people who are present the less
people help a suffering other (Fischer et al., 2011; Latané & Nida, 1981). The bystander
effect is, inter alia, explained by the diffusion of responsibility and reduced
accountability of one’s behavior (Garcia, Weaver, Moskowitz, & Darley, 2002; Weesie,
1993). In contrast to the studies concerning the bystander effect, activated social
presence via a subtle cue of being watched is not activated by the presence of several
people (and eyes). It seems reasonable to assume (van Bommel et al., in press, put this
argument forward) that subtle cues of being watched do not reduce one’s level of
accountability. When facing a decision (e.g., in a dictator game) it should be clear that
the decider is responsible for the decision. The decision cannot be passed on to other
present individuals as there are none. Thus, it seems unlikely that subtle cues of being
watched reduce one’s accountability and that they can be used to replicate the bystander
effect.
EMPIRICAL PAPERS 45
Another interesting finding in terms of methodology comes from Sparks and Barclay
(2013) who show that prolonged exposure to a subtle cue of being watched did not
increase prosocial behavior in a dictator game. It is unclear, however, why this result
emerged (e.g., because the eyes are consciously processed and individuals more
explicitly regulate their behavior). Nonetheless, Sparks and Barclay (2013) emphasize
that the prolonged presentation of subtle cues of being watched may not result in the
effects typically found with these cues (increased prosocial behavior and decreased
antisocial behavior). This indicates that a precondition for subtle cues of being watched
to work is relatively short exposure.
In sum, subtle cues of being watched seem to be a valid approach to investigating social
presence and, accordingly, reputational concerns. This conclusion is worth noting
because research on watching eyes has assumed but not empirically tested the hypothesis
that a sense of being seen is induced by subtle cues of being watched (e.g., Izuma, 2012;
Nettle et al., 2013). Therefore, the presented work was indeed necessary in that it has
addressed a remarkable gap in this field of research. In the next paper presented, a subtle
cue of being watched is applied to test whether individuals are more or less likely to
engage in costly punishment when reputational concerns are activated.
3.3 Punishing for the sake of reputation? Effects of reputational concerns on costly
punishment (Pfattheicher, Landhäußer, & Keller, 2014, Appendix C)
As mentioned above, some scholars have suggested that individuals strategically costly
punish in order to gain a valuable reputation and to benefit from that good reputation
thereafter (e.g., Barclay, 2006; Dos Santos et al., 2011; Fehr & Gächter, 2002; Gächter &
Herrmann, 2009; Hagen & Hammerstein, 2006; Hilbe & Traulsen, 2012; Kurzban et al.
2007; Nelissen, 2008; Piazza & Bering, 2008; Shaw & Santons, 2012; Sylwester et al.,
2013; Tennie, 2012). As outlined in the introduction (see Chapter 2.2.3), this strategic
perspective on costly punishment implies that costly punishment is executed to
maximize benefits, that is, costly punishment is considered to be part of a benefitmaximizing strategy. If individuals strategically punish to gain a beneficial good
reputation which serves them in future interactions (Milinski, Semmann, & Krambeck,
EMPIRICAL PAPERS 46
2002; Nowak & Sigmund, 1998; Roberts, 1998; van Vugt et al., 2007; Wedekind &
Braithwaite, 2002) they should be more likely to engage in costly punishment when
reputational concerns are activated. In the empirical studies outlined in detail below we
experimentally tested this strategic perspective.
In the manuscript we made use of costly signaling theory (Grafen, 1990; Zahavi, 1975)
to provide a theoretical basis for the empirical studies. The theoretical framework of
costly signaling theory suggests that specific types of costly behavior are likely to
emerge in conditions where the actor can gain a valuable reputation through engagement
in these behaviors (Griskevicius et al., 2010; van Vugt et al., 2007). Accordingly, these
types of costly behavior should be more likely to emerge when reputational concerns are
activated (vs. not activated). Thus, costly signaling theory constitutes a meaningful
general theoretical framework regarding the relation of reputational concerns and costly
punishment (costly signaling theory is discussed in Chapter 3.3.3).23
Remarkably, to the best of my knowledge, there is only one serious empirical work
(Kurzban et al., 2007) that has tested the notion of whether individuals costly punish for
reputational concerns.24 However, there are several articles that consider the relation
23
To establish transparency, I want to mention that this paragraph includes sentences taken from
Pfattheicher et al. (2014, Appendix C).
24
Kurzban et al. (2007) reported on audience effects on costly punishment in a third party punishment
scenario and found increased costly punishment under conditions where an experimenter and other
participants were present (vs. an explicit statement towards anonymity in the control condition). However,
the studies have several shortcomings such as the possibility of demand characteristics (Orne, 1962),
Pygmalion effects (Rosenthal & Jacobson, 1968), the lack of a neutral control condition (explicitly
inducing anonymity in the “control” condition), and confounding costly punishment and the money
participants kept, that is, participants behaved more egoistically in the anonymous condition (punishing
less, keeping more) which is in line with several empirical studies (Hoffman, McCabe, & Smith, 1996;
Milinski, Semmann, Krambeck, & Marotzke, 2006; Rege & Telle, 2004). The work of Kurzban et al.
(2007) in comparison with our approach is further discussed in the general discussion of Pfattheicher et al.
(2014, Appendix C). To complete the list, Piazza & Bering (2008) also manipulated the degree of
anonymity in three conditions but did not find any significant differences (even with collapsed groups all
ps were > .24). Notably, this study lacks statistical power to detect even medium effects (three conditions,
total N = 42, power was below 68% to detect an existing effect (p < .05) with medium effect size, f = .25).
EMPIRICAL PAPERS 47
between reputation and costly punishment theoretically (e.g., Fehr & Gächter, 2002;
Gächter & Herrmann, 2009; Hilbe & Traulsen, 2012; Shaw & Santons, 2012; Sylwester
et al., 2013; Tennie, 2012), test how costly punishment is perceived, that is, whether
costly punishers are ascribed a good reputation (Barclay, 2006; Nelissen, 2008), and
make assumptions regarding the relation for computer simulations (Dos Santos et al.,
2011; Hilbe & Traulsen, 2012). The absence of valid empirical work is even more
noteworthy given that costly punishment and reputation both represent two essential
concepts in evolutionary psychology and biology. Therefore, investigating the relation of
reputation and costly punishment addresses a remarkable gap in this field of study.
3.3.2 The empirical studies
In the empirical paper in which we tested whether reputational concerns increase or
decrease costly punishment, we first introduced and validated a public self-awareness
manipulation to render reputational concerns salient. Specifically, we asked participants
in a validation study to describe the self that they should possess in the eyes of others (cf.
the method section in Pfattheicher et al., 2014, Appendix C). Reputation refers to the
judgments of relevant others about one’s personal qualities and how one behaves (Emler,
1990). Reputational concerns, therefore, refer to concerns regarding how others think
about one’s personal qualities. Accordingly, reputational concerns correspond to the
psychological concept of public self-awareness (Gervais & Norenzayan, 2012; Mifune et
al., 2010) which reflects a state in which one’s attention is directed to how one appears in
the eyes of others (Fenigstein et al., 1975; Fenigstein & Vanable, 1992). Therefore, we
manipulated reputational concerns and tested whether the manipulation was valid in that
it affected responses on the situational public self-awareness scale by Govern and
Marsch (2001). Using structural equation modeling, we were able to show increased
public self-awareness when participants described the self that they should possess in the
eyes of others as compared with a neutral control condition. Importantly, the
manipulation significantly affected situational public self-awareness but not private selfawareness and awareness of surroundings. The findings of the validation study speak to
EMPIRICAL PAPERS 48
the applicability of the manipulation which was then implemented in two consecutive
studies.25
In these studies, we applied a public goods game under conditions of anonymity so as to
minimalize default reputational concerns. We also precluded mutual interactions (i.e., we
used one-shot games) between participants to minimize default strategic concerns when
engaging in costly punishment. As outlined in detail in Chapter 1.4, if strategic
considerations had not been minimized (i.e., if it was possible that costly punishment
was strategically executed) the likelihood that we would have been able to have observed
an effect in terms of reputation on top of already strategically executed costly
punishment would have been fairly low. To put it another way, we could not have
investigated the central research question of whether costly punishment is executed to
strategically gain a valuable reputation.
We manipulated reputational concerns as outlined above. In two studies using a
hypothetical public goods game and a public goods game involving real interactions (via
z-Tree; Fischbacher, 2007) and real monetary rewards, we were able to document that
costly punishment is reduced when reputational concerns were rendered salient. No
effects were found regarding the contribution to the public good (cf. the discussion
below regarding this finding).
In an additional study we replicated these results by showing that costly punishment is
also reduced when reputational concerns are rendered salient by means of a subtle cue of
being watched (as discussed in the last chapter and in Pfattheicher and Keller, 2014,
Appendix B). In sum, in the three studies using different manipulations of reputational
concerns we showed that costly punishment is actually reduced under conditions of
reputational concerns.
The question remains of why exactly individuals reduce costly punishment under
reputational concerns conditions. In the empirical paper, we built on findings that
25
To establish transparency, I want to mention that this paragraph includes sentences taken from
Pfattheicher et al. (2014, Appendix C).
EMPIRICAL PAPERS 49
antisocial behavior is typically reduced under social presence and when individuals are
motivated to gain a valuable reputation, that is, when reputational concerns are activated
(Andreoni & Bernheim, 2009; Ernest-Jones, Nettle, & Bateson, 2010; Haley & Fessler,
2005; Mifune et al., 2010; Milinski et al., 2002; Milinski et al., 2006; Oda et al., 2011).
We argued that costly punishment reflects negative reciprocity (cf. Bowles & Gintis,
2003; Brandstätter & Königstein, 2001; Reuben & van Winden, 2008) and revenge (de
Quervain et al., 2004; Gollwitzer & Bushman, 2012; McCullough, 2008) and, building
on the considerations outlined above (cf. Pfattheicher & Keller, in press, Appendix A),
one can add reactive aggression (Crockett et al., 2010; Osumi et al., 2012; Stirrat &
Perrett, 2010). I argue that negative reciprocity, revenge, and reactive aggression reflect
antisocial tendencies which are, as reviewed above, less likely to be shown when
reputational concerns are activated. This implies that potential punishers hold beliefs
about how costly punishment is socially construed by others and they shape their
behavior accordingly when others are relevant (i.e., when reputational concerns are
activated). The findings of the three studies presented do suggest that costly punishment
is considered to be a reputation reducing, socially undesirable tendency by the potential
punisher.
A fourth study was conducted to examine this notion directly. We tested whether
individuals consider engagement in negative reciprocity and revenge socially
undesirable. If costly punishment and the corresponding tendencies of negative
reciprocity and revenge are considered to be reputation reducing, socially undesirable
tendencies, we should have found modest support for negative reciprocity and revenge
with significantly lower approval for negative reciprocity and revenge when reputational
concerns were activated. This was revealed to be the case (cf. Study 4 in Pfattheicher et
al., 2014, Appendix C).26
26
To establish transparency, I want to mention that this paragraph includes sentences taken from
Pfattheicher et al. (2014, Appendix C).
EMPIRICAL PAPERS 50
3.3.1 Discussion
The findings obtained strongly speak against the strategic perspective of costly
punishment. Costly punishment does not seem to be strategically executed to gain
reputational benefits. The findings are important because they contradict the assumptions
made by several scholars (Barclay, 2006; Dos Santos et al., 2011; Fehr & Gächter, 2002;
Gächter & Herrmann, 2009; Hagen & Hammerstein, 2006; Hilbe & Traulsen, 2012;
Kurzban et al. 2007; Nelissen, 2008; Piazza & Bering, 2008; Shaw & Santons, 2012;
Sylwester et al., 2013; Tennie, 2012) and should, therefore, be taken into account in
future (especially evolutionary-focused) theorizing involving the interplay of costly
punishment and reputation.
Our findings are in line with research done by Rockenbach and Milinski (2011) who told
their participants that their decision to punish could be observed by other subjects, and
then gave their participants the possibility of hiding the amount they invested in costly
punishment. They found that the probability of individuals hiding their punishment
increased the more money was spent on costly punishment. Rockenbach and Milinski
(2011) interpreted this result as evidence that individuals fear losing their reputations
when engaging in costly punishment. That is, Rockenbach and Milinski (2011) argued
that individuals’ reputational concerns are activated when they are observed and engage
in costly punishment. As such, their argument is in line with ours. However, Rockenbach
and Milinski (2011) did not manipulate the psychological state of reputational concern.
That is, their conclusions rely on correlational evidence, and their explanations were
made ex post. As such, our findings are in line with the findings of Rockenbach and
Milinski (2011) but extend them in a meaningful way in that we directly show that the
manipulation of reputational concerns decreased costly punishment.
I do not claim that costly punishment cannot have benefits for punishing individuals. For
instance, in line with the conceptualization of costly punishment as reactive aggression,
costly punishment can benefit an individual in that costly punishment serves as an
instrumental means of achieving the goal of harming an uncooperative interaction
partner. Reaching a goal can indeed satisfy a need (e.g., Higgins, 2012a), that is, benefit
EMPIRICAL PAPERS 51
an individual. However, I do criticize the assumption that costly punishment is
strategically executed to gain reputational benefits, thereby questioning the
considerations of the aforementioned scholars (Barclay, 2006; Dos Santos et al., 2011;
Fehr & Gächter, 2002; Gächter & Herrmann, 2009; Hagen & Hammerstein, 2006; Hilbe
& Traulsen, 2012; Kurzban et al. 2007; Nelissen, 2008; Piazza & Bering, 2008; Shaw &
Santons, 2012; Sylwester et al., 2013; Tennie, 2012) and how they specifically
conceptualize the strategic nature of costly punishment.
In all studies, no significant effects were found regarding the contribution to the public
good. At first glance, this finding seems to be a non-replication of several studies that
have shown that activating reputational concerns leads to an increased contribution to the
public good (e.g., Burnham & Hare, 2007; Milinski et al., 2002; Rege & Telle, 2004).
However, all studies exclusively investigated the effects of reputational concerns on the
contribution to the public good in the absence of costly punishment. This is worth noting
because the presence of the option to costly punish others substantially changes the
nature and amount of contributions to the public good in that individuals contribute
significantly more to the public good if an option to costly punish is present (Fehr &
Gächter, 2002). That is to say, when an option to costly punish is available, the
contribution to the public good is far from being purely prosocial behavior and is often
executed strategically to avoid punishment (Berg et al., 1995). I do not deny that some
contributions to the public good are prosocial when costly punishment is possible.
However, when an option to costly punish is present, strategic contributions executed to
avoid being punished increase substantially so that one cannot distinguish between “real”
prosocial contributions and “strategic” non-prosocial contributions. In such a situation,
individuals cannot reliably signal prosocial motives. From the perspective of costly
signaling theory it is important that the behavior that may function as a signal must
reliably signal the underlying characteristics or motives of an individual (Bliege Bird,
Smith, & Bird, 2001; Griskevicius et al., 2010; Smith & Bliege Bird, 2000).27 This is not
27
Nonetheless, one could argue that costly punishment of uncooperative individuals signals prosocial
tendencies because costly punishment is positively correlated with contributions to the public good (e.g.,
Pfattheicher & Keller, 2013, Appendix D). However, costly punishment of uncooperative individuals is
EMPIRICAL PAPERS 52
sufficiently the case when strategic non-prosocial contributions cannot be differentiated
from real prosocial contributions. Therefore, the contributions to the public good when
an option to punish is available should not increase when reputational concerns are
activated, which is in line with our results.28
Another argument points in this direction. If reputational concerns should increase
contributions to the public good, neglecting the argument above for a moment, this effect
must be present on top of the high level of cooperation that already exists when the
option of costly punishment is present (Fehr & Gächter, 2002). That is, if contributions
are already high, the likelihood that an additional factor further increases the
contributions decreases (i.e., a ceiling effect; see, e.g., Rigdon, Ishii, Watabe, &
Kitayama, 2009, for a similar argument). That is to say, only a strong effect could run
against a ceiling effect. Essentially, these two arguments speak to the notion that it was
unlikely that we would have observed increased contributions to the public good when
reputational concerns were activated.
3.3.2 Further remarks
I have argued that it is unlikely that the ultimate motives in costly punishing are
prosocial. Rather, I have argued that antisocial tendencies drive costly punishment. I
want to note that I do not argue that costly punishment and the corresponding concepts
of negative reciprocity, revenge, and reactive aggression do not have prosocial
consequences. There is a lot of research that shows that negative reciprocity, revenge,
and reactive aggression can lead to valued outcomes such as cooperation in groups,
status, and a state of peace through deterrence (e.g., Banny, Heilbron, Ames, & Prinstein,
2011; Fehr & Gächter, 2002; Griskevicius et al., 2009; Hawley & Vaughn, 2003). Using
the terms of Batson and colleagues (Batson et al., 2008; Batson et al., 2002), I take the
perspective that the prosocial effects of costly punishment reflect unintended
consequences whereas the ultimate goal (the actual motive) reflects revenge and
also positively correlated with antisocial punishment (e.g., Pfattheicher et al., in press, Appendix E). Thus,
based on this logic, one could equally argue that costly punishment may signal antisocial tendencies.
28
To establish transparency, I want to mention that this paragraph includes sentences taken from
Pfattheicher et al. (2014, Appendix C).
EMPIRICAL PAPERS 53
antisocial tendencies (representing tendencies that are likely to reduce the actor’s social
reputation).
Moreover, I do not argue that individuals generally disapprove of the punishment of
norm deviants. The legal punishment of criminals, for instance, is well in line with
individuals’ beliefs and values (Carlsmith & Darley, 2008; Carlsmith, Darley, Robinson,
2002). In this case, the punishment is executed by legitimized institutions. However, the
work presented is about individuals’ costly punishment in public good games, executed
by one’s own hand. Costly punishment executed by one’s own hand is fundamentally
different from the punishment of criminals by legitimized institutions. Indeed, research
in the field of justice has shown that individuals typically consider self-administered
justice, that is, the punishment of a potential criminal executed by one’s own hand,
socially undesirable (Raney, 2009). Of course, there are substantial differences between
the punishment of a potential criminal and costly punishment of an uncooperative
individual in a public goods game. These considerations emphasize that it is important to
take into account who is punishing and whether a punishing entity is legitimized (this
issue is considered again below).
3.3.3 Discussing costly signaling theory
As outlined above, the theoretical framework of costly signaling theory suggests that
specific types of costly behavior are likely to emerge in conditions where the actor can
gain a valuable reputation through engagement in these behaviors (Griskevicius et al.,
2010; van Lange et al, 2013; van Vugt et al., 2007). Costly signaling theory was
originally developed in order to explain characteristics that seem to be costly rather than
beneficial at first sight. The classic example is the peacock’s tail (Petrie, Halliday, &
Sanders, 1991; Zahavi & Zahavi, 1997). The peacock’s tail is a remarkable handicap in
terms of fitness at first sight as it is more costly (to grow and when escaping predators;
Zahavi, 1975) than beneficial. However, the notion emerges that the peacock’s tail
signals underlying phenotypic qualities. Specifically, a peacock with an impressive tail
survives although it is such a fitness handicap. On this basis, one can infer that the
peacock must have good phenotypic qualities that can compensate for the costs.
EMPIRICAL PAPERS 54
Crucially, peacocks possessing the most costly and impressive tails gain benefits in the
future because they are more likely to be chosen as a mating partner (i.e., reproductive
success, see Petrie et al., 1991). Thus, costly signaling theory explains why apparently
unreasonable, costly characteristics and behavior persist: costly signal provides benefits
in the future.29
It is worth noting that costly signaling theory is not an exclusive evolutionary approach
in terms of sexual selection and fitness benefits (e.g., Bliege Bird et al., 2001; Hamilton
& Zuk, 1982). Spence (1973), for instance, applied the logic of costly signaling to the
job market. Basically, high compared with low quality applicants for a job have fewer
difficulties to gain good education credentials due to their higher abilities. The idea in
the approach of Spence (1973) is that job applicants signal their underlying working
qualities through good education credentials because especially high quality applicants
will gain good education credentials. Thus, high quality applicants can be distinguished
from low quality applicants on the basis of education credentials which, as a
consequence, represent a signal of high working qualities. Given that costly signaling
theory is not an exclusive evolutionary approach and costly signaling takes place in
domains other than sexual selection, our work did not aim to contribute to the literature
on evolutionary psychology. Rather, we used costly signaling theory to build a
theoretical basis for the assumption that socially desirable, prosocial behavior is likely to
emerge under conditions of reputational concerns (Griskevicius et al., 2010; van Vugt et
al., 2007).
In critically discussing costly signaling theory, one has to acknowledge that postulations
that a specific characteristic may function as a costly signal are often made ex post. This
is, for instance, the case for the peacock’s tail (Petrie et al., 1991; Zahavi & Zahavi,
1997) where the emergence of the tail is explained ex post by the fitness benefit it
provides. Costly signaling theory cannot explain why other birds and animals in general
do not possess such an impressive tail. That is, costly signaling theory cannot explain
29
To establish transparency, I want to mention that this paragraph includes sentences taken from
Pfattheicher et al. (2014, Appendix C).
EMPIRICAL PAPERS 55
why specific behavior exists but only why a specific characteristic persists (because of
benefits). These considerations also correspond to the fundamental axiom that all
explanations are causal explanations, and that behavior cannot be explained by its’
(beneficiary) consequences (Elster, 2007). Of course, beneficial consequences strengthen
the appearance of a behavior or characteristic in future (i.e., classic positive
reinforcement). However, beneficial consequences cannot explain the appearance of a
behavior or characteristic in the first place, because at this point no beneficial
consequences had occurred. Explaining behavior by its consequences is, however, often
done in evolutionary psychology (Elster, 2007; West, El Mouden, & Gardner, 2011).
The work presented made use of costly signaling theory because this theory proposes
that prosocial behavior, which typically results in a good reputation as outlined above, is
more likely to be shown when reputational concerns are activated. In contrast, costly
signaling theory proposes that antisocial behavior, which typically results in a bad
reputation, is less likely to be shown when reputational concerns are activated. Building
on these considerations, it makes sense that costly punishment, which I linked to the
antisocial tendencies of negative reciprocity, revenge, and reactive aggression, is reduced
when reputational concerns are activated. The purpose of applying costly signaling
theory was not to explain the existence or persistence of costly punishment per se.
Rather, costly signaling theory was used to theoretically integrate the concepts of
reputational concerns and costly punishment.
On the basis of costly signaling theory, Barclay (2006) and Nelissen (2008) argue that
costly punishment can function as a costly signal, signaling prosocial tendencies. Indeed,
Nelissen (2008) showed that strong costly punishers gained reputational benefits from
punishment: they were perceived as fair, friendly, and generous by third parties.
Additionally, they received more trust in a standard trust game as compared with nonpunishers (Berg et al., 1995). Congruently, Barclay (2006) reported that costly punishers
(vs. non-punishers) are perceived to be more trustworthy, group focused, and worthy of
respect. He also showed that costly punishers benefited in a subsequent trust game. This
research actually shows that costly punishment can have reputational benefits (cf.
EMPIRICAL PAPERS 56
Barclay, 2013). However, receiving (reputational) benefits is only one side of the coin.
Sundie et al. (2011; see also Bliege Bird et al., 2001; Griskevicius et al., 2007)
emphasized that behavior that is said to be a costly signal needs also be signaled by an
actor under conditions where the signal can work, that is, under social presence. If costly
punishment is considered a reputation-increasing tendency by individuals, costly
punishment should increase when reputational concerns are activated. However, as
outlined above, we have shown reduced costly punishment under reputational concerns
conditions.
3.3.4 Explaining actor-observer discrepancies
Obviously, there is a discrepancy between the actors’ construal of the engagement in
costly punishment as compared with the construal by observers. The folk conceptual
theory of Malle and colleagues (Knobe & Malle, 2002; Malle & Knobe, 1997, 2001;
Malle, Knobe, & Nelson, 2007) provides a theoretical framework to explain this issue. It
is described in detail below in order to establish a theoretical grounding for explaining
actor-observer discrepancies regarding costly punishment.
Folk conceptual theory was originally developed to explain actor-observer discrepancies,
as shown in experiments on the fundamental attribution error (Jones & Harris, 1967;
Ross, Amabile, & Steinmetz, 1977). The fundamental attribution error reflects the idea
that observers tend to explain behavior with an emphasis on personal factors whereas
actors tend to explain behavior with an emphasis on situational factors. 30 Malle and
colleagues (Malle et al., 2007, see also Knobe & Malle, 2002) were able to show that the
distinction between personal and situational attributions regarding actors and observers is
not crucial. Instead, they postulate (and empirically show) three other dimensions in
which actors and observers differ which were finally implemented in folk conceptual
theory. First, actors tend to explain behavior by reasons, because actors have direct
access (although subjectively and potentially biased) to the reasons and motives of their
behavior. Also, actors tend to see themselves as a meaningful, reasonably acting actor
30
Later, it was shown in a meta-analysis (Malle, 2006) that this attribution holds, if ever, only for negative
events that happen to an actor.
EMPIRICAL PAPERS 57
and are motivated to present themselves in such a way. Observers, on the other hand,
tend to explain behavior by describing the actor as relatively less rational, as acting less
consciously, and, speaking in the terms used by Malle and colleagues, observers use
rather a causal history of reasons to explain behavior. Second, actors tend to explain
behavior through beliefs whereas observers tend to explain behavior through desires.
Third, actors are less likely to use marked belief reasons (e.g., “I work so much because I
think it is good to earn some money”; using “I think” is relatively uncommon, cf. Malle
et al., 2007) whereas observers do indeed use marked belief reasons (e.g., “Dan works so
much because he wants to earn some money”; using “he wants” is relatively usual).31
The first aspect of folk conceptual theory in particular is relevant in the context of actorobserver discrepancies regarding costly punishment. That is, Malle and colleagues
explain actor-observer differences by arguing that actors in particular take their motives,
reasons, and the decision process into account when evaluating their behavior, given that
they have cognitive access to their motives, their reasons, and the decision process
underlying their actions. Therefore, it seems reasonable to assume that potential
punishers do indeed take their anger (Fehr & Gächter, 2002; Pillutla & Murnighan, 1996;
cf. Chapter 2.2.1) and revenge motives (de Quervain et al., 2004; Gollwitzer &
Bushman, 2012; McCullough, 2008) into account when judging how they should behave
appropriately (i.e., when reputational concerns are activated). Folk conceptual theory
emphasizes that observers do not have direct access to the decision process of an actor
and therefore have to rely on other inferences to which they have access. These are
causal history of reason explanations such as traits or cultural norms (Malle, 1999; Malle
31
This third dimension, emphasizing linguistic differences between actors’ and observers’ explanations,
actually explains why the fundamental attribution error is apparently empirically found (e.g., in Jones &
Harris, 1967). Raters of observer statements such as “He took the umbrella because he thinks it is raining”
are more likely categorized as a “person attribution” (due to the use of “he thinks”) whereas raters of actor
statements categorize the same statement with only the source varying (now the actor) more likely to the
situation (“I took the umbrella because it is raining”). The actual explanation (“it is raining”) is given in
both the actors’ and observers’ explanations. So, actually, actors and observers do not differ in their
explanation in this example, but their explanations were categorized either as “person attribution” or
“situation attribution”.
EMPIRICAL PAPERS 58
et al., 2007). Thus, it seems reasonable to assume that observers explain costly
punishment with reference to a fairness norm and ascribe fairness motives and prosocial
traits to costly punishers, which is in line with the findings of Barclay (2006) and
Nelissen (2008).
Another reasonable explanation for the actor-observer difference regarding costly
punishment can be found in the notion of emotion appropriateness (d’Arms & Jacobson,
2000; Aune, 1997; Aune, Aune, & Buller, 1994). This notion implies that an expressed
emotion should be appropriate in interpersonal situations, that is, when others are
present. Building on this notion, one can assume that individuals who experienced anger
as a reaction to a fairness norm violation do not consider their anger to be appropriate or
its expression socially desirable, which may be especially relevant in situations in which
they are concerned about their reputation and present others. Thus, from this perspective,
one can argue that costly punishment (i.e., the expression of anger) is not considered to
be an appropriate response, which may be particularly relevant when reputational
concerns are activated. As a result, individuals are less likely to engage in costly
punishment when reputational concerns are rendered salient. In contrast, perceivers of
costly punishment do not directly observe (at least in the experiments of Barclay, 2006,
and Nelissen, 2008) the anger of the costly punishers. Thus, they may not be able to
consider anger when making a judgment about an individual who engaged in costly
punishment. Reconsidering folk conceptual theory (Knobe & Malle, 2002; Malle &
Knobe, 1997, 2001; Malle et al., 2007), observers must rely on other inferences. In sum,
folk conceptual theory and the notion of emotion appropriateness suggest that it is
important to differentiate between the actor’s and the observer’s perspective and to
explain actor-observer discrepancies in costly punishment.
Another line of research may be able to explain why the pattern of reduced costly
punishment under conditions of reputational concerns emerges. Specifically, social
psychological research on deindividuation suggests that individuals in a state of
decreased self-evaluation and increased anonymity (i.e., deindividuation; Festinger,
Pepitone, & Newcomb, 1952) are more likely to engage in disinhibited, aggressive, and
EMPIRICAL PAPERS 59
antisocial behavior (Diener, Fraser, Beaman, & Kelem, 1976; Diener, Lusk, DeFour, &
Flax, 1980). In his theory of deindividuation, Zimbardo (1969) explained these effects,
inter alia, by a reduced concern for one’s evaluation by others. He wrote: “If others can’t
identify or single you out, they can’t evaluate, criticize, judge, or punish you” (p. 255).
Empirically, Rogers and Ketchen (1979) have shown that when individuals were aroused
they were more likely to engage in aggressive behavior (using the Buss, 1961,
aggression machine to give electroshocks for false answers in a learning experiment), but
only if they were de-individuated (participants wore white lab coats and hoods vs. casual
clothes). These considerations fit our reasoning on costly punishment and reputational
concerns. Specifically, it may be possible that in the control, as compared with the
reputational concerns condition, individuals experience an increased sense of anonymity
and behave in a less inhibited way whereas in the reputational concerns condition
individuals are more inhibited (for a similar argument, see Bushman and Huesmann,
2010). As a result, individuals in the control condition are more likely to follow their
emotions and engage in costly punishment whereas participants in the reputational
concerns condition are more likely to show inhibitory tendencies and are less likely to
engage in costly punishment (which is in line with our work on behavioral inhibition, as
presented in the following paper, see Chapter 3.4 below and Pfattheicher and Keller,
2013, Appendix D).
One must note that the explanations offered by deindividuation, emotion appropriateness
as well as the application of folk conceptual theory are made ex post. A direct, ex ante,
empirical test is therefore needed. For instance, one could directly investigate the
explanation of costly punishment by actors versus observers of costly punishment and
apply the dimensions proposed by Malle et al. (2007). Using this strategy, one could
shed light on the distinct evaluations of costly punishment by actors and observers. In
sum, it seems a valuable approach for future research to examine actor-observer
discrepancies in costly punishment.
It still remains an open question as to why exactly reputational concerns reduced costly
punishment. As outlined above, it could be the case that individuals consider their
EMPIRICAL PAPERS 60
emotional state, that is, the expression of anger, as inappropriate and not socially
desirable. It is also possible that costly punishment itself is not considered appropriate or
socially desirable or that individuals do not feel legitimized to engage in costly
punishment, particularly when they are observed. Moreover, research on deindividuation
suggests that when individuals act without the activation of what others think (i.e.,
without the activation of reputational concerns), they are more likely to do what they
want (i.e., follow their emotions and punish uncooperative others).
Another question is which specific others one needs to be concerned about regarding
one’s reputation. That is to say, the specific audience seems of importance when it comes
to costly punishment. Froming, Walker, and Lopyan (1982), for instance, showed that
participants are less likely to attack another individual (using the Buss aggression
machine) when the electroshocks had to be executed in the presence of an audience (vs.
no audience) that was against the use of electroshocks. The reverse was found when the
audience was in favor of electroshocks, that is, the electroshocks were more frequently
used when the electroshocks were socially desirable (vs. no audience). Congruently, one
could imagine that when an audience is in favor of costly punishment, it will be more
likely to emerge (vs. no audience). In the studies presented we did not include this
boundary condition as this was beyond the scope of the research. Our research question
was more basic, that is, whether “general” reputational concerns increase or decrease
costly punishment. In this way we could target the general claim by several scholars that
costly punishment is strategically executed to gain reputational benefits. As outlined
above, we were able to refute this claim.
I do not deny that costly punishment can have some reputational benefits. Rather, I
suggest that the hypothesis that costly punishment is generally used to gain reputational
benefits needs to be reformulated. In line with Griskevicius et al. (2009), it could be that
costly punishment is sometimes used to impress others which corresponds to the idea of
“aggress to impress.” Investigating the boundary condition of the relation between costly
punishment and reputational concerns appears to be an interesting approach for future
research.
EMPIRICAL PAPERS 61
In sum, there are obviously some questions that could not be answered in the studies
presented, but could (and should) be targeted in future research. Nonetheless, the work
presented represents a new and promising approach to the study of costly punishment
and reputational concerns. Our findings demonstrate that the activation of reputational
concerns reduced costly punishment. The findings are important because they contradict
the suggestions of several scholars (e.g., Barclay, 2006; Fehr & Gächter, 2002; Gächter
& Herrmann, 2009; Hagen & Hammerstein, 2006; Kurzban et al. 2007; Nelissen, 2008)
in that the findings strongly speak against the strategic perspective: costly punishment
does not seem to be strategically executed to gain reputational benefits. The work
presented should therefore be taken into account in future (especially evolutionaryfocused) theorizing involving the interplay of costly punishment and reputation.
3.4 Vigilant self-regulation and costly punishment in public goods situations
(Pfattheicher & Keller, 2013, Appendix D)
The last paper presented here (see Chapter 3.3 above) documents that costly punishment
is reduced under reputational concerns conditions. In the following work (Pfattheicher &
Keller, 2013, Appendix D) it is shown that chronically prevention-focused individuals,
that is, individuals who possess a strong ought-self (which one could argue corresponds
to the notion of reputational concerns, see below), are more likely to engage in costly
punishment as compared with relatively weakly prevention-focused individuals.
Before discussing these counter-intuitive findings, I will integrate and discuss regulatory
focus theory with respect to other motivational theories (beyond what is already said in
Pfattheicher & Keller, 2013, Appendix D). That is, the following section will outline in
detail the distinctiveness and specificity of regulatory focus theory in comparison with
other motivational theories. From this basis, then, the prediction that the vigilant
orientation of regulatory focus theory (i.e., the prevention focus) is positively related to
costly punishment, but not vigilant orientations as conceptualized in other motivational
theories, will become clearer. This is why several motivational theories are outlined in
the following section.
EMPIRICAL PAPERS 62
3.4.1 Theories of motivation and regulatory focus theory
Self-regulation is the operation by individuals to (consciously or unconsciously) modify
and adjust their own habits or actual state to bring these into alignment with a positive
standard (Vohs & Baumeister, 2004; Vohs & Schmeichel, 2003). Self-regulation refers
to the general hedonic principle that individuals are motivated to approach pleasure and
to avoid pain and that individuals regulate their current state according to this principle.
This notion is implemented in several motivational theories, going back to Kurt Lewin
(1943). Lewin (among others, e.g., Fishbein & Ajzen, 1975; Rotter, 1954) can be
considered to be a theorist postulating that individuals do not generally strive for positive
end-states but depending on Value × Expectancy considerations. That is to say,
individuals are more likely to be motivated to achieve a specific goal (i.e., regulate
toward a desired end-state) the higher they value it and if there is a sufficient expectation
that the goal will be reached. If expectancies are low, however, individuals’ motivational
energy and drive (to use the terms of early motivational theorists such as Hull, 1943,
1952) are reduced even if the goal is considered highly valuable.
This theoretical approach was further developed by Atkinson and colleagues (Atkinson,
1957, 1964; McClelland, Atkinson, Clark, & Lowell, 1953) who added specific
motivational orientations to the Value × Expectancy approach, resulting in the model of
Value × Expectancy × Motive. Atkinson (1957) focused on the field of task performance
and postulated two different motivational orientations that are relevant in goal striving.
The first one is the basal motivation of achieving and approaching success (termed “hope
for success”). The relevant standards of this motivational orientation are “achievement,
affiliation, power” (Atkinson, 1957, p. 360). The second orientation is the basal
motivation of avoiding failure (termed “fear of failure”). The relevant standard of this
motivational orientation is to minimize pain. According to Atkinson’s model, approach
motivation and corresponding actions of approaching a target or standard are elicited if
hope for success is situationally stronger than fear of failure, and pride is experienced
when a goal is successfully approached. Conversely, avoidance motivation and
corresponding actions of avoiding a target or standard are elicited if fear of failure is
EMPIRICAL PAPERS 63
situationally stronger than hope for success, and shame and humiliation are experienced
when a goal is not reached.
The core of Atkinson’s theory of motivation is that individuals generally approach
pleasure (i.e., hope for success) or avoid pain (i.e., fear of failure), a notion also found in
other motivational theories postulating two systems of self-regulation. In Carver and
Scheier’s self-discrepancy theory (Carver, 2006; Carver, Lawrence, & Scheier, 1999;
Carver & Scheier, 1981, 1990, 2011) motivation is understood as movement toward
pleasure and away from pain. The movement aspect was also emphasized above when
defining the term self-regulation. If one’s current state is compared to a positively valued
reference point, discrepancy reducing approach motivation is elicited; if one’s current
state is compared to a negatively valued end-state, discrepancy increasing avoidance
motivation is elicited.
Gray’s widely applied reinforcement sensitivity theory (RST; Gray, 1970, 1982) of
personality also deserves a mention. Gray’s approach focuses on two orthogonal
personality dimensions (cf. Corr, 2009): the first dimension is the behavioral inhibition
system (BIS), which is sensitive to conditioned signals of punishment, non-reward
(frustration), and (less strongly) novelty. A strong BIS should be reflected in a greater
proneness to anxiety, fear, frustration, and the use of vigilant strategies in goal striving.
The motivational aspect of the BIS, that is, how this system regulates movements toward
goals, is that the BIS inhibits movements toward negative outcomes in order to avoid
punishment and/or non-rewards, that is, an avoidance motivation. The second dimension
is the behavioral activation system (BAS), which is sensitive to conditioned appetitive
stimuli and is activated by signals of non-punishment and reward. A strong BAS should
be reflected in the experience of positive affect. The motivational aspect of the BAS is
that it initiates movements toward positive outcomes (e.g., reward and/or nonpunishment), that is, an approach motivation. Moreover, Gray developed a third system,
the fight-flight system (FFS), which is sensitive to unconditioned aversive stimuli and
mediates the emotions of panic and rage, but was not implemented as a personality factor
EMPIRICAL PAPERS 64
(cf. Carver & White, 1994). As in the other motivational two system models, individuals
self-regulate to approach pleasure and to avoid pain.
In an advancement of classic motivational theories, Elliot and colleagues (Elliot, 1999;
Elliott & Dweck, 1988; Elliot & McGregor, 1999, 2001; Moller & Elliot, 2006) went
beyond the simple notion that positive end-states are approached and negative end-states
are avoided. Elliot and colleagues emphasized that it is necessary to differentiate
between specific standards that can either be approached or avoided. Specifically, it is
proposed that individuals can possess mastery approach goals, which relates to the
development of competence through task mastery; individuals can also possess
performance approach goals, which relates to the demonstration of competence relative
to other individuals (Elliot & McGregor, 2001). Congruently, it is proposed that
individuals can have mastery avoidance goals, which relates to preventing failures in
terms of task mastery; individuals can also possess performance avoidance goals, which
relates to not performing worse than other individuals (Elliot & McGregor, 2001). So,
the specific standard of what is approached or avoided is fundamental to this theory. The
standard of mastery is related to absolute, intrapersonal standards of what is ideally being
approached or ideally being avoided. Performances goals are related to normative
standards of what ought to be approached or ought to be avoided in comparison to
others.32
The theories outlined above share the idea that the vigilant avoidance system regulates
movements away from negative outcomes (i.e., “pain” in a general sense). That is,
individuals in a vigilant avoidance mode use avoidance and inhibitory tendencies and
behaviors to reach the goal of avoiding negative outcomes. In contrast to all of the
32
Other theoretical models make similar assumptions and postulate two systems of motivation,
emphasizing different standards and regulatory strategies (e.g., locomotion and assessment concerns when
moving to and evaluating behavioral options, see Kruglanski et al., 2002; or individualistic concerns that
regulate the self in an independent way in distinction to others, and collectivist concerns that regulate the
self in a way dependent on others; Markus & Kitayama, 1991). These models emphasize different
standards and regulatory strategies; the core, however, remains the same as in the models of Atkinson,
Gray, Carver, and Elliot: approaching pleasure and avoiding pain.
EMPIRICAL PAPERS 65
motivational theories outlined, regulatory focus theory (RFT; Higgins, 1997, 1998, 2012;
Scholer & Higgins, 2008, 2011) goes beyond the notion that the vigilant avoidance
system is exclusively related to avoidance and inhibitory tendencies. RFT proposes that
the vigilant avoidance system can also entail approach strategies and the active initiation
of behavior. Specifically, RFT emphasizes that it is necessary to differentiate between
specific standards as well as specific preferred strategies of how positively evaluated
standards are approached and how negatively evaluated standards are avoided. These are
fundamental and important departures from all of the other theories mentioned that
generally postulate that positive end-states are approached and negative end-states are
avoided.
RFT proposes two distinct regulatory systems – a prevention-focused orientation and a
promotion-focused orientation (Higgins, 1997, 1998, 2012; Scholer & Higgins, 2008,
2011). The input factors (i.e., valued standards or reference points) of the preventionfocused orientation are safety and security needs. Individuals possessing a prevention
focus are oriented toward significant others, that is, they are concerned with fulfilling
oughts and responsibilities that reflect normative standards. Moreover, preventionfocused individuals are motivated to avoid losses and to approach non-losses. If a goal is
reached,
prevention-focused
individuals
experience
quiescence/calmness-related
emotions whereas if a goal is missed, prevention-focused individuals experience
agitation/anxiety-related emotions (Molden, Lee, & Higgins 2008; Higgins, 1997).
The input factors of the promotion-focused orientation are growth, advancement, and
accomplishment. Individuals possessing a promotion focus are oriented toward ideals,
wishes, and aspirations. Promotion-focused individuals are, moreover, motivated to
avoid non-gains and to approach gains. If a goal is reached, promotion-focused
individuals experience cheerfulness/happiness-related emotions whereas if a goal is
missed romotion-focused individuals experience dejection/sadness-related emotions
(Molden et al., 2008; Higgins, 1997).33
33
Of note, promotion focus represents a distinct construct and is not the opposite dimension to prevention
focus. An intuitive parallel are the “Big Five” personality dimensions in which, for instance, neuroticism is
EMPIRICAL PAPERS 66
In a recent advancement of RFT, three independent levels of self-regulation have been
proposed: the system, the strategic, and the tactical level (Higgins, 2012b; Scholer &
Higgins, 2008, 2011). On the system level, prevention focus and promotion focus are
orthogonal to the classic approach-avoidance distinction, because prevention focus
includes approaching safety, security, and non-losses but also avoiding insecurity and
losses. Congruently, promotion focus includes approaching growth, advancement, and
gains but also avoiding the absence of these standards.
On the strategic level, prevention and promotion focus differ in the preferred strategies
of how the respective standards are approached and avoided. In a prevention-focused
mode, individuals approach a specific end-state (e.g., security) in a vigilant and rather
defensive manner. That is, prevention-focused individuals typically prefer avoidance
strategies and are concerned with the presence or absence of negative outcomes and
information (Higgins, 2012b; Scholer & Higgins, 2008, 2011). In a promotion-focused
mode, individuals approach specific end-states (e.g., accomplishment) in an eager
manner. That is, promotion-focused individuals typically prefer approach strategies and
are concerned with the presence or absence of positive outcomes and information
(Higgins, 2012b; Scholer & Higgins, 2008, 2011).
The tactical level reflects a more concrete level and refers to which behavioral patterns
prevention- and promotion-focused individuals typically use in particular situations. As
with the system and the strategic level, the tactical level is independent from the other
two levels. Thus, it is possible that an individual is motivated to avoid pain on the system
level (e.g., generally avoiding nosocomial infections), use a vigilant avoidance strategy
(e.g., being careful and vigilant regarding infections across different situations) but use
approach tactics to reach the goal (e.g., approaching hand hygiene, participating in
workshops on how to prevent nosocomial infections). Typically, avoidance strategies
tend to fit avoidance tactics. For instance, Crowe and Higgins (1997) show in a signal
clearly distinct from contentiousness (Zillig, Hemenover, & Dienstbier, 2002). When investigating
neuroticism, contentiousness is rather neglected (e.g., Scheier, Carver, & Bridges, 1994). Congruently,
promotion focus is neglected in this work.
EMPIRICAL PAPERS 67
detection task that prevention-focused individuals have a defensive response bias, that is,
they vigilantly use an avoidance tactic to insure correct rejections and insure against
errors of commission whereas promotion-focused individuals have a risky response bias,
that is, they eagerly use an approach tactic to ensure correct hits and insure against errors
of omission. If prevention-focused individuals constantly use avoidance tactics one
should find that prevention-focused individuals always show a defensive and inhibitory
conservative bias, for instance in the domain of risk behavior. However, Scholer, Zou,
Fujita, Stroessner, and Higgins (2010) have shown that prevention-focused individuals
show a risky bias in a state of a loss, that is, they adopt an approach tactic to prevent a
negative event.
The previous section emphasized the distinctiveness of RFT in comparison with other
motivational theories. Only the vigilant orientation in RFT explicitly entails the notion
that approach strategies are used to reach a goal, that is, it is in line with RFT that
prevention-focused individuals take action and show disinhibited tendencies (rather than
merely executing defensive and inhibitory tendencies). This should be particularly true
when it comes to the negative event of a norm violation, given that prevention-focused
individuals are concerned with negative events as well as oughts and responsibilities
which reflect normative standards. That is to say, it is reasonable to assume that strongly
prevention-focused individuals are likely to engage in costly punishment. The paper
presented in the following section was devoted to examining the relation of the
prevention focus and costly punishment. The distinctiveness of RFT is also one
important part of the paper; specifically, prevention focus was theoretically and
empirically contrasted to general inhibitory tendencies. The empirical paper is discussed
in detail below.
3.4.2 The empirical paper
As outlined above, prevention-focused individuals are particularly sensitive to negative
outcomes and information; they are, moreover, concerned with oughts and
responsibilities, reflecting normative standards. To look again at the literature on costly
punishment, I have outlined that costly punishment is a response to a negative event (i.e.,
EMPIRICAL PAPERS 68
uncooperative behavior resulting in the violation of a fairness norm). Due to the
sensitivity to negative outcomes involving normative standards, it seems reasonable to
assume that prevention-focused individuals react particularly strongly to such a negative
outcome and are prone to punish those identified as the cause for the negative outcome.
This is shown in Pfattheicher and Keller (2013, Appendix D); the disposition of
prevention-focused self-regulation was significantly positively related to costly
punishment. This finding fits the theoretical considerations discussed above. On a
strategic level, prevention focus is related to vigilant avoidance strategies. However, as
outlined, this does not mean that prevention-focused individuals always use avoidant
tactics, as reflected in behavioral inhibition or inaction (Gable, Reis, & Elliot, 2000;
Scholer & Higgins, 2008). That is exactly what we can document; strongly preventionfocused individuals were more likely to take action and punish as compared with weakly
prevention-focused individuals.
Scholer and Higgins (2008) wrote: “It is common to associate approach and avoidance at
the behavioral level with action and inaction, respectively” (p. 491). If the assumption
holds true that an avoidance tactic should lead to behavioral inhibition or inaction, that
is, the omission of costly punishment, we should be able to document a negative relation,
or at least a non-relation, between behavioral inhibition and costly punishment. Indeed,
behavioral inhibition, as measured with Carver and White’s (1994) BIS scale, had a
close to zero-order relation to costly punishment; when prevention and promotion were
controlled for, a negative relation was found (although this relation did not reach
conventional levels of significance, B = -.15, p = .14). These findings support our
arguments that prevention focus is related to a special concern with normative standards
and entails regulatory tactics to take action. This is not the case for the BIS as
conceptualized in Gray’s theory (1970, 1982) and Carver and White’s (1994) approach
(cf. Pfattheicher and Keller, 2013, Appendix D).
In sum, the empirical paper (Pfattheicher & Keller, 2013, Appendix D) made three
important contributions to the scientific literature. First, we were able to differentiate
prevention focus from behavioral inhibition on the tactical level which is well in line
EMPIRICAL PAPERS 69
with recent developments of RFT (Higgins, 2012a, 2012b; Scholer & Higgins, 2008,
2011). Second, our approach to examine the association between basal motivational
orientations (specifically, prevention-focused self-regulatory orientation and behavioral
inhibition) and costly punishment addresses a gap in the field of costly punishment in
social dilemma situations. Third, from the perspective of individual differences, the
presented work is the first study that relates costly punishment in a public goods game to
individual differences. Interpersonal differences regarding cooperative behavior in social
dilemma situations have previously been repeatedly shown (e.g., Hilbig, Zettler, &
Heydasch, 2012; Hirsh & Peterson, 2009; Kurzban & Houser, 2001). The work
presented here, however, is the first study to empirically relate costly punishment in
public goods games to individual differences.
3.4.3 Integrating reputational concerns and oughts
As outlined, reputation refers to judgments of relevant others about ones’ personal
qualities and how one behaves (Emler, 1990), and reputational concerns refer to the
concerns regarding how others think about one’s personal qualities and behavior.
Reputational concerns therefore include how one thinks one ought to behave, that is,
one’s perceived duties, responsibilities, and social expectations as set by others. In other
words, one can argue that reputational concerns correspond to the ought-self. In
Pfattheicher et al. (2014, Appendix C; cf. Chapter 3.3 above) it is documented that costly
punishment was reduced when the ought-self was activated, that is, when reputational
concerns were rendered salient. In contrast, prevention-focused individuals, that is,
individuals who possess a strong ought-self, are more likely to engage in costly
punishment compared with relatively weakly prevention-focused individuals. How can
these apparently inconsistent findings be integrated?
I postulate that the ought-system in prevention-focused individuals needs to be
differentiated. Specifically, I argue that prevention-focused individuals possess an oughtfocus that is relevant for the self, but they also possess an ought-focus that is relevant for
others. The ought-self-focus is activated in situations when it is relevant how “oneself”
behaves in the eyes of others. These are situations in which an individual should fulfill
EMPIRICAL PAPERS 70
social expectations, duties, and responsibilities as is the case when reputational concerns
are rendered salient. The ought-other-focus is activated in situations when it is relevant
how others behave in one’s own eyes. These are situations in which others should fulfill
social norms.
I propose that the ought-other-focus is activated when other group members behave
uncooperatively, that is, when others violate a fairness norm. In contrast, the ought-focus
that is relevant for the self is activated in situations when it is relevant how one behaves
in the eyes of others, that is, when reputational concerns are rendered salient. As
documented, when the ought-self-focus is activated through rendering reputational
concerns salient, different effects can be found than one would expect when the oughtother-focus is activated (i.e., reduced costly punishment when the ought-self-focus is
activated vs. increased costly punishment when the ought-other-focus is activated).
Of course, the differentiation of the ought-foci needs direct empirical validation. For
instance, one could experimentally test the proposed differentiation by activating the
ought-self-focus in one experimental group (e.g., activating one’s focus on how oneself
ought to behave, for instance through activation of reputational concerns), the oughtother-focus in another (e.g., through activating one’s focus on how others ought to
behave), and comparing these two conditions to a neutral control condition. Assuming
that the uncooperative behavior of another individual should already activate the oughtother-focus (also in the neutral control group), one should not necessarily find a
difference between the control condition and the condition in which the ought-otherfocus is rendered salient. In line with our studies (Pfattheicher et al., 2014, Appendix C),
however, one should find reduced costly punishment when the ought-self-focus is
rendered salient.
3.4.4 Additional theoretical remarks
One has to acknowledge that the empirical paper presented remains silent regarding the
causal nature of the prevention-costly punishment relation. Several different methods
have been proposed to manipulate prevention focus such as to activate current security
concerns (Friedman & Förster, 2001), or current ought-self concerns (Scholer et al.,
EMPIRICAL PAPERS 71
2010), or by simultaneously activating security and ought-self concerns (Higgins et al.,
2001). Indeed, it seems a valuable approach to experimentally manipulate prevention
focus and to use a causal-analytic approach. However, one must note that it appears
important which dimension of prevention focus is manipulated. In line with the
considerations regarding the differentiation of the ought-foci (see above), a
differentiation of prevention focus could also be considered. In specific domains, an
ought-self manipulation may not result in the same outcomes as a manipulation
rendering security concerns salient. This may hold true because different regulatory
tactics are applied depending on the activated dimension. For instance, it seems
reasonable that rendering security concerns salient results in a defensive, risk-averse
tactic (Friedman & Förster, 2001) whereas an ought-self manipulation results in active
impression management.
As a consequence one could ask whether it is actually meaningful to consider prevention
focus to reflect one construct. The evaluative dimensions of whether prevention focus
should be considered to be one construct could be (a) the parsimoniousness of a theory
and (b) the ability of a theory to make predictions. Emphasizing the parsimoniousness of
a theory, one should reject the differentiation of prevention focus. Neglecting the
parsimoniousness of a theory and moving toward a theory of middle range or in an even
more concrete direction (Merton, 1968), one should consider prevention focus to be one
construct with different dimensions. This is not the case in current conceptualizations of
RFT; prevention focus (and promotion focus) is considered to be one construct.
Taking the other evaluative dimension into account, that is, the ability of a theory to
make predictions, it seems reasonable to differentiate the prevention focus. This claim is
supported by the present findings that ought-self concerns reduced costly punishment
whereas ought-other concerns probably relate positively to costly punishment, and thus
move in exactly the opposite direction. Congruently, one can expect distinct effects if
one renders security concerns or ought concerns salient, because different regulatory
tactics are likely to be executed (see above). That is, without a differentiating prevention
focus precise predictions are not possible.
EMPIRICAL PAPERS 72
In this regard, a comparative model could be the conceptualization of “social class” or
socioeconomic status (SES), defined as the rank of an individual in comparison to others
in a society with regard to wealth, occupation, and education (Adler, Epel, Castellazzo,
& Ickovics, 2000). Piff, Stancato, Côté, Mendoza-Denton, and Keltner (2012) suggest
that upper-class individuals, relative to lower-class individuals, behave more unethically.
Specifically, they have shown that individuals driving upper-class (vs. lower-class)
vehicles in terms of make, age, and appearance are more likely to cut off other vehicles
and pedestrians. Moreover, they have shown that individuals’ subjective SES was
negatively correlated with tendencies toward a variety of unethical decisions, and it was
positively correlated with positive attitudes toward greed and lying. Piff et al. (2012)
focused on specific dimensions of social class, that is, individuals’ subjective belief of
being superior to others (as measured by individuals’ subjective SES) and individuals’
(economic) ability to drive so-called “upper-class vehicles”.
This surely neglects certain dimensions of social class, for instance one’s objective
educational achievement or one’s subjective job prestige.34 Interestingly, Krause (2011)
has shown a small negative correlation between individuals’ objective educational
achievement and the antisocial attitude of hostility. Likewise, Hessing, Elffers, Robben,
and Webley (1993)
documented that
individuals
who fraudulently claimed
unemployment benefits were less educated than a honest control group. These findings
suggest that one dimension of social class (i.e., education) may in fact be negatively
related to antisocial behavior. This comparative example emphasizes that in order to
make precise predictions about a general construct one has to distinguish its specific
dimensions. Otherwise, one would make false theoretical conclusions, for instance as
done by Piff et al. (2012). They concluded that “upper-class individuals behave more
unethically than lower-class individuals” (p. 4086) which has to be refuted in light of the
findings by Krause (2011). A solution would be to differentiate the concept of social
34
It is another question whether subjective and objective class constitute one dimension, as Piff et al.
(2012) indirectly assumed; already Karl Marx (1847/2009) emphasized the importance of this
differentiation when he spoke about a class “in itself” (i.e., objective class, or common standard in terms of
means of production) as opposed to a class “for itself” (i.e., subjective class, or class consciousness).
EMPIRICAL PAPERS 73
class (as typically done in the field of sociology, for instance by Bourdieu, 1979) and to
outline that the different dimensions of social class may be differentially related to
unethical behavior.
Likewise, one can propose a general prevention focus dimension including specific
domains such as security concerns and ought concerns with a differentiation of the latter
into ought-self and ought-other concerns. This claim is in line with Karl Popper’s idea of
falsifying (instead of verifying) theories (Popper, 2002). If a general prevention focus
dimension cannot predict outcomes whereas a prevention focus including specific
domains can, the general prevention focus dimension should be rejected in favor of a
prevention focus including specific domains.
3.4.5 Integrating findings on cortisol and prevention focus
I have shown that endogenous cortisol is negatively related to costly punishment whereas
prevention-focused self-regulation is positively related to costly punishment. By
integrating these findings and considerations one can argue that prevention focus should
only be positively related to costly punishment when cortisol levels are relatively low,
because high cortisol levels are related to behavioral withdrawal and inhibition (Blair et
al., 2004; Brown et al., 1996; Nava et al., 2006; Tops & Boksem, 2011; van Honk et al.,
2010; Windle, 1994). Consequently, one can assume that cortisol may inhibit the
pathway between prevention focus and costly punishment.
A re-analysis of a data set in which cortisol and prevention focus were measured (N =
72) indeed revealed a significant interaction between endogenous cortisol and prevention
focus (p < .01). Simple slope analyses showed that prevention was only significantly
positively related to costly punishment when basal cortisol was relatively low (1 SD
below the mean; p < .001), but not when basal cortisol was relatively high (1 SD above
the mean; p = .46). These findings strengthened the notion that cortisol is indeed related
to behavioral withdrawal and inhibition and they suggest that endogenous cortisol may
indeed inhibit the pathway between prevention focus and costly punishment.
In sum, the empirical work presented documents that self-regulatory orientations play a
crucial role in guiding individuals’ thoughts and behavior in social dilemma situations.
EMPIRICAL PAPERS 74
Considering the case of costly punishment in particular, I emphasized that it is necessary
to elaborate on self-regulatory orientations in order to reach a comprehensive
understanding of costly punishment. Therefore, it seems important to take individuals’
self-guide into account, that is, whether individuals act in accordance to the social
expectations of others (the proposed ought-self-focus) or whether individuals act in
accordance to the social expectations of oneself (the proposed ought-other-focus). To
conclude, the work presented addressed a gap in the literature on costly punishment and
individual differences as well as basal motivational orientations. Notably, we were able
to add valuable empirical evidence regarding the distinctiveness of prevention focus in
comparison with another vigilant self-regulatory orientation. As such, the work presented
represents a valuable contribution to the literature on costly punishment, individual
differences, and basal motivational orientations.
So far, the present dissertation has focused on costly punishment of uncooperative
individuals in public goods situations. Interestingly, a specific destructive behavior
emerges in public goods situations when a system of costly punishment is implemented:
antisocial punishment, that is, costly punishing cooperative individuals (cf. Herrmann et
al., 2008). The “dark side” of costly punishment is the focus of the following chapter.
3.5 Individual differences in antisocial punishment in public goods situations: The
interplay of cortisol with testosterone and dominance (Pfattheicher, Landhäußer,
& Keller, in press, Appendix E)
When reviewing the literature on antisocial punishment, it can be seen that only a few
empirical papers have investigated this interesting behavioral phenomenon, that is, the
investigation of antisocial punishment has been almost completely neglected in social
dilemma research thus far (cf. Dreber & Rand, 2012; Herrmann et al., 2008). Published
empirical papers exclusively focus on the boundary conditions of antisocial punishment.
For instance, Herrmann et al. (2008) showed that antisocial punishment is more likely to
emerge in societies with relatively weak norms of civic cooperation and an established
and functioning judiciary (see also Gächter & Herrmann, 2009). Lamba and Mace (2011)
showed that antisocial punishment is more likely to emerge when an outgroup member
EMPIRICAL PAPERS 75
(as compared with an ingroup member) can be punished (for similar findings see also
Bernhard, Fischbacher, & Fehr, 2006, and Goette, Huffman, Meier, & Sutter, 2012).
Building on these findings Sylwester et al. (2013) argue that antisocial punishment of
outgroup members is used to increase one’s own group’s status through the devaluation
of an outgroup.
A
biopsychological
perspective
on
antisocial
punishment
that
investigates
endocrinological and psychological factors is completely lacking. Our empirical work
(Pfattheicher et al., in press, Appendix E) has examined antisocial punishment from a
biopsychological perspective and has, therefore, addressed a remarkable gap in this line
of research. In our paper we have argued that antisocial punishment in a typical public
goods game is in line with destructive dominant behavior, which we define as a
propensity toward forceful, competitive, and aggressive behavior to attain social rank
(Cheng, Tracy, Foulsham, Kingstone, & Henrich, 2013; Mazur & Booth, 1998).
Basically, in a typical public goods game as applied in the present work, punishing other
group members has the effect that one’s own payoff is reduced by a specific amount of
money and the punished individual’s payoff is reduced by a factor of three. So,
individuals can increase their rank by destructively punishing others (because the costs
of punishment are greater for the punished than for the punisher). That is to say,
punishing cooperative individuals reflects an aggressive act that allows an increase in
rank.35
These considerations exactly fit those of Sylwester et al. (2013) who also emphasized the
aggressive component in antisocial punishment and the possibility of increasing one’s
rank in comparison to others. These authors wrote: “This suggests that some instances of
costly punishment, in particular antisocial punishment, may function as aggressive acts,
and are not contingent on the previous cooperative behavior of the punished individuals.
In sum, in apparently irrational costly antisocial behavior, the decisions to punish are, at
least in some studies, logically tied to the effectiveness of such punishment and to the
35
To establish transparency, I want to mention that this paragraph includes sentences taken from
Pfattheicher et al. (in press, Appendix E).
EMPIRICAL PAPERS 76
ability to increase the positive difference between others’ payoffs and one’s own” (p.
172, emphases added).
Given that antisocial punishment corresponds to the basic notion of dominance, we
proposed that antisocial punishment should be related to psychological dominance and
the dominance-related hormone testosterone (Mehta et al., 2008; Sellers, Mehl, &
Josephs, 2007). Additionally, we applied the dual-hormone perspective according to
which testosterone is related to dominant and destructive behavior but only in individuals
with a low level of endogenous cortisol (Carré & Mehta, 2011; Mehta & Josephs, 2010).
Congruently, we have argued that psychological dominance (i.e., forceful, competitive,
and aggressive orientations to attain social rank) should be positively related to antisocial
punishment and, in line with the dual-hormone perspective, this relation should only hold
for individuals with a low level of endogenous cortisol. Indeed, in two studies we were
able to document that antisocial punishment was positively related to testosterone and to
psychological dominance, but only in individuals with a relatively low level of cortisol.
Cortisol again seems to inhibit a pathway, this time the pathway between testosterone
and dominance concerns about antisocial punishment.
In contrast, testosterone and psychological dominance were not significantly related to
costly punishment of uncooperative individuals, neither as a main effect nor as an
interaction involving cortisol (see the discussion below regarding this point and
Pfattheicher et al., in press, Appendix E).
As mentioned above, research on costly punishment in social dilemma situations has
almost completely neglected the case of antisocial punishment (Dreber & Rand, 2012;
Herrmann et al., 2008). As such, the work presented addressed a remarkable gap in this
line of research by showing that one has to take the endocrinological factor of
testosterone and the psychological factor of dominance as well as their interaction with
cortisol into account in order to reach a comprehensive understanding of antisocial
punishment. To conclude, antisocial punishment does indeed seem to be driven by
dominance concerns, as suggested by Sylwester and colleagues (2013). The work
presented is the first work that has empirically tested this notion.
EMPIRICAL PAPERS 77
3.5.1 Relations to other work
In the empirical paper we discussed possible underlying mechanisms of the findings and
related our work to interesting work by other researchers (cf. the General Discussion in
Pfattheicher et al., in press, Appendix E). In order to avoid redundancies, this discussion
is not repeated. There is one line of research that is not addressed in the paper.
Specifically, it has been shown that men with a relatively high level of endogenous
testosterone are more likely to reject unfair offers in an ultimatum game as compared
with men with a relatively low level of testosterone (Burnham, 2007; Mehta & Beer,
2010). That is to say, costly punishment of unfair others is positively related to
endogenous testosterone (assuming that one can treat the rejection in an ultimatum game
as costly punishment of uncooperative behavior). In contrast, we did not find a positive
association between testosterone and costly punishment of uncooperative individuals in
the public goods game. Given that testosterone is related to dominance concerns
(Eisenegger, Naef, Snozzi, Heinrichs, & Fehr, 2010), it seems possible that costly
punishment of uncooperative others in an ultimatum game but not in a public goods
game can more easily fulfill one’s need for dominance. For instance, given the direct
head-to-head constellation in the ultimatum game but not in the public goods game it
may be the case that one can more easily prevent the first player being “better off” than
oneself by destroying both payoffs, that is, by engaging in costly punishment.
The differences found in costly punishment with respect to costly punishment in a public
goods game and in an ultimatum game are grist to the mill for people emphasizing the
difference between costly punishment in an ultimatum game and in a public goods game.
Future research should investigate the different motives of these two forms of costly
punishment (cf. Chapter 1.4).
Other work has claimed to test the causal effect of testosterone on costly punishment.
Specifically, Zak et al. (2009) showed that testosterone administration is associated with
increased costly punishment in an ultimatum game in men, which was, however, not
replicated in women (Eisenegger et al., 2010). Zethraeus et al. (2009) also did not find
this effect in women after long-term testosterone administration. However, one has to be
EMPIRICAL PAPERS 78
cautious when interpreting these experimental findings. What can be said is that the
administration of testosterone can have a causal effect on ultimatum game rejections
(Zak et al., 2009). What cannot be said is that testosterone itself directly affects costly
punishment because hormones, and specifically testosterone, cross-talk on several levels
(Viau, 2002). For instance, it is shown that testosterone inhibits the cortisol releasing
HPA axis (Rubinow et al., 2005), which speaks to the notion that testosterone
administration lowers endogenous cortisol levels. As shown in our work, individuals
with a relatively low (vs. high) cortisol level are more likely to engage in costly
punishment. From this basis one could assume that cortisol levels may also drive the
effects of testosterone administrations. That is to say, testosterone administration
influences several biological pathways to behavior. In other words, testosterone
administrations are confounded in that it is not only endogenous testosterone levels that
are manipulated. What can additionally be said is that negative feedback loops following
the exogenous administration of testosterone are very possible, that is, the actual downregulation of the testosterone releasing HPG axis (Eisenegger, Haushofer, & Fehr, 2011).
As with the investigation of exogenous cortisol (cf. Chapter 3.1), the investigation of
exogenous testosterone is problematic, at least when one would like to draw causal
inferences regarding testosterone itself.
3.5.2 Proposals for future research
In line with Sylwester et al. (2013), we (Pfattheicher et al., in press, Appendix E)
proposed that individuals increase their rank by destructively punishing others because
the costs of punishment are greater for the punished than for the punisher. If antisocial
punishment in our studies is, at least in some part, driven by status and dominance
concerns, antisocial punishment should be reduced when status and dominance concerns
cannot be fulfilled through antisocial punishment. For instance, one could reduce the
effectiveness of costly punishment to a level where the amount that is invested in costly
punishment is equal to the reduction of another individuals’ payoff (cf. Shinada,
Yamagishi, & Ohmura, 2004). Therefore, one could not, at least, increase their monetary
rank, which should reduce antisocial punishment if status and dominance motives are at
work when engaging in antisocial punishment.
EMPIRICAL PAPERS 79
An intriguing approach in the investigation of antisocial punishment would be to relate
antisocial punishment to individual differences. If antisocial punishment is indeed driven
by interindividual comparisons (i.e., status and dominance concerns), individuals
possessing a strong competitive orientation, as implemented in the social value
orientation of van Lange, De Bruin, Otten, and Joireman (1997), should be likely to
engage in antisocial punishment. Another intriguing approach would be to relate
antisocial punishment to social values, as conceptualized in Schwartz’s model of basic
human values (Schwartz, & Bilsky, 1990; Schwartz et al., 2012). In the recently revised
model of basic human values (Schwartz et al., 2012), three candidates may be related to
antisocial punishment as conceptualized in this dissertation (i.e., antisocial punishment
as destructive and dominant aggression toward cooperative group members). First,
antisocial punishment should be negatively related to the prosocial human value of
“benevolence-caring” which is defined as the devotion of welfare to ingroup members.
Antisocial punishment seems to involve rather opposite tendencies in this regard.
Second, antisocial punishment should be negatively related to the prosocial human value
of “benevolence-dependability” which is defined as being reliable and trustworthy
toward ingroup members. Again, antisocial punishment seems to involve rather opposite
tendencies in this regard. Third, if antisocial punishment is executed to control other
individuals’ payoff and to increase one’s status, antisocial punishment should be
positively related to the human value of “power-resources” because this value reflects
the need for power through controlling material and social resources. Basically, one can
conclude that research on antisocial punishment would benefit from an individual
difference perspective because this approach could shed at least some light on the
motives for antisocial punishment.
Integrating research on reputational concerns and costly punishment, Sylwester et al.
(2013) proposed that antisocial punishment can function as a costly signal. Specifically,
antisocial punishment can be executed to attain a reputation for aggressiveness and to
signal social dominance. Griskevicius et al. (2009), for instance, have shown that
aggression can be used to impress others. However, as suggested in our work on
reputational concerns and costly punishment of uncooperative individuals (Pfattheicher
EMPIRICAL PAPERS 80
et al., 2014, Appendix C), costly punishment may be considered socially undesirable.
Therefore, one must consider specific situations in which antisocial punishment may not
be socially undesirable and can therefore function as a costly signal (e.g., competitive
male-to-male aggression in young men to acquire a reputation for dominance, as
documented by Hilton, Harris, & Rice, 2000).
To sum up, I have shown that testosterone and psychological dominance (given low
cortisol) relate to one interesting behavior that emerges in public goods games when
costly punishment is possible, that is, antisocial punishment. As such, the work presented
contributes to a better understanding of antisocial punishment which is important given
that research has almost completely neglected antisocial punishment thus far (Dreber &
Rand, 2012; Herrmann et al., 2008). Moreover, I have outlined that future research can
aim to answer questions regarding psychological boundary conditions, individual
differences, and the specific cognitive processes that explain antisocial punishment. The
present work can surely represent a meaningful basis in this regard.
GENERAL DISCUSSION 81
4 General discussion
In social dilemma situations, individuals face a mutually interdependent situation in that
their benefits depend on the behavior of the other individuals involved in the situation.
Social dilemmas are further characterized by conflicting interests; specifically,
maximizing one’s individual benefit stands in conflict with maximizing the benefit of the
collective. This is the specific quality that is inherent in social dilemma situations
(Camerer, 2003) and that is why Hardin (1968) speaks of a tragedy: behaving selfishly is
always the most benefitting strategy from a benefit-maximizing perspective. The
problem is also manifested in a Nash equilibrium that exists when all individuals behave
uncooperatively, because no one has any incentive to individually change their behavior.
That is to say, the quality of social dilemmas is that they are intractable, strong situations
in which individuals benefit from acting against the collective (Esser, 2000).
Another aspect of social dilemmas makes them worth investigating, that is, the quantity
of their appearance (van Lange et al., 2007; van Lange et al., 2013). Humans face social
dilemmas on several levels (Yurdusev, 1993), on the macro-level, for instance when it
comes to the issue of climate change, on the meso-level, when it comes to diverging
interests in a village concerning the water irrigation system (cf. Ostrom, 1990), and on
the micro-level, when students undertake group work and can benefit from being a freerider. The quality and the quantity argument constitute the rationale for why factors that
foster cooperation in social dilemma situations should be investigated.
Kollock (1998) has systematically outlined three different perspectives on how to solve
social dilemmas. Whereas the structural perspective and the strategic perspective
emphasize the fundamental goal of maximizing benefits (Kollock, 1998), the
motivational (psychological) perspective emphasizes that individuals not only take their
own benefit into account but also that of other individuals involved in the situation.
Batson and colleagues (Batson et al., 2008; Batson, et al., 2002) specified the
motivational perspective and claimed three general non-egoistic motivations for why
people act for the public good (collectivism, principlism, and altruism). In the present
work I have extended this view to one important behavior that increases cooperation in a
GENERAL DISCUSSION 82
remarkable way: costly punishment (Balliet et al., 2011; Fehr & Gächter, 2002).
Specifically, I have outlined that costly punishment can represent a structural solution for
establishing cooperation (cf. Fehr & Gächter, 2002). Furthermore, I have discussed the
possibility that costly punishment is strategically executed to obtain benefits.
Specifically, I have focused on claims that costly punishment is strategically executed to
improve one’s mood and to gain reputational benefits. These claims could be refuted, as
summarized below. Generally, I have emphasized the third perspective introduced by
Kollock (1998), that is, the question of why individuals are motivated, beyond strategic
and structural considerations, to punish uncooperative and cooperative others. This
motivational perspective and the corresponding empirical findings from the literature and
the studies presented are summarized in the following section.
4.1 Summarizing the findings
As outlined in the introduction, research on costly punishment in public goods situations
has documented that costly punishment of uncooperative others can be considered to be
predominantly driven by anger on the basis of a fairness norm violation by low
contributors (e.g., Cubitt et al., 2011; Fehr & Fischbacher, 2004). Costly punishment
reflects a behavior that seems to result from the impulsive system rather than the
reflective system, that is, costly punishment can be considered to be an impulsive act
(e.g., Crockett et al., 2010; Halali et al., in press). Moreover, research on costly
punishment suggests that costly punishment corresponds to the idea of reactive
aggression (e.g., Crockett et al., 2010; Osumi et al., 2012; Stirrat & Perrett, 2010).
Research on anticipated reward (Gollwitzer & Bushman, 2012) and our work on
reputation (Pfattheicher et al., 2014, Appendix C) show that costly punishment is neither
strategically executed in order to improve one’s mood nor to gain a valuable reputation.
The findings therefore question the benefit-maximizing strategic perspective. This
acquired knowledge is indeed a valuable contribution to the literature because the
findings contradict the conceptualization of costly punishment as a benefit-maximizing
strategy as proposed by several scholars in the field of costly punishment (Barclay, 2006;
Dos Santos et al., 2011; Fehr & Gächter, 2002; Gächter & Herrmann, 2009; Hagen &
GENERAL DISCUSSION 83
Hammerstein, 2006; Hilbe & Traulsen, 2012; Kurzban et al. 2007; Nelissen, 2008;
Piazza & Bering, 2008; Shaw & Santons, 2012; Sylwester et al., 2013; Tennie, 2012).
Although research on costly punishment has produced a large number of articles in
recent years, there still exist remarkable research gaps. The empirical papers I have
presented address several neglected areas in this field of research. I have shown that
endogenous cortisol is negatively related to costly punishment. From this finding one can
conclude that a relatively low level of cortisol can function as a condition for engaging in
costly punishment. This consideration is supported by the finding that strongly
prevention-focused individuals are more likely to engage in costly punishment of
uncooperative individuals as compared with weakly prevention-focused individuals, but
only when strongly prevention-focused individuals have a relatively low level of cortisol.
In distinguishing prevention focus from a general inhibitory tendency, I have
documented that only prevention focus significantly predicts costly punishment whereas
behavioral inhibition is instead negatively related to costly punishment (although not
significantly at a conventional level). Therefore, this work did not only contribute to
research on costly punishment in social dilemma situations but also to a better
understanding of prevention focus in distinction to other motivational theories.
Following this, I outlined and discussed our study showing that subtle cues of being
watched lead to an increased sense of being seen, an assumption that was made in other
research but had not been empirically tested before. Applying this subtle cue of being
watched to render reputational concerns salient, I documented that activating (vs. not
activating) reputational concerns reduced costly punishment. In this regard, I have
proposed that one has to differentiate an ought-self-focus from an ought-other-focus
because one can make different predictions depending on which focus is activated. These
considerations are also relevant for RFT because they emphasize that, when considering
the ought-self of the prevention focus, one has to specify which ought-focus is relevant
in a given situation in order to make precise predictions.
In a final step I shed light on biopsychological factors that can explain the punishment of
cooperative individuals, that is, antisocial punishment. A biopsychological analysis of
GENERAL DISCUSSION 84
this interesting behavioral phenomenon has not been done before. Showing that
testosterone and psychological dominance predict antisocial punishment given a
relatively low cortisol level provides a solid basis for future research on antisocial
punishment. In sum, the empirical papers presented have addressed several research gaps
in the field of costly punishment. Together, they have surely contributed to a better
understanding of costly punishment in public goods situations, which was the main aim
of the present dissertation. However, it should have become clear in the course of this
dissertation that there are a remarkable number of research questions still to be
considered.
In the following sections I want to discuss methodological issues that apply to the
specific paradigm applied in our studies as well as methodological issues that apply to
the biopsychological perspective.
4.2 Discussion of the applied paradigm
In the studies reported, we used the paradigm of Fehr and Gächter (2002) to investigate
costly punishment in public goods situations. This paradigm has several methodological
strengths. First, our participants behaved under anonymous conditions. This approach
ensured that individuals’ behavior was not influenced by strategic behavioral shifts due
to a lack of anonymity. Indeed, strategic behavioral shifts are commonly observed when
anonymity is taken away (e.g., Bushman & Huesmann, 2010; Diener, 1976). Second,
one-shot games were applied, that is, individuals interacted only once with another
individual. Thus, strategic considerations where individuals try to strategically shape the
behavior of others in order to gain personal benefits in future interactions were
minimized because there were no future interactions (at least within the studies). This is
particularly important given that we investigated whether costly punishment is
strategically executed in order to gain a valuable reputation; if costly punishment had
already been strategically executed (i.e., if strategic considerations had not been
minimized, for instance when mutual interactions were possible) the likelihood that we
would have been able to have observed an effect in terms of reputation on top of already
strategically executed costly punishment would have been fairly low.
GENERAL DISCUSSION 85
Moreover, by applying a reduced situation we were able to investigate costly punishment
beyond strategic considerations, thus focusing on the essence of costly punishment.
Generally speaking, our aim was to understand the basic of costly punishment beyond
strategic considerations. This was possible using the parsimonious paradigm of Fehr and
Gächter (2002).
4.3 Discussion of the biopsychological perspective
Another notable strength of the work presented is the application of a biopsychological
perspective. This approach enabled us to show that costly punishment of uncooperative
individuals can be predicted by basal levels of cortisol. Hence, we could relate costly
punishment to processes typically associated with basal levels of cortisol, that is,
impulsivity, vigilance, and behavioral withdrawal. This is worth mentioning because
individuals often have difficulties and a restricted ability to correctly report the motives
and processes underlying their behavior (Nisbett & Wilson, 1977), and may engage in
socially desirable responding (which is not the case when assessing hormone levels). Of
course, future research is needed to provide more direct empirical evidence regarding the
process (e.g., via a manipulation; cf. Jacoby & Sassenberg, 2011; Spencer et al., 2005).
Our research can constitute a meaningful basis in this regard.
By relating antisocial punishment to testosterone and dominance concerns (given a
relatively low level of cortisol) we were able to shed light on the underlying factors of
antisocial punishment. Indeed, taking interactions between hormones (cortisol,
testosterone) and psychological motivations (dominance concerns) into account
prevented us from drawing false conclusions. If we had not assessed cortisol, we could
have come to the probably false conclusion that dominance concerns and testosterone
may not play a role in antisocial punishment.
Interestingly, the relation of the prevention focus and costly punishment was moderated
by basal cortisol in that a positive relation could only be documented in individuals with
a relatively low level of cortisol. Thus, a biopsychological perspective also offers a
meaningful boundary condition (given cortisol’s positive link to behavioral withdrawal;
Blair et al., 2004; Brown et al., 1996; Nava et al., 2006; Tops & Boksem, 2011; van
GENERAL DISCUSSION 86
Honk et al., 2010; Windle, 1994) of the relation of one important basal motivational
orientation and costly punishment. In sum, a biopsychological perspective offers the
potential to draw conclusions but also to prevent false or no conclusions. That is to say, a
biopsychological perspective does indeed provide a meaningful framework to reach a
better understanding of costly punishment in public goods situations.
I want to acknowledge, nonetheless, that the biopsychological perspective also has its
shortcomings. First of all, the present research remains silent regarding the causal nature
of the observed relations and the more specific endocrinological pathway. Whereas the
specific endocrinological pathway was not the focus of the present research and requires
the knowledge of endocrinologists to reach a more fine-grained level of analysis,
showing causal relations is deeply rooted in the field of psychology. As outlined above,
it is possible to manipulate cortisol (this applies to testosterone levels, too) via
exogenous cortisol or via situational stressful demands. This approach, however, has
major setbacks given that exogenous cortisol and typically applied manipulations to
increase cortisol levels, for instance the Trier Social Stress Test (TSST), have a nonnegligible amount of confounds (cf. Chapter 3.1.1). In sum, one has to acknowledge that
the present work remains silent regarding the causal nature of the observed relations.
This should not, however, limit the valuable conclusions that could be drawn using the
existing studies.
In the following section I want to go beyond specific research findings and discuss costly
punishment in social dilemma situations from a more general perspective. Specifically, I
will address three main questions: first, given that the present work is based on studies
from the laboratory, I ask how adequate it is to investigate social dilemma situations and
costly punishment in the laboratory. The second question builds on these considerations
and asks what can be deduced from the studies presented for behavior in “real life”.
Third, given that Fehr and Gächter (2002) termed costly punishment of uncooperative
individuals “altruistic punishment” and given that this term is frequently used in the
literature, I will discuss the role of altruism in costly punishment.
GENERAL DISCUSSION 87
4.4 The subjective construction of social dilemmas
The famous Thomas-Theorem is grounded in the sentence “if men define situations as
real, they are real in their consequences” (Thomas & Thomas, 1928). That is, individuals
behave according to their subjective construction of reality (Esser, 2000; Smith &
Mackie, 2007). When participants face a public goods game in the laboratory, the
subjective conditions of such a situation indeed reflect a social dilemma situation.
Typically, individuals ascribe value to money, that is, they subjectively construe the
value of money (e.g., Bruner & Goodman, 1947). Having subjectively ascribed value to
money and wanting to maximize the amount they have of it (regardless of the actual
motive such as greed, selfishness, satisfying security needs etc.), individuals face a social
dilemma situation when confronted with a public goods game in the laboratory. If
individuals subjectively want more money the best strategy in this regard would be to
behave uncooperatively. The strategy of maximizing one’s own benefit has the effect
that the benefit of the collective is reduced as compared with a cooperative strategy. This
subjectively construed situation does reflect a social dilemma situation. As such, one can
certainly investigate social dilemma situations per se in the laboratory. There are
substantial advantages to investigating social dilemmas in the laboratory. Only
laboratories allow the investigation of social dilemma situations in a simple, reduced
situation in which external influences can be controlled and causal analytic approaches
applied, which would be rather difficult when investigating social dilemma situations in
the field (cf. van Lange et al., 2014).
With regard to the subjective construction of reality, individuals construe social dilemma
situations differently. Halevy et al. (2012) have impressively shown that when
individuals should describe a conflicting two-person situation, they mainly use either the
benefit matrix of a Prisoner’s Dilemma (in which the benefits of the two persons
maximally diverge), a Chicken Game (in which the benefits of the two persons
moderately diverge), an Assurance Game (in which the benefits of the two persons also
moderately diverge), or a Maximizing Difference Game (in which the benefits of the two
GENERAL DISCUSSION 88
persons correspond positively).36 Given that individuals construe a conflict differently, it
seems likely that individuals also construe an objective benefit matrix differently and in
accordance with their basic needs, motivations, attitudes, and values (Halevy, Chou, &
Murnighan, 2011). In other words, I propose that individuals construe their subjective
benefit within a given objective payoff matrix of a theoretical game paradigm. For
instance, individuals with a basic prosocial orientation (e.g., chronically altruistic
individuals; Kuhl & Kazén, 1997) should ascribe a greater value to cooperation in
comparison with competitive individuals (van Lange, 1997) because cooperation is more
in line with their basic values (Halevy et al., 2011).
While I argue that it is possible to investigate social dilemmas in the laboratory, the issue
of how specifically individuals subjectively construe an objectively given payoff matrix
in a social dilemma game still needs investigating. This also holds true for the subjective
construction of costly punishment. It has been shown that the fewer objective resources
one has to invest in costly punishment and the higher the impact on the punished is the
more individuals use costly punishment (Anderson & Putterman, 2006; Egas & Riedl,
2005). However, given that reality is a subjective construction, this question should also
be investigated from a subjective perspective, that is, does the same effect emerge when
individuals subjectively ascribe higher as opposed to lower costs to costly punishment?
This question is important given the strong influence of costly punishment on human
cooperation (Fehr & Gächter, 2002). Specifically, if individuals subjectively ascribe too
many costs to punishment the problem of how to establish cooperation is surely
pronounced.
Another future line of research could also contribute to a better understanding of how
individuals construe costly punishment. Specifically, one could investigate how
prevention-focused individuals construe the benefit of costly punishment and,
consequently, whether a possible specific subjective construction predicts their costly
punishment behavior. In this regard, an investigation of the subjective value of costly
36
This game actually represents no conflict because mutual cooperation is the joint optimum and a Nash
equilibrium.
GENERAL DISCUSSION 89
punishment when others are present (e.g., when a subtle cue of being watched is given)
also seems to be a valuable approach. Specifically, if one’s own reputational concerns
drive the effects, the subjectively considered benefit from engaging in costly punishment
should be reduced when others are present. This, in turn, should predict the actual
reduction of costly punishment under conditions of social presence.
Generally, it would be a valuable approach to investigate how individuals subjectively
construe costly punishment as well as the consequences of this subjectively construed
reality on objective cooperation levels. This strategy would, indeed, provide a better
understanding of why and under which conditions individuals engage in costly
punishment.
4.5 Implications for behavior in “real life”
Of course, studies in the laboratory remain silent (and should remain silent) about
humans’ general behavior in social dilemma situations in the “real world” (Mook, 1983).
This is, however, not considered when Camerer and Fehr (2004) state that “[f]or
example, if many people in a society exhibit inequality aversion or reciprocity, they will
be willing to punish those who do not share food, so no formal mechanism is needed to
govern food sharing. Without such preferences, formal mechanisms are needed to sustain
food sharing (or sharing does not occur at all)” (p. 56). This direct transfer from the
laboratory to general human behavior in real life is not an exceptional case (cf. Guala,
2012). It remains unclear whether motives, processes, or the function of specific
behavior observed in the laboratory also holds true in real life. For instance, it is
plausible that specific forms of punishment in real life are not driven by impulsivity
(e.g., when a ticket inspector punishes a free-riding fare dodger). That is, situations in
real life are often qualitatively different from those in the laboratory. Consequently, one
should avoid a direct transfer from the laboratory to general human behavior in real life.
Elinor Ostrom is a good example of how to integrate research from the laboratory and
research from the field. In her research, Ostrom (1990) investigated costly punishment in
social dilemma situations in the laboratory. In a next step, she observed punishment of
uncooperative individuals and its effects on free-riders in the real world. Not till then did
GENERAL DISCUSSION 90
she conclude that “graduated sanctions” (p. 94) represent one possibility for reducing the
free-rider problem in real world social dilemma situations. Of note, Ostrom speaks about
sanctions, not about costly sanctions or costly punishment. In line with this view, I want
to emphasize that the aspect of costs was not at the center of this dissertation and with
good cause given that the specificity of costly (monetary) punishment in comparison
with other forms of punishment (e.g., social disapproval) needs solid investigation in
future research. This was also reflected in the articles presented when I spoke about
reactive aggression but not about costly reactive aggression. It is still an unsolved issue
whether costly punishment, as conceptualized by evolutionary psychologists and
biologists (i.e., the costs of punishing are greater than the benefits) emerges in real life
(cf. Guala, 2012).
As a consequence of these considerations, I want to take an in-between step and
emphasize first what can be inductively concluded from the present research’s findings
and what can, from this theoretical basis, be deductively concluded for behavior in real
life.
Given the quantity of the appearances of social dilemmas in real life (van Lange et al.,
2013) I will pick one incisive example to make my argument. This example relates to
social ostracism, defined as the act of ignoring or excluding an individual from social
interaction (Williams, 1997; Williams, Cheung, & Choi, 2000). Ostracism emerges in a
variety of situations in real life (cf. Williams, 2007), for instance when it comes to
bullying at school, when women are not allowed to join students’ fraternities, or when an
obstructionist is excluded from a political party. Recently, Feinberg, Willer, and Schultz
(in press; see also Feinberg et al., 2012) have argued that social ostracism reflects one
form of punishment that leads to cooperation in social dilemma situations. In fact, these
authors showed that punishing uncooperative group members using ostracism in a social
dilemma situation leads to (a) increased cooperation on the collective level and (b) an
increase in the cooperative behavior of ostracized individuals after being ostracized (i.e.,
the individual level). In real life, a student study group may face a social dilemma
because individuals are best off when they do not contribute to the common group
project but let the others do all the work. In such situations, free-riding group members
may be ignored or even excluded from the group. In that way an uncooperative
GENERAL DISCUSSION 91
individual is punished by the other group members. This behavior can also be framed as
reactive aggression, given that the uncooperative individual is harmed in response to a
provocation. In this regard, I do not directly claim that the punishment toward a freerider in a student study group is more likely to be executed by individuals possessing
relative low (vs. high) cortisol levels, or more likely to be executed by strongly rather
than weakly prevention-focused individuals, or less likely to be executed when
individuals are in a state of reputational concern. Instead, I propose a theoretical step inbetween. Specifically, what can be inductively concluded is that cortisol negatively
relates to the punishment of unfair and uncooperative group members which is likely
impulsively executed. Moreover, strongly prevention-focused individuals are more likely
than weakly prevention-focused individuals to engage in costly punishment, presumably
because of their specific sensitivity to negative social information involving norm
violations. Finally, in a state of reputational concern individuals engage less strongly in
costly punishment, most likely because costly punishment may not be considered to be a
reputation-increasing behavior. These considerations represent a theoretical framework
for when and under which conditions individuals may engage in the punishment of
uncooperative others. From this basis then one has to analyze whether social ostracism in
a study group fits into the framework. This is questionable because ostracism in a study
group may not reflect impulsive behavior but is deliberately debated in the group, and
there may be a local social norm to ostracize (i.e., it is socially desirable to exclude an
uncooperative group member). On the other hand it is possible that strongly preventionfocused individuals particularly react to this negative event which involves a norm
violation. Thus, from this theoretical basis one can also come to different conclusions
regarding how two variables are related in real life, for instance that cortisol levels must
not be related to social ostracism in a study group.
Nonetheless, the investigation of the main concepts of the present work in real life is an
interesting approach when one has the aim of understanding these concepts as they apply
in real life. For instance, how are cortisol levels related to the support of punishment of
deviant behavior in the context of tax evasion? How is testosterone related to proactive
aggression, for instance in football supporters? Does cortisol inhibit this relation? Can
the analysis of reputational concerns in this context contribute to a better understanding
GENERAL DISCUSSION 92
of proactive aggression, especially when varying the source of reputational concerns
(e.g., security cameras, the police, or other supporters from one’s own team)?
To conclude, what I wish to emphasize here is that deduction for real life needs to result
from the rigorous analysis of a theoretical approach and the behavior of interest in real
life. The present work can surely offer a theoretical framework in this regard.
4.6 The role of altruism in costly punishment
In this section I will discuss the third issue under consideration, that is, the role of
altruism in punishing uncooperative group members. I am not the first to discuss the
question of whether costly punishment is an altruistic act, be it from a social
psychological perspective or from the perspective of evolutionary psychology and
biology (e.g., Clavien & Klein, 2010; Elster, 2007; Kurzban & DeScioli, 2013; Peacock,
2007; Sober & Wilson, 1998; West et al., 2011). The discussion, and the confusion,
started with Fehr and Gächter (2002) terming costly punishment of uncooperative
individuals “altruistic punishment.” The use of this term built on the notion that any
behavior that is “costly and yields no pecuniary benefits” (Fehr & Gächter, 2002, p. 137)
for the acting individual but benefits any other individual in any way reflects altruism.
Fehr and Gächter (2002) wrote: “punishment may well benefit the future group members
of a punished subject, if that subject responds to the punishment by raising investments
in the following periods. In this sense, punishment is altruistic” (p. 137).
Let us take a step back and reflect on what researchers in social psychology and
evolutionary psychology and biology typically imply when they use the term “altruism”.
From the perspective of social psychology, altruism reflects behavior with the ultimate
goal of increasing the welfare of one or more other individuals (Batson et al., 2002). This
conceptualization does not deny that behaving altruistically can benefit the altruistic
individual. However, an altruistic act from the perspective of social psychology is that an
individual’s ultimate motive is to increase the welfare of others. Benefits that are
obtained (e.g., the relief of a negative state) reflect unintended consequences (cf. Batson
et al., 2002).
From the perspective of evolutionary psychology and biology, altruism reflects a
behavior that directly benefits others at a cost to oneself whereas the actual
GENERAL DISCUSSION 93
psychological motive of the behavior must not be “psychologically altruistic” (Kerr,
Godfrey-Smith, & Feldman, 2004; Sober & Wilson, 1998). In other words, the basis of
sociobiological altruism is that individual A benefits individual B and individual A bears
the costs of doing so (Peacock, 2007; Sober & Wilson, 1998; West et al., 2011).
Fehr and Gächter (2002) reduce the concept of altruism ad absurdum (cf. Peacock,
2007). The definition of altruism given by these authors involves the logic that individual
A can be spiteful toward individual B which leads individual B to behave altruistically
(from a sociobiological viewpoint) toward individual C. On this basis, Fehr and Gächter
(2002) would claim that the spiteful act of individual A is altruistic although individual
A has no direct influence or control over the benefit of individual C. One can illustrate
the absurdity of this logic by claiming that if individual A kills individual B which
benefits individual C (e.g., through inheriting the estate of deceased individual B), the
killing of individual A would be altruistic in the sense dictated by Fehr and Gächter
(2002) because individual A killing B benefits C (cf. Peacock, 2007). Accordingly, this
conceptualization of altruism by Fehr and Gächter (2002) does not fit the common
definition of altruism found in the field of evolutionary psychology and biology, given
that the common definitions restrict altruism to the notion of individual A helping B.
Individual C is not considered.
Additionally, one can criticize the Fehr and Gächter (2002) definition of altruism
because they neglect what evolutionary theorists often emphasize, that is, costs and
benefits are calculated on the basis of the lifetime fitness consequences of a behavior
(e.g., Hamilton, 1964; West et al., 2011). Fehr and Gächter (2002), however, calculate
costs and benefits of a specific situation (i.e., the public goods game). So, it is possible
that individuals engage in costly behavior that benefits some other individuals (e.g.,
through costly punishment), and Fehr and Gächter (2002) would conclude that this is
altruism. However, when calculated on the basis of lifetime benefits, it could actually be
that the costly behavior, in sum over a lifetime, leads to a reduced benefit for the actor
(e.g., because costly punishment destroys a sense of community and individuals have
less trust in others when costly punishment is possible; cf. Shinada et al., 2004). As such,
GENERAL DISCUSSION 94
Fehr and Gächter (2002) do not distinguish between the classic rational choice approach
(i.e., maximizing benefits in a specific situation) and evolutionary models (i.e.,
maximizing benefits in the long run; cf. West et al., 2011). When evaluating whether a
behavior is altruistic in an evolutionary sense, one has to apply a lifetime cost-benefit
analysis and not, as Fehr and Gächter (2002) did, a cost-benefit analysis of a specific
situation (i.e., the public goods game). To conclude, the definition of altruism given by
Fehr and Gächter (2002) is definitely not in line with the typical conceptualization of
altruism in the field of evolutionary biology.
There is still the possibility that costly punishment reflects altruism in a social
psychological sense, that is, individuals costly punish uncooperative others with the
ultimate prosocial motive of increasing the welfare of others with whom the punished
individual will interact in future (possibly in a more cooperative manner). For instance, it
could be the case that the observation of uncooperative behavior leads to anger which in
turn elicits the motivation to reinstall justice and, accordingly, to help people (Haidt,
2003). Peacock (2007), however, rightly states in relation to this that “upholding justice
carries no necessary intent to benefit others” (p. 14). Our own work (Pfattheicher et al.,
2014, Appendix C) can contribute to this issue. The finding that prosocial behavior with
prosocial motives is more likely shown when reputational concerns are activated (Nettle
et al., 2013; Sparks & Barclay, 2013) whereas costly punishment was actually reduced in
our experiments, does not bolster the notion that prosocial, altruistic motives drive costly
punishment. Additionally, costly punishment of uncooperative individuals is considered
to be driven by revenge (Gollwitzer & Bushman, 2012; de Quervain et al., 2004) and
revenge includes the ultimate goal of harming another individual (McCullough, 2008),
which contradicts an altruistic motivation. This is in line with our own work
(Pfattheicher & Keller, in press, Appendix A) which conceptualizes costly punishment as
reactive aggression and shows corresponding empirical findings.
To sum up, it is unlikely that costly punishment of uncooperative individuals is driven by
altruistic motives (although I do not want to completely exclude this motive for all
individuals). The term “altruistic punishment” does not seem appropriate, be it from a
GENERAL DISCUSSION 95
social psychological perspective or from the perspective of evolutionary psychology and
biology and should, therefore, be avoided.
A few last words: the present dissertation proposed that investigating cooperation and
punishment in social dilemma situations and, more generally, corresponding concepts
such as aggression, dominance, impulsivity, selfishness, status and fairness concerns,
group orientations, and social norms is a worthwhile approach. In fact, the analysis of
behavior in social dilemma situations offers fruitful possibilities to acquire knowledge
about human behavior. The present dissertation represents the reaping of some of these
fruits, but certainly not the end of the harvest.
REFERENCES 96
References
Adler, N. E., Epel, E. S., Castellazzo, G., & Ickovics, J. R. (2000). Relationship of
subjective and objective social status with psychological and physiological
functioning: Preliminary data in healthy white women. Health Psychology, 19,
586-592.
Alink, L. R., van IJzendoorn, M. H., Bakermans-Kranenburg, M. J., Mesman, J., Juffer,
F., & Koot, H. M. (2008). Cortisol and externalizing behavior in children and
adolescents: Mixed meta-analytic evidence for the inverse relation of basal cortisol
and cortisol reactivity with externalizing behavior. Developmental Psychobiology,
50, 427-450.
Anderson, C. M., & Putterman, L. (2006). Do non-strategic sanctions obey the law of
demand? The demand for punishment in the voluntary contribution mechanism.
Games and Economic Behavior, 54, 1-24.
Anderson, C. A., & Bushman, B. J. (2002). Human aggression. Annual Review of
Psychology, 53, 27-51.
Andreoni, J., & Bernheim, B. D. (2009). Social image and the 50-50 norm: A theoretical
and experimental analysis of audience effects. Econometrica, 77, 1607-1636.
Arch, J. J., & Craske, M. G. (2006). Mechanisms of mindfulness: Emotion regulation
following a focused breathing induction. Behaviour Research and Therapy, 44,
1849-1858.
Atkinson, J. W. (1957). Motivational determinants of risk-taking behavior.
Psychological Review, 64, 359-372.
Atkinson, J. W. (1964). An introduction of motivation. Princeton, NJ: van Nostrand.
Aune, K. S. (1997). Self and partner perceptions of the appropriateness of emotions.
Communication Reports, 10, 133-142.
REFERENCES 97
Aune, K. S., Aune, R. K., & Buller, D. B. (1994). The experience, expression, and
perceived appropriateness of emotions across levels of relationship development.
The Journal of Social Psychology, 134, 141-150.
Axelrod, R., & Hamilton, W. D. (1981). The evolution of cooperation. Science, 211,
1390-1396.
Balleine, B. W., Delgado, M. R., & Hikosaka, O. (2007). The role of the dorsal striatum
in reward and decision-making. The Journal of Neuroscience, 27, 8161-8165.
Balliet, D., Mulder, L. B., & van Lange, P. A. (2011). Reward, punishment, and
cooperation: A meta-analysis. Psychological Bulletin, 137, 594-615.
Banny, A. M., Heilbron, N., Ames, A., & Prinstein, M. J. (2011). Relational benefits of
relational aggression: Adaptive and maladaptive associations with adolescent
friendship quality. Developmental Psychology, 47, 1153-1166.
Barclay, P. (2006). Reputational benefits for altruistic punishment. Evolution and Human
Behavior, 27, 325-344.
Barclay, P. (2013). Strategies for cooperation in biological markets, especially for
humans. Evolution and Human Behavior, 34, 164-175.
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.
Bateson, M., Callow, L., Holmes, J. R., Redmond Roche, M. L., & Nettle, D. (2013). Do
images of ‘watching eyes’ induce behaviour that is more pro-social or more
normative? A field experiment on littering. PLOS ONE, 8, e82055.
Bateson, M., Nettle, D., & Roberts, G. (2006). Cues of being watched enhance
cooperation in a real-world setting. Biology Letters, 2, 412-414.
REFERENCES 98
Batson, C. D., Ahmad, N., Powell, A. A., & Stocks, E. L. (2008). Prosocial motivation.
In J. Y. Shah & W. L. Gardner (Eds.), The handbook of motivation science (pp.
135-149). New York: Guilford Press.
Batson, C. D., Ahmad, N., & Tsang, J. A. (2002). Four motives for community
involvement. Journal of Social Issues, 58, 429-445.
Baumeister, R. F. (2002). Yielding to temptation: Self-control failure, impulsive
purchasing, and consumer behavior. Journal of Consumer Research, 28, 670-676.
Becker, G. S. (1976). The economic approach to human behavior. Chicago, IL:
University of Chicago Press.
Bendor, J., & Mookherjee, D. (1990). Norms, third-party sanctions, and cooperation.
Journal of Law, Economics, & Organization, 6, 33-63.
Ben-Shakhar, G., Bornstein, G., Hopfensitz, A., & van Winden, F. (2007). Reciprocity
and emotions in bargaining using physiological and self-report measures. Journal
of Economic Psychology, 28, 314-323.
Berg, J., Dickhaut, J., & McCabe, K. (1995). Trust, reciprocity and social history. Games
and Economic Behavior, 10, 122-142.
Bernhard, H., Fischbacher, U., & Fehr, E. (2006). Parochial altruism in humans. Nature,
442, 912-915.
Blair, C., Peters, R., & Granger, D. (2004). Physiological and neuropsychological
correlates of approach/withdrawal tendencies in preschool: Further examination of
the behavioral inhibition system/behavioral activation system scales for young
children. Developmental Psychobiology, 45, 113-124.
Bless, H., Fiedler, K., & Strack, F. (2004). Social cognition: How individuals construct
social reality. Hove: Psychology Press.
REFERENCES 99
Bliege Bird, R., Smith, E. A., & Bird, D. W. (2001). The hunting handicap: Costly
signaling in human foraging strategies. Behavioral Ecology and Sociobiology 50,
9-19.
Binmore, K. G. (2005). Economic man-or straw man? Behavioral and Brain Sciences,
28, 817-818.
Binmore, K., & Shaked, A. (2010). Experimental economics: Where next? Journal of
Economic Behavior & Organization, 73, 87-100.
Böhnke, R., Bertsch, K., Kruk, M. R., & Naumann, E. (2010). The relationship between
basal and acute HPA axis activity and aggressive behavior in adults. Journal of
Neural Transmission, 117, 629-637.
Böhnke, R., Bertsch, K., Kruk, M. R., Richter, S., & Naumann, E. (2010). Exogenous
cortisol
enhances
aggressive behavior in
females,
but
not
in
males.
Psychoneuroendocrinology, 35, 1034-1044.
Bolton, G. E., & Ockenfels, A. (2000). A theory of equity, reciprocity, and competition.
American Economic Review, 90, 166-193.
Bourdieu, P. (1979). Distinction: A social critique of the judgment of taste. Cambridge,
MA: Harvard University Press.
Bourrat, P., Baumard, N., & McKay, R. (2011). Surveillance cues enhance moral
condemnation. Evolutionary Psychology, 9, 193-199.
Bowles, S., & Gintis, H. (2003). Origins of human cooperation. In P. Hammerstein (Ed.),
Genetic and cultural evolution of cooperation (pp. 429-443). Cambridge, MA:
MIT Press.
Boyd, R., & Richerson, P. J. (1992). Punishment allows the evolution of cooperation (or
anything else) in sizable groups. Ethology and Sociobiology, 13, 171-195.
REFERENCES 100
Brandstätter, H., & Königstein, M. (2001). Personality influences on ultimatum
bargaining decisions. European Journal of Personality, 15, 53-70.
Bromberg-Martin, E. S., Matsumoto, M., & Hikosaka, O. (2010). Dopamine in
motivational control: Rewarding, aversive, and alerting. Neuron, 68, 815-834.
Brown, L. L., Tomarken, A. J., Orth, D. N., Loosen, P. T., Kalin, N. H., & Davidson, R.
J. (1996). Individual differences in repressive-defensiveness predict basal salivary
cortisol levels. Journal of Personality and Social Psychology, 70, 362-371.
Brüne, M., Juckel, G., & Enzi, B. (2013). “An eye for an eye”? Neural correlates of
retribution and forgiveness. PLOS ONE, 8, e73519.
Brüne, M., Scheele, D., Heinisch, C., Tas, C., Wischniewski, J., & Güntürkün, O.
(2012). Empathy moderates the effect of repetitive transcranial magnetic
stimulation of the right dorsolateral prefrontal cortex on costly punishment. PLOS
ONE, 7, e44747.
Bruner, J. S., & Goodman, C. C. (1947). Value and need as organizing factors in
perception. The Journal of Abnormal and Social Psychology, 42, 33-44.
Bullock, J. G., Green, D. P., & Ha, S. E. (2010). Yes, but what’s the mechanism? (Don’t
expect an easy answer). Journal of Personality and Social Psychology, 98, 550558.
Burnham, T. C. (2007). High-testosterone men reject low ultimatum game offers.
Proceedings of the Royal Society B: Biological Sciences, 274, 2327-2330.
Burnham, T. C., & Hare, B. (2007). Engineering human cooperation. Human Nature, 18,
88-108.
Bushman, B. J., & Anderson, C. A. (2002). Violent video games and hostile
expectations: A test of the General Aggression Model. Personality and Social
Psychology Bulletin, 28, 1679-1686.
REFERENCES 101
Bushman, B. J., & Huesmann, L. R. (2010). Aggression. In S. T. Fiske, D. T. Gilbert &
G. Lindzey (Eds.), Handbook of social psychology (5th ed., pp. 833-863). New
York: John Wiley & Sons.
Buss, A. H. (1961). The psychology of aggression. New York: Wiley.
Caldwell, M. D. (1976). Communication and sex effects in a five-person Prisoner’s
Dilemma Game. Journal of Personality and Social Psychology, 33, 273-280.
Calogero, A. E., Gallucci, W. T., Gold, P. W., & Chrousos, G. P. (1988). Multiple
feedback regulatory loops upon rat hypothalamic corticotropin-releasing hormone
secretion. Potential clinical implications. Journal of Clinical Investigation, 82,
767-774.
Camerer, C. F. (2003). Behavioral game theory: Experiments in strategic interaction.
New York: Russell Sage Foundation.
Camerer, C. F., & Fehr, E. (2004). Measuring social norms and preferences using
experimental games: A guide for social scientists. In J. Henrich, R. Boyd, S.
Bowles, C. Camerer, E. Fehr & H. Gintis (Eds.), Foundations of human sociality
(pp. 55-95). Oxford: Oxford University Press.
Caporael, L. R., Dawes, R. M., Orbell, J. M., & van de Kragt, A. J. (1989). Selfishness
examined: Cooperation in the absence of egoistic incentives. Behavioral and Brain
Sciences, 12, 683-739.
Cappelletti, D., Goth, W., & Ploner, M. (2011). Being of two minds: Ultimatum offers
under cognitive constraints. Journal of Economic Psychology, 32, 940-950.
Carbon, C., & Hesslinger, V. M. (2011). Bateson et al.’s (2006) cues-of-being-watched
paradigm revisited. Swiss Journal of Psychology, 70, 203-210.
Carlsmith, K. M., & Darley, J. M. (2008). Psychological aspects of retributive justice.
Advances in Experimental Social Psychology, 40, 193-236.
REFERENCES 102
Carlsmith, K. M., Wilson, T. D., & Gilbert, D. T. (2008). The paradoxical consequences
of revenge. Journal of Personality and Social Psychology, 95, 1316-1324.
Carlsmith, K. M., Darley, J. M., & Robinson, P. H. (2002). Why do we punish?
Deterrence and just deserts as motives for punishment. Journal of Personality and
Social Psychology, 83, 284-299.
Carré, J. M., & Mehta, P. H. (2011). Importance of considering testosterone-cortisol
interactions in predicting human aggression and dominance. Aggressive Behavior,
37, 1-3.
Carré, J. M., Putnam, S. K., & McCormick, C. M. (2009). Testosterone responses to
competition predict future aggressive behaviour at a cost to reward in men.
Psychoneuroendocrinology, 34, 561-570.
Carver, C. S. (2006). Approach, avoidance, and the self-regulation of affect and action.
Motivation and Emotion, 30, 105-110.
Carver, C. S., Lawrence, J. W., & Scheier, M. F. (1999). Self-discrepancies and affect:
Incorporating the role of feared selves. Personality and Social Psychology Bulletin,
25, 783-792.
Carver, C. S., & Scheier, M. F. (1981). Attention and self-regulation: A control-theory
approach to human behavior. New York: Springer.
Carver, C. S., & Scheier, M. F. (1990). Origins and functions of positive and negative
affect: A control-process view. Psychological Review, 97, 19-35.
Carver, C. S., & Scheier, M. F. (2011). Self regulation of action and affect. In K. D.
Vohs & R. F. Baumeister (Eds.), Handbook of self-regulation: Research, theory,
and application (pp. 3-21). New York: Guilford Press.
REFERENCES 103
Carver, C., & White, T. (1994). Behavioral inhibition, behavioral activation, and
affective responses to impending reward and punishment: The BIS/BAS scales.
Journal of Personality and Social Psychology, 67, 319-333.
Castrogiovanni, P., Capone, M. R., Maremmani, I., & Marazziti, D. (1994). Platelet
serotonergic
markers
and
aggressive
behaviour
in
healthy
subjects.
Neuropsychobiology, 29, 105-107.
Cheng, J. T., Tracy, J. L., Foulsham, T., Kingstone, A., & Henrich, J. (2013). Two ways
to the top: Evidence that dominance and prestige are distinct yet viable avenues to
social rank and influence. Journal of Personality and Social Psychology, 104, 103125.
Clavien, C., & Klein, R. A. (2010). Eager for fairness or for revenge? Psychological
altruism in economics. Economics and Philosophy, 26, 267-290.
Cialdini, R. B., Reno, R. R., & Kallgren, C. A. (1990). A focus theory of normative
conduct: Recycling the concept of norms to reduce littering in public places.
Journal of Personality and Social Psychology, 58, 1015-2026.
Cohrs, J. C. (2013). Threat and authoritarianism: Some theoretical and methodological
comments. International Journal of Psychology, 48, 50-54.
Corr, P. J. (2009). The reinforcement sensitivity theory of personality. In P. J. Corr & G.
Matthews (Eds.), The Cambridge handbook of personality psychology (pp. 347376). Cambridge, MA: Cambridge University Press.
Crick, N. R., & Dodge, K. A. (1996). Social information‐processing mechanisms in
reactive and proactive aggression. Child Development, 67, 993-1002.
Critchley, H. D., Elliott, R., Mathias, C. J., & Dolan, R. J. (2000). Neural activity
relating to generation and representation of galvanic skin conductance responses:
A functional magnetic resonance imaging study. The Journal of Neuroscience, 20,
3033-3040.
REFERENCES 104
Crockett, M. J. (2009). The neurochemistry of fairness: Clarifying the link between
serotonin and prosocial behavior. Annals of the New York Academy of Sciences,
1167, 76-86.
Crockett, M. J., Clark, L., Lieberman, M. D., Tabibnia, G., & Robbins, T. W. (2010).
Impulsive choice and altruistic punishment are correlated and increase in tandem
with serotonin depletion. Emotion, 10, 855-862.
Crowe, E., & Higgins, E. T. (1997). Regulatory focus and strategic inclinations:
Promotion and prevention in decision-making. Organizational Behavior and
Human Decision Processes, 69, 117-132.
Crowne, D. P., & Marlowe, D. (1960). A new scale of social desirability independent of
psychopathology. Journal of Consulting Psychology, 24, 349-354.
Cubitt, R. P., Drouvelis, M., Gächter, S., & Kabalin, R. (2011). Moral judgments in
social dilemmas: How bad is free riding? Journal of Public Economics, 95, 253264.
Cumming, D. C., Quigley, M. E., & Yen, S. S. C. (1983). Acute suppression of
circulating testosterone levels by cortisol in men. Journal of Clinical
Endocrinology & Metabolism, 57, 671-673.
d’Arms, J., & Jacobson, D. (2000). The moralistic fallacy: On the ‘appropriateness’ of
emotions. Philosophical and Phenomenological Research, 61, 65-90.
Dawes, C. T., Fowler, J. H., Johnson, T., McElreath, R., & Smirnov, O. (2007).
Egalitarian motives in humans. Nature, 446, 794-796.
Dawes, R. M. (1980). Social dilemmas. Annual Review of Psychology, 31, 169-193.
Delgado, M. R., Li, J., Schiller, D., & Phelps, E. A. (2008). The role of the striatum in
aversive learning and aversive prediction errors. Philosophical Transactions of the
Royal Society B: Biological Sciences, 363, 3787-3800.
REFERENCES 105
Denson, T. F., Pedersen, W. C., Ronquillo, J., & Nandy, A. S. (2009). The angry brain:
Neural correlates of anger, angry rumination, and aggressive personality. Journal
of Cognitive Neuroscience, 21, 734-744.
de Quervain, D. J.-F., Fischbacher, U., Treyer, V., Schellhammer, M., Schnyder, U.,
Buck, A., & Fehr, E. (2004). The neural basis of altruistic punishment. Science
305, 1254-1258.
Denson, T. F., von Hippel, W., Kemp, R. I., & Teo, L. S. (2010). Glucose consumption
decreases impulsive aggression in response to provocation in aggressive
individuals. Journal of Experimental Social Psychology, 46, 1023-1028.
DeWall, C. N., Anderson, C. A., & Bushman, B. J. (2011). The General Aggression
Model: Theoretical extensions to violence. Psychology of Violence, 1, 245-258.
DeWall, C. N., Baumeister, R. F., Gailliot, M. T., & Maner, J. K. (2008). Depletion
makes the heart grow less helpful: Helping as a function of self-regulatory energy
and genetic relatedness. Personality and Social Psychology Bulletin, 34, 16531662.
DeWall, C. N., Baumeister, R. F., Stillman, T. F., & Gailliot, M. T. (2007). Violence
restrained: Effects of self-regulation and its depletion on aggression. Journal of
Experimental Social Psychology, 43, 62-76.
Dickerson, S. S., & Kemeny, M. E. (2004). Acute stressors and cortisol responses: A
theoretical integration and synthesis of laboratory research. Psychological Bulletin,
130, 355-391.
Diekmann, A. (2009). Spieltheorie: Einführung, Beispiele, Experimente [Game theory:
Introduction, examples, experiments]. Hamburg: Rowohlt.
Diener, E. (1976). Effects of prior destructive behavior, anonymity, and group presence
on deindividuation and aggression. Journal of Personality and Social Psychology,
33, 497-507.
REFERENCES 106
Diener, E., Fraser, S. C., Beaman, A. L., & Kelem, R. T. (1976). Effects of
deindividuation variables on stealing among Halloween trick-or-treaters. Journal
of Personality and Social Psychology, 33, 178-183.
Diener, E., Lusk, R., DeFour, D., & Flax, R. (1980). Deindividuation: Effects of group
size, density, number of observers, and group member similarity on selfconsciousness and disinhibited behavior. Journal of Personality and Social
Psychology, 39, 449-459.
Dos Santos, M., Rankin, D. J., & Wedekind, C. (2011). The evolution of punishment
through reputation. Proceedings of the Royal Society B: Biological Sciences, 278,
371-377.
Dreber, A., & Rand, D. G. (2012). Retaliation and antisocial punishment are overlooked
in many theoretical models as well as behavioral experiments. Behavioral and
Brain Sciences, 35, 24-24.
Egas, M., & Riedl, A. (2005). The economics of altruistic punishment and the demise of
cooperation (Working Paper). Amsterdam: Tinbergen Institute.
Eisenegger, C., Haushofer, J., & Fehr, E. (2011). The role of testosterone in social
interaction. Trends in Cognitive Sciences, 15, 263-271.
Eisenegger, C., Naef, M., Snozzi, R., Heinrichs, M., & Fehr, E. (2010). Prejudice and
truth about the effect of testosterone on human bargaining behaviour. Nature, 463,
356-359.
Ekström, M. (2011). Do watching eyes affect charitable giving? Evidence from a field
experiment. Experimental Economics, 15, 530-546.
Elliot, A. J. (1999). Approach and avoidance motivation and achievement goals.
Educational Psychologist, 34, 169-189.
REFERENCES 107
Elliot, A. J., & McGregor, H. A. (1999). Test anxiety and the hierarchical model of
approach and avoidance achievement motivation. Journal of Personality and
Social Psychology, 76, 628-644.
Elliot, A. J., & McGregor, H. A. (2001). A 2 × 2 achievement goal framework. Journal
of Personality and Social Psychology, 80, 501-519.
Elliott, E. S., & Dweck, C. S. (1988). Goals: An approach to motivation and
achievement. Journal of Personality and Social Psychology, 54, 5-12.
Elster, J. (2007). Explaining social behavior. Cambridge, MA: Cambridge University
Press.
Emanuele, E., Brondino, N., Bertona, M., Re, S., & Geroldi, D. (2008). Relationship
between platelet serotonin content and rejections of unfair offers in the ultimatum
game. Neuroscience Letters, 437, 158-161.
Emler, N. (1990). A social psychology of reputation. European Review of Social
Psychology, 1, 171-193.
Ernest-Jones, M., Nettle, D., & Bateson, M. (2010). Effects of eye images on everyday
cooperative behavior: A field experiment. Evolution and Human Behavior, 32,
172-178.
Esser, H. (2000). Soziologie. Soziales Handeln [Sociology. Social action] (3rd ed.).
Frankfurt am Main: Campus.
Eysenck, H. J. (1970). The biological basis of personality. Springfield, IL: Charles C.
Thomas Publisher Ltd.
Fazio, R. H. (1990). The MODE model as an integrative framework. Advances in
Experimental Social Psychology, 23, 75-109.
Fehr, E., & Fischbacher, U. (2003). The nature of human altruism. Nature, 425, 785-791.
REFERENCES 108
Fehr, E., & Fischbacher, U. (2004). Social norms and human cooperation. Trends in
Cognitive Sciences, 8, 185-190.
Fehr, E., & Gächter, S. (2002). Altruistic punishment in humans. Nature, 415, 137-140.
Fehr, E., & Rockenbach, B. (2003). Detrimental effects of sanctions on human altruism.
Nature, 422, 137-140.
Fehr, E., & Schmidt, K. M. (1999). A theory of fairness, competition, and cooperation.
The Quarterly Journal of Economics, 114, 817-868.
Fehr, E., & Schneider, F. (2010). Eyes are on us, but nobody cares: Are eye cues relevant
for strong reciprocity? Proceedings of the Royal Society B: Biological Sciences, 277,
1315-1323.
Feilhauer, J., Cima, M., Korebrits, A., & Nicholson, N. A. (2013). Salivary cortisol and
psychopathy dimensions in detained antisocial adolescents. Psychoneuroendocrinology, 38, 1586-1595.
Feinberg, M., Cheng, J. T., & Willer, R. (2012). Gossip as an effective and low-cost
form of punishment. Behavioral and Brain Sciences, 35, 25-25.
Feinberg, M., Willer, R., & Schultz, M. (in press). Gossip and ostracism promote
cooperation in groups. Psychological Science.
Fenigstein, A., Scheier, M., & Buss, A. (1975). Public and private self-consciousness:
Assessment and theory. Journal of Consulting and Clinical Psychology, 43, 522527.
Fenigstein, A., & Vanable, P. A. (1992). Paranoia and self-consciousness. Journal of
Personality and Social Psychology, 62, 129-138.
Festinger, L., Pepitone, A., & Newcomb, T. (1952). Some consequences of deindividuation in a group. The Journal of Abnormal and Social Psychology, 47,
382-389.
REFERENCES 109
Fetchenhauer, D., & Huang, X. (2004). Justice sensitivity and distributive decisions in
experimental games. Personality and Individual Differences, 36, 1015-1029.
Fiedler, K., Schott, M., & Meiser, T. (2011). What mediation analysis can (not) do.
Journal of Experimental Social Psychology, 47, 1231-1236.
Fischbacher, U. (2007). z-Tree: Zurich toolbox for ready-made economic experiments.
Experimental Economics, 10, 171-178.
Fischer, P., Krueger, J., Greitemeyer, T., Kastenmüller, A., Vogrincic, C., Frey, D., … &
Kainbacher, M. (2011). The bystander-effect: A meta-analytic review on bystander
intervention in dangerous and non-dangerous emergencies. Psychological Bulletin,
137, 517-537.
Fishbein, A. P., & Ajzen, I. (1975). Belief, attitude, intention, and behavior: An
introduction to theory and research. Reading, MA: Addison-Wesley.
Fowler, J. H., Johnson, T., & Smirnov, O. (2005). Egalitarian motive and altruistic
punishment. Nature, 433, E1-E2.
Fox, J., & Guyer, M. (1978). “Public” choice and cooperation in n-person Prisoner’s
Dilemma. Journal of Conflict Resolution, 22, 469-481.
Francey, D., & Bergmüller, R. (2012). Images of eyes enhance investments in a real-life
public good. PLOS ONE, 7, e37397.
Friedman, R. S., & Förster, J. (2001). The effects of promotion and prevention cues on
creativity. Journal of Personality and Social Psychology, 81, 1001-1013.
Frijda, N. H. (1994). The lex talionis: On vengeance. In S. H. M. van Goozen, N. E. van
der Poll & J. A. Sergeant (Eds.), Emotions: Essays on emotion theory (pp. 263289). Hillsdale, NJ: Lawrence Erlbaum.
REFERENCES 110
Froming, W. J., Walker, G. R., & Lopyan, K. J. (1982). Public and private selfawareness: When personal attitudes conflict with societal expectations. Journal of
Experimental Social Psychology, 18, 476-487.
Gable, S. L., Reis, H. T., & Elliot, A. J. (2000). Behavioral activation and inhibition in
everyday life. Journal of Personality and Social Psychology, 78, 1135-1149.
Gächter, S., & Herrmann, B. (2009). Reciprocity, culture and human cooperation:
Previous insights and a new cross-cultural experiment. Philosophical Transactions
of the Royal Society B, 364, 791-806.
Garcia, S. M., Weaver, K., Moskowitz, G. B., & Darley, J. M. (2002). Crowded minds:
The implicit bystander effect. Journal of Personality and Social Psychology, 83,
843-853.
Gawronski, B., & Bodenhausen, G. V. (2006). Associative and propositional processes
in evaluation: An integrative review of implicit and explicit attitude change.
Psychological Bulletin, 132, 692-731.
Gawronski, B., & Bodenhausen, G. V. (2007). Unraveling the processes underlying
evaluation: Attitudes from the perspective of the APE model. Social Cognition, 25,
687-717.
Gervais, W., & Norenzayan, A. (2012). Like a camera in the sky? Thinking about God
increases public self-awareness and socially desirable responding. Journal of
Experimental Social Psychology, 48, 298-302.
Gilovich, T., Kruger, J., & Medvec, V. H. (2002). The spotlight effect revisited:
Overestimating the manifest variability of our actions and appearance. Journal of
Experimental Social Psychology, 38, 93-99.
Gilovich, T., Medvec, V., & Savitsky, K. (2000). The spotlight effect in social judgment:
An egocentric bias in estimates of the salience of one’s own actions and
appearance. Journal of Personality and Social Psychology, 78, 211-222.
REFERENCES 111
Goette, L., Huffman, D., Meier, S., & Sutter, M. (2012). Competition between
organizational groups: Its impact on altruistic and anti-social motivations.
Management Science, 58, 948-960.
Gollwitzer, M., & Bushman, B. J. (2012). Do victims of injustice punish to improve their
mood? Social Psychological and Personality Science, 3, 572-580.
Govern, J., & Marsch, L. (2001). Development and validation of the situational SelfAwareness Scale. Consciousness and Cognition, 10, 366-378.
Grafen, A. (1990). Biological signals as handicaps. Journal of Theoretical Biology, 144,
517-546.
Gray J. A. (1970). The psychophysiological basis of introversion-extraversion. Behavior
Research and Therapy, 8, 249-266.
Gray, J. A. (1982). The neuropsychology of anxiety: An enquiry into the functions of the
septo-hippocampal system. Oxford: Oxford University Press.
Grier, R. A., Warm, J. S., Dember, W. N., Matthews, G., Galinsky, T. L., Szalma, J. L.,
& Parasuraman, R. (2003). The vigilance decrement reflects limitations in effortful
attention, not mindlessness. The Journal of the Human Factors and Ergonomics
Society, 45, 349-359.
Griskevicius, V., Tybur, J. M., & Den Bergh, B. V. (2010). Going green to be seen:
Status, reputation, and conspicuous conservation. Journal of Personality and
Social Psychology, 98, 392-404.
Griskevicius, V., Tybur, J. M., Gangestad, S. W., Perea, E. F., Shapiro, J. R., & Kenrick,
D. T. (2009). Aggress to impress: Hostility as an evolved context-dependent
strategy. Journal of Personality and Social Psychology, 96, 980-994.
Guala, F. (2012). Reciprocity: Weak or strong? What punishment experiments do (and
do not) demonstrate. Behavioral and Brain Sciences, 35, 1-59.
REFERENCES 112
Guney, S., & Newell, B. R. (2013). Fairness overrides reputation: The importance of
fairness norms in altruistic cooperation. Frontiers in Human Neuroscience, 7, 1-12.
Gürerk, Ö., 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, 367388.
Hagen, E. H., & Hammerstein, P. (2006). Game theory and human evolution: A critique
of some recent interpretations of experimental games. Theoretical Population
Biology, 69, 339-348.
Haidt, J. (2003). The moral emotions. In R. J. Davidson, K. R. Scherer & H. H.
Goldsmith (Eds.), Handbook of affective sciences (pp. 852-870). Oxford: Oxford
University Press.
Haidt, J., & Kesebir, S. (2010). Morality. In S. Fiske, D. Gilbert & G. Lindzey (Eds.),
Handbook of social psychology (pp. 797-832), Hobeken, NJ: Wiley.
Halali, E., Bereby-Meyer, Y., & Meiran, N. (in press). Between self-interest and
reciprocity: The social bright side of self-control failure. Journal of Experimental
Psychology: General.
Haley, K. J., & Fessler, D. M. T. (2005). Nobody’s watching? Subtle cues affect
generosity in an anonymous economic game. Evolution and Human Behavior, 26,
245-256.
Halevy, N., Chou, E. Y., & Murnighan, J. K. (2011). Games groups play: Mental models
in intergroup conflict and negotiation. Research on Managing Groups and Teams,
14, 79-107.
REFERENCES 113
Halevy, N., Chou, E. Y., & Murnighan, J. K. (2012). Mind games: The mental
representation of conflict. Journal of Personality and Social Psychology, 102, 132148.
Hamilton, W. D. (1964). The genetical evolution of social behaviour. Journal of
Theoretical Biology, 7, 17-52.
Hamilton, W. D., & Zuk, M. (1982). Heritable true fitness and bright birds: A role for
parasites? Science, 218, 384-387.
Hansen, J., Kharecha, P., Sato, M., Masson-Delmotte, V., Ackerman, F., Beerling, D. J.,
... & Zachos, J. C. (2013). Assessing “dangerous climate change”: Required
reduction of carbon emissions to protect young people, future generations and
nature. PLOS ONE, 8, e81648.
Hardin, G. (1968). The tragedy of the commons. Science, 162, 1243-1248.
Hawley, P. H., & Vaughn, B. E. (2003). Aggression and adaptive functioning: The bright
side to bad behavior. Merrill-Palmer Quarterly, 49, 239-242.
Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process
analysis. New York: The Guilford Press.
Henrich, J., McElreath, R., Barr, A., Ensminger, J., Barrett, C., Bolyanatz, … & Ziker, J.
(2006). Costly punishment across human societies. Science, 312, 1767-1770.
Hessing, D. J., Elffers, H., Robben, H. S. J., & Webley, P. (1993). Needy or greedy? The
social psychology of individuals who fraudulently claim unemployment benefits.
Journal of Applied Social Psychology, 23, 226-243.
Herrmann, B., Thöni, C., & Gächter, S. (2008). Antisocial punishment across societies.
Science, 319, 1362-1367.
Higgins, E. T. (1997). Beyond pleasure and pain. American Psychologist, 52, 1280-1300.
REFERENCES 114
Higgins, E. T. (1998). Promotion and prevention: Regulatory focus as a motivational
principle. In M. P. Zanna (Ed.), Advances in experimental social psychology (pp.
1-46). San Diego, CA: Academic Press.
Higgins, E. T. (2012a). Beyond pleasure and pain: How motivation works. Oxford:
Oxford University Press.
Higgins, E. T. (2012b). Regulatory focus theory. In P. A. van Lange, A. W. Kruglanski
& E. T. Higgins (Eds.), Handbook of theories of social psychology (Vol. 1, pp.
483-504). London: Sage Publications.
Higgins, E. T., Friedman, R., Harlow, R., Idson, L., Ayduk, O., & Taylor, A. (2001).
Achievement orientations from subjective histories of success: Promotion pride
versus prevention pride. European Journal of Social Psychology, 31, 3-23.
Hilbe, C., & Traulsen, A. (2012). Emergence of responsible sanctions without second
order free riders, antisocial punishment or spite. Scientific Reports, 2, 458-458.
Hilbig, B. E., Zettler, I., & Heydasch, T. (2012). Personality, punishment and public
goods: Strategic shifts towards cooperation as a matter of dispositional honestyhumility. European Journal of Personality, 26, 245-254.
Hilton, N. Z., Harris, G. T., & Rice, M. E. (2000). The functions of aggression by male
teenagers. Journal of Personality and Social Psychology, 79, 988-994.
Hirsh, J. B., & Peterson, J. B. (2009). Extraversion, neuroticism, and the prisoner’s
dilemma. Personality and Individual Differences, 46, 254-256.
Hoffman, E., McCabe, K., & Smith, V. L. (1996). Social distance and other-regarding
behavior in dictator games. The American Economic Review, 86, 653-660.
Horne, C. (2001). The enforcement of norms: Group cohesion and meta-norms. Social
Psychology Quarterly, 64, 253-266.
REFERENCES 115
Hubbard, J. A., Dodge, K. A., Cillessen, A. H., Coie, J. D., & Schwartz, D. (2001). The
dyadic nature of social information processing in boys’ reactive and proactive
aggression. Journal of Personality and Social Psychology, 80, 268-280.
Hull, C. L. (1943). Principles of behavior. New York: Appleton-Century-Crofts.
Hull, C. L. (1952). A behavior system: An introduction to behavior theory concerning
the individual organism. New Haven, CT: Yale University Press.
Isaac, R. M., & Walker, J. M. (1988). Communication and free‐riding behavior: The
voluntary contribution mechanism. Economic Inquiry, 26, 585-608.
Irwin, K., & Berigan, N. (2013). Trust, culture, and cooperation: A social dilemma
analysis of pro‐environmental behaviors. The Sociological Quarterly, 54, 424-449.
Izuma, K. (2012). The social neuroscience of reputation. Neuroscience Research, 72,
283-288.
Jacoby, J., & Sassenberg, K. (2011). Interactions do not only tell us when, but can also
tell us how: Testing process hypotheses by interaction. European Journal of Social
Psychology, 41, 180-190.
Jerdee, T. H., & Rosen, B. (1974). Effects of opportunity to communicate and visibility
of individual decisions on behavior in the common interest. Journal of Applied
Psychology, 59, 712-716.
Jones, E. E., & Harris, V. A. (1967). The attribution of attitudes. Journal of
Experimental Social Psychology, 3, 1-24.
Josephs, R. A., Sellers, J. G., Newman, M. L., & Mehta, P. H. (2006). The mismatch
effect: When testosterone and status are at odds. Journal of Personality and Social
Psychology, 90, 999-1013.
Kahneman, D., Knetsch, J. L., & Thaler, R. H. (1991). Anomalies: The endowment
effect, loss aversion, and status quo bias. The Journal of Economic Perspectives, 5,
193-206.
REFERENCES 116
Kavoussi, R., Armstead, P., & Coccaro, E. (1997). The neurobiology of impulsive
aggression. Psychiatric Clinics of North America, 20, 395-403.
Keller, J., & Pfattheicher, S. (2011). Vigilant self-regulation, cues of being watched and
cooperativeness. European Journal of Personality, 25, 363-372.
Keller, J., & Pfattheicher, S. (in press). The Compassion-Hostility-Paradox: The
interplay of vigilant, prevention-focused self-regulation, compassion and hostility.
Personality and Social Psychology Bulletin.
Kelley, H. H., & Grzelak, J. (1972). Conflict between individual and common interest in
an n-person relationship. Journal of Personality and Social Psychology, 21, 190197.
Kerr, B., Godfrey-Smith, P., & Feldman, M. W. (2004). What is altruism? Trends in
Ecology & Evolution, 19, 135-140.
Kerr, N. L., & Kaufman-Gilliland, C. M. (1994). Communication, commitment, and
cooperation in social dilemma. Journal of Personality and Social Psychology, 66,
513-529.
Kessler, T., & Cohrs, J. C. (2008). The evolution of authoritarian processes: Fostering
cooperation in large-scale groups. Group Dynamics: Theory, Research, and
Practice, 12, 73-84.
Khalfa, S., Isabelle, P., Jean-Pierre, B., & Manon, R. (2002). Event-related skin
conductance responses to musical emotions in humans. Neuroscience Letters, 328,
145-149.
Kirschbaum,
C.,
&
Hellhammer,
psychoneuroendocrine
research:
D.
H.
Recent
Psychoneuroendocrinology, 19, 313-333.
(1994).
Salivary
developments
and
cortisol
in
applications.
REFERENCES 117
Kirschbaum, C., Pirke, K. M., & Hellhammer, D. H. (1993). The ‘Trier Social Stress
Test’- A tool for investigating psychobiological stress responses in a laboratory
setting. Neuropsychobiology, 28, 76-81.
Klimesch, W. (1999). EEG alpha and theta oscillations reflect cognitive and memory
performance: A review and analysis. Brain Research Reviews, 29, 169-195.
Klinesmith, J., Kasser, T., & McAndrew, F. T. (2006). Guns, testosterone, and
aggression: An experimental test of a mediational hypothesis. Psychological
Science, 17, 568-571.
Knetsch, J. L. (1989). The endowment effect and evidence of nonreversible indifference
curves. The American Economic Review, 79, 1277-1284.
Knobe, J., & Malle, B. F. (2002). Self and other in the explanation of behavior: 30 years
later. Psychologica Belgica, 42, 113-130.
Knoch, D., & Fehr, E. (2007). Resisting the power of temptations. Annals of the New
York Academy of Sciences, 1104, 123-134.
Knoch, D., Gianotti, L. R., Baumgartner, T., & Fehr, E. (2010). A neural marker of
costly punishment behavior. Psychological Science, 21, 337-342.
Knoch, D., Pascual-Leone, A., Meyer, K., Treyer, V., & Fehr, E. (2006). Diminishing
reciprocal fairness by disrupting the right prefrontal cortex. Science, 314, 829-832.
Knutson, B. (2004). Sweet revenge. Science, 305, 1246-1247.
Kollock, P. (1998). Social dilemmas: The anatomy of cooperation. Annual Review of
Sociology, 24, 183-214.
Krakowski, M. (2003). Violence and serotonin: Influence of impulse control, affect
regulation, and social functioning. The Journal of Neuropsychiatry and Clinical
Neurosciences, 15, 294-305.
REFERENCES 118
Kroneberg, C., & Kalter, F. (2012). Rational choice theory and empirical research:
Methodological and theoretical contributions in Europe. Annual Review of
Sociology, 38, 73-92.
Kruglanski, A. W., Shah, J. Y., Fishbach, A., Friedman, R., Chun, W. Y., & SleethKeppler, D. (2002). A theory of goal systems. Advances in Experimental Social
Psychology, 34, 331-378.
Kruk, M. R., Halasz, J., Meelis, W., & Haller, J. (2004). Fast positive feedback between
the adrenocortical stress response and a brain mechanism involved in aggressive
behavior. Behavioral Neuroscience, 118, 1062-1070.
Kuhl, J., & Kazén, M. (1997). Persönlichkeits-Stil- und Störungs-Inventar (PSSI)
[Personality styles and dysfunction inventory]. Göttingen: Hogrefe.
Kuhlmann, S., & Wolf, O. T. (2006). A non-arousing test situation abolishes the
impairing effects of cortisol on delayed memory retrieval in healthy women.
Neuroscience Letters, 399, 268-272.
Kurzban, R., & DeScioli, P. (2013). Adaptationist punishment in humans. Journal of
Bioeconomics, 15, 269-279.
Kurzban, R., & Houser, D. (2001). Individual differences in cooperation in a circular
public goods game. European Journal of Personality, 15, 38-52.
Kurzban, R., DeScioloi, P., & O’Brien, E. (2007). Audience effects on moralistic
punishment. Evolution and Human Behavior, 28, 75-84.
Krause, N. (2011). Neighborhood conditions and helping behavior in late life. Journal of
Environmental Psychology, 31, 62-69.
Lamba, S., & Mace, R. (2011). Demography and ecology drive variation in cooperation
across human populations. Proceedings of the National Academy of Sciences, 108,
14426-14430.
REFERENCES 119
Latané, B., & Nida, S. (1981). 10 Years of research on group-size and helping.
Psychological Bulletin, 89, 308-324.
Leary, M. R. (1995). Self-presentation: Impression management and interpersonal
behavior. New York: Brown & Benchmark Publishers.
Leary, M. R., & Kowalski, R. M. (1990). Impression management: A literature review
and two-component model. Psychological Bulletin, 107, 34-47.
Lewin, K. (1943). Forces behind food habits and methods of change. Bulletin of the
National Resources Council, 108, 35-65.
MacKinnon, D. P., Krull, J. L., & Lockwood, C. M. (2000). Equivalence of the
mediation, confounding and suppression effect. Prevention Science, 1, 173-181.
Malle, B. F. (1999). How people explain behavior: A new theoretical framework.
Personality and Social Psychology Review, 3, 21-43.
Malle, B. F. (2006). The actor-observer asymmetry in attribution: A (surprising) metaanalysis. Psychological Bulletin, 132, 895-919.
Malle, B. F., & Knobe, J. (1997). The folk concept of intentionality. Journal of
Experimental Social Psychology, 33, 101-121.
Malle, B. F., & Knobe, J. (2001). The distinction between desire and intention: A folkconceptual analysis. In B. F. Malle, L. J. Moses & D. A. Baldwin (Eds.), Intentions
and intentionality: Foundations of social cognition (pp. 45-67). Cambridge, MA:
MIT Press.
Malle, B. F., Knobe, J. M., & Nelson, S. E. (2007). Actor-observer asymmetries in
explanations of behavior: New answers to an old question. Journal of Personality
and Social Psychology, 93, 491-514.
Markus, H. R., & Kitayama, S. (1991). Culture and the self: Implications for cognition,
emotion, and motivation. Psychological Review, 98, 224-253.
REFERENCES 120
Marlowe, F. W., Berbesque, J. C., Barr, A., Barrett, C., Bolyanatz, A., Cardenas, J. C.,
… & Tracer, D.(2008). More ‘altruistic’ punishment in larger societies.
Proceedings of the Royal Society B: Biological Sciences, 275, 587-592.
Marx, K. (1847/2009). Das Elend der Philosophie [The poverty of philosophy]. Zittau:
Bernd Müller Verlag.
Matthews, B. A., & Norris, F. H. (2002). When is believing “seeing”? Hostile attribution
bias as a function of self-reported aggression. Journal of Applied Social
Psychology, 32, 1-31.
Mazur, A., & Booth, A. (1998). Testosterone and dominance in men. Behavioral and
Brain Sciences, 21, 353-397.
McClelland, D. C., Atkinson, J. W., Clark, R. A., & Lowell, E. L. (1953). The
achievement motive. New York: Appleton-Century-Crofts.
McCullough, M. (2008). Beyond revenge. San Francisco: Jossey-Bass.
Mehta, P. H., & Beer, J. (2010). Neural mechanisms of the testosterone-aggression
relation: The role of orbitofrontal cortex. Journal of Cognitive Neuroscience, 22,
2357-2368.
Mehta, P. H., Jones, A. C., & Josephs, R. A. (2008). The social endocrinology of
dominance: Basal testosterone predicts cortisol changes and behavior following
victory and defeat. Journal of Personality and Social Psychology, 94, 1078-1093.
Mehta, P. H., & Josephs, R. A. (2010). Testosterone and cortisol jointly regulate
dominance: Evidence for a dual-hormone hypothesis. Hormones and Behavior, 58,
898-906.
Merton, R. K. (1968). Social theory and social structure. Glencoe, IL: Free Press.
Mifune, N., Hashimoto, H., & Yamagishi, T. (2010). Altruism toward in-group members
as a reputation mechanism. Evolution and Human Behavior, 31, 109-117.
REFERENCES 121
Milinski, M., Semmann, D., & Krambeck, H. (2002). Reputation helps solve the tragedy
of the commons. Nature, 415, 424-426.
Milinski, M., Semmann, D., Krambeck, H., & Marotzke, J. (2006). Stabilizing the
earth’s climate is not a losing game: Supporting evidence from public goods
experiments. Proceedings of the National Academy of Sciences, 103, 3994-3998.
Miller, E. K., & Cohen, J. D. (2001). An integrative theory of prefrontal cortex function.
Annual Review of Neuroscience, 24, 167-202.
Mischel, W. (1961). Delay of gratification, need for achievement and acquiescence in
another culture. Journal of Abnormal and Social Psychology, 62, 543-552.
Mischel, W. (2012). Self-control theory. In P. A. van Lange, A. W. Kruglanski & E. T.
Higgins (Eds.), Handbook of theories of social psychology (Vol. 2, pp. 1-22).
London: Sage Publications.
Molden, D. C., Lee, A. Y., & Higgins, E. T. (2008). Motivations for promotion and
prevention. In J. Y. Shah & W. L. Gardner (Eds.), Handbook of motivation science
(pp. 169-187). New York: Guilford Press.
Moller, A. C., Elliot, A. J. (2006). The 2 × 2 achievement goal framework: An overview
of empirical research. In A. V. Mittel (Ed.), Focus on educational psychology (pp.
307-326). Hauppauge, NY: Nova Science Publishers.
Montoya, E. R., Terburg, D., Bos, P. A., & van Honk, J. (2012). Testosterone, cortisol,
and serotonin as key regulators of social aggression: A review and theoretical
perspective. Motivation and Emotion, 36, 65-73.
Mook, D. G. (1983). In defense of external invalidity. American Psychologist, 38, 379387.
Nash, J. F. (1950). Equilibrium points in n-person games. Proceedings of the National
Academy of Sciences, 36, 48-49.
REFERENCES 122
Nava, F., Caldiroli, E., Premi, S., & Lucchini, A. (2006). Relationship between plasma
cortisol levels, withdrawal symptoms and craving in abstinent and treated heroin
addicts. Journal of Addictive Diseases, 25, 9-16.
Nelissen, R. M. (2008). The price you pay: Cost-dependent reputation effects of altruistic
punishment. Evolution and Human Behavior, 29, 242-248.
Nettle, D., Harper, Z., Kidson, A., Stone, R., Penton-Voak, I. S., & Bateson, M. (2013).
The watching eyes effect in the Dictator Game: It’s not how much you give, it’s
being seen to give something. Evolution and Human Behavior, 43, 35-40.
Nettle, D., Nott, K., & Bateson, M. (2012). ‘Cycle thieves, we are watching you’: Impact
of a simple signage intervention against bicycle theft. PLOS ONE, 7, e51738.
Nisbett, R., & Cohen, D. (1996). Culture of honour: The psychology of violence in the
south. Boulder, CO: Westview Press.
Nisbett, R. E., & Wilson, T. D. (1977). Telling more than we can know: Verbal reports
on mental processes. Psychological Review, 84, 231-259.
Nowak, M. A., & Sigmund, K. (1998). Evolution of indirect reciprocity by image
scoring. Nature, 393, 573-577.
Oda, R., Niwa, Y., Honma, A., & Hiraishi, K. (2011). An eye-like painting enhances the
expectation of a good reputation. Evolution and Human Behavior, 32, 166-171.
Olson, M. (1968). Die Logik des kollektiven Handelns [The logic of collective action].
Tübingen: Mohr.
Orne, M. T. (1962). On the social psychology of the psychological experiment: With
particular reference to demand characteristics and their implications. American
Psychologist, 17, 776-783.
Ostrom, E. (1990). Governing the commons. Cambridge, MA: Cambridge University
Press.
REFERENCES 123
Osumi, T., Nakao, T., Kasuya, Y., Shinoda, J., Yamada, J., & Ohira, H. (2012).
Amygdala dysfunction attenuates frustration-induced aggression in psychopathic
individuals in a non-criminal population. Journal of Affective Disorders, 142, 311338.
Panagopoulos, C. (2014). I’ve got my eyes on you: Implicit social-pressure cues and
prosocial behavior. Political Psychology, 35, 23-33.
Panagopoulos, C. (in press). Watchful eyes: Implicit observability cues and voting.
Evolution and Human Behavior.
Papies, E. K., Barsalou, L. W., & Custers, R. (2012). Mindful attention prevents
mindless impulses. Social Psychological and Personality Science, 3, 291-299.
Peacock, M. S. (2007). The conceptual construction of altruism Ernst Fehr’s
experimental approach to human conduct. Philosophy of the Social Sciences, 37, 323.
Petrie, M., Halliday, T., & Sanders, C. (1991). Peahens prefer peacocks with elaborate
trains. Animal Behaviour, 41, 323-331.
Petty, R. E., Cacioppo, J. T., & Goldman, R. (1981). Personal involvement as a
determinant of argument-based persuasion. Journal of Personality and Social
Psychology, 41, 847-855.
Pfattheicher, S., & Keller, J. (2013). Vigilant self-regulation and costly punishment in
public goods situations. European Journal of Personality, 27, 346-354.
Pfattheicher, S., & Keller, J. (2014). The eye cue manipulation as subtle induction of a
sense of being watched: Empirical validation and theoretical integration.
Manuscript under review.
Pfattheicher, S., & Keller, J. (in press). Towards a biopsychological understanding of
costly punishment: The role of basal cortisol. PLOS ONE.
REFERENCES 124
Pfattheicher, S., Landhäußer, A., & Keller, J. (2014). Punishing for the sake of
reputation? Effects of reputational concerns on costly punishment. Manuscript
under review.
Pfattheicher, S., Landhäußer, A., & Keller, J. (in press). Individual differences in
antisocial punishment in public goods situations: The interplay of cortisol with
testosterone and dominance. Journal of Behavioral Decision Making.
Pfattheicher, S., Sassenrath, C., & Schindler, S. (2014). Feelings for the suffering of
others and the environment: Compassion fosters proenvironmental tendencies.
Manuscript under review.
Piazza, J., & Bering, J. M. (2008). The effects of perceived anonymity on altruistic
punishment. Evolutionary Psychology, 6, 487-501.
Piff, P. K., Stancato, D. M., Côté, S., Mendoza-Denton, R., & Keltner, D. (2012). Higher
social class predicts increased unethical behavior. Proceedings of the National
Academy of Science, 109, 4086-4091.
Pillutla, M. M., & Murnighan, J. K. (1996). Unfairness, anger, and spite: Emotional
rejections of ultimatum offers. Organizational Behavior and Human Decision
Processes, 68, 208-224.
Popper, K. (2002). The logic of scientific discovery. London: Routledge.
Poustka, L., Maras, A., Hohm, E., Fellinger, J., Holtmann, M., Banaschewski, T., ... &
Laucht, M. (2010). Negative association between plasma cortisol levels and
aggression in a high-risk community sample of adolescents. Journal of Neural
Transmission, 117, 621-627.
Powell, K. L., Roberts, G., & Nettle, D. (2012). Eye images increase charitable
donations: Evidence from an opportunistic field experiment in a supermarket.
Ethology, 118, 1096-1101.
REFERENCES 125
Raihani, N. J., & Bshary, R. (2012). A positive effect of flowers rather than eye images
in a large-scale, cross-cultural dictator game. Proceedings of the Royal Society B:
Biological Sciences, 279, 3556-3564.
Raine, A., Dodge, K., Loeber, R., Gatzke‐Kopp, L., Lynam, D., Reynolds, C., ... & Liu,
J. (2006). The Reactive-Proactive Aggression Questionnaire: Differential
correlates of reactive and proactive aggression in adolescent boys. Aggressive
Behavior, 32, 159-171.
Rand, D. G., Greene, J. D., & Nowak, M. A. (2012). Spontaneous giving and calculated
greed. Nature, 489, 427-430.
Raney, A. A. (2009). Punishing media criminals and moral judgment: The impact on
enjoyment. Media Psychology, 7, 145-163.
Rege, M., & Telle, K. (2004). The impact of social approval and framing on cooperation
in public good situations. Journal of Public Economics, 88, 1625-1644.
Reuben, E., & Van Winden, F. (2008). Social ties and coordination on negative
reciprocity: The role of affect. Journal of Public Economics, 92, 34-53.
Richetin, J., & Richardson, D. S. (2008). Automatic processes and individual differences
in aggressive behavior. Aggression and Violent Behavior, 13, 423-430.
Rigdon, M., Ishii, K., Watabe, M., & Kitayama, S. (2009). Minimal social cues in the
dictator game. Journal of Economic Psychology, 30, 358-367.
Roberts, G. (1998). Competitive altruism: From reciprocity to the handicap principle.
Proceeding of the Royal Society, 265, 427-431.
Robinson, M. D., Schmeichel, B. J., & Inzlicht, M. (2010). A cognitive control
perspective of self-control strength and its depletion. Social and Personality
Psychology Compass, 4, 189-200.
REFERENCES 126
Rockenbach, B., & Milinski, M. (2011). To qualify as a social partner, humans hide
severe punishment, although their observed cooperativeness is decisive.
Proceedings of the National Academy of Sciences, 108, 18307-18312.
Rogers, R. W., & Ketchen, C. M. (1979). Effects of anonymity and arousal on
aggression. The Journal of Psychology, 102, 13-19.
Rosenthal, R., & Jacobson, L. (1968). Pygmalion in the classroom. The Urban Review, 3,
16-20.
Ross, L., & Nisbett, R. E. (1991). The person and the situation: Perspectives of social
psychology. New York: Mcgraw-Hill Book Company.
Ross, L. D., Amabile, T. M., & Steinmetz, J. L. (1977). Social roles, social control, and
biases in social-perception processes. Journal of Personality and Social
Psychology, 35, 485-494.
Rotemberg, J. J. (2008). Minimally acceptable altruism and the ultimatum game. Journal
of Economic Behavior & Organization, 66, 457-476.
Rotter, J. B. (1954). Social learning and clinic psychology. Englewood Cliffs, NJ:
Prentice-Hall.
Rubinow, D. R., Roca, C. A., Schmidt, P. J., Danaceau, M. A., Putnam, K., Cizza, G., ...
& Nieman, L. (2005). Testosterone suppression of CRH-stimulated cortisol in
men. Neuropsychopharmacology, 30, 1906-1912.
Rucker, D. D., Preacher, K. J., Tormala, Z. L., & Petty, R. E. (2011). Mediation analysis
in social psychology: Current practices and new recommendations. Social and
Personality Psychology Compass, 5, 359-371.
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,
1755-1758.
REFERENCES 127
Scheier, M. F., Carver, C. S., & Bridges, M. W. (1994). Distinguishing optimism from
neuroticism (and trait anxiety, self-mastery, and self-esteem): A reevaluation of the
Life Orientation Test. Journal of Personality and Social Psychology, 67, 10631076.
Schmidt, B., Mussel, P., & Hewig, J. (2013). I’m too calm – Let’s take a risk! On the
impact of state and trait arousal on risk taking. Psychophysiology, 50, 498-503.
Schmitt, M. J., Baumert, A., Fetchenhauer, D., Gollwitzer, M., Rothmund, T., &
Schlösser, T. (2009). Sensibilität für Ungerechtigkeit [Sensitivity for injustice].
Psychologische Rundschau, 60, 8-22.
Schmitt, M. J., Neumann, R., & Montada, L. (1995). Dispositional sensitivity to befallen
injustice. Social Justice Research, 8, 385-407.
Scholer, A. A., & Higgins, E. T. (2008). Distinguishing levels of approach and
avoidance: An analysis using regulatory focus theory. In A. J. Elliot (Ed.),
Handbook of approach and avoidance motivation (pp. 489-503). New York:
Psychology Press.
Scholer, A., & Higgins, E. T. (2011). Promotion and prevention systems: Regulatory
focus dynamics within self-regulatory hierarchies. In K. D. Vohs & R. F.
Baumeister (Eds.), Self-regulation: Research, theory, and applications (pp. 143161). New York: Guilford Press.
Scholer, A. A., Zou, X., Fujita, K., Stroessner, S. J., & Higgins, E. T. (2010). When risk
seeking becomes a motivational necessity. Journal of Personality and Social
Psychology, 99, 215-231.
Schulz, J. F., Fischbacher, U., Thöni, C., & Utikal, V. (in press). Affect and fairness:
Dictator games under cognitive load. Journal of Economic Psychology.
Schwartz, S. H., & Bilsky, W. (1990). Toward a theory of the universal content and
structure of values: Extensions and cross-cultural replications. Journal of
Personality and Social Psychology, 58, 878-891.
REFERENCES 128
Schwartz, S. H., Cieciuch, J., Vecchione, M., Davidov, E., Fischer, R., Beierlein, C., ...
& Konty, M. (2012). Refining the theory of basic individual values. Journal of
Personality and Social Psychology, 103, 663-688.
Sellers, J. G., Mehl, M. R., & Josephs, R. A. (2007). Hormones and personality:
Testosterone as a marker of individual differences. Journal of Research in
Personality, 41, 126-138.
Selten, R. (1998). Features of experimentally observed bounded rationality. European
Economic Review, 42, 413-436.
Shaw, A., & Santos, L. (2012). Lab support for strong reciprocity is weak: Punishing for
reputation rather than cooperation. Behavioral and Brain Sciences, 35, 39-39.
Shinada, M., & Yamagishi, T. (2007). Bringing back Leviathan in social dilemmas. In A.
Biel, D. Eek, T. Gärling & M. Gustafsson (Eds.), New issues and paradigms in
research on social dilemmas (pp. 93-123). New York: Springer.
Shinada, M., Yamagishi, T., & Ohmura, Y. (2004). False friends are worse than bitter
enemies: “Altruistic” punishment of in-group members. Evolution and Human
Behavior, 25, 379-393.
Simonsohn, U. (2013). Small telescopes: Detectability and the evaluation of replication
results. The Wharton School, University of Pennsylvania. Retrieved December 18,
2013, from http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2259879.
Small, D. A., & Loewenstein, G. (2005). The devil you know: The effects of
identifiability on punishment. Journal of Behavioral Decision Making, 18, 311318.
Smith, E. A., & Bliege Bird, R. (2000). Turtle hunting and tombstone opening: Public
generosity as costly signaling. Evolution and Human Behavior, 21, 245-261.
Smith, E. R., & Mackie, D. M. (2007). Social psychology (3rd ed.). Hove: Psychology
Press/Taylor & Francis.
REFERENCES 129
Smith, M., Edwards, R., & McCann, U. (2007). The effects of sleep deprivation on pain
inhibition and spontaneous pain in women. Sleep, 30, 494-505.
Sober, E., & Wilson, D. S. (1998). Unto others. The evolution and psychology of
unselfish behavior. Cambridge, MA: Harvard University Press.
Sparks, A., & Barclay, P. (2013). Eye images increase generosity, but not for long: The
limited effect of a false cue. Evolution and Human Behavior, 5, 317-322.
Spence, M. (1973). Job market signaling. The Quarterly Journal of Economics, 87, 355374.
Spencer, S. J., Zanna, M. P., & Fong, G. T. (2005). Establishing a causal chain: Why
experiments are often more effective than mediational analyses in examining
psychological processes. Journal of Personality and Social Psychology, 89, 845851.
Srivastava, J., Espinoza, F., & Fedorikhin, A. (2009). Coupling and decoupling of
unfairness and anger in ultimatum bargaining. Journal of Behavioral Decision
Making, 22, 475-489.
Stirrat, M., & Perrett, D. I. (2010). Valid facial cues to cooperation and trust male facial
width and trustworthiness. Psychological Science, 21, 349-354.
Strack, F., & Deutsch, R. (2004). Reflective and impulsive determinants of social
behavior. Personality and Social Psychology Review, 8, 220-247.
Strack, F., & Deutsch, R. (2012). A theory of impulse and reflection. In P. A. van Lange,
A. W. Kruglanski & E. T. Higgins (Eds.), Handbook of theories of social
psychology (Vol. 1, pp. 97-117). London: Sage Publications.
Stucke, T. S., & Baumeister, R. F. (2006). Ego depletion and aggressive behavior: Is the
inhibition of aggression a limited resource? European Journal of Social
Psychology, 36, 1-13.
REFERENCES 130
Sundie, J. M., Kenrick, D. T., Griskevicius, V., Tybur, J. M., Vohs, K. D., & Beal, D. J.
(2011). Peacocks, Porsches, and Thorstein Veblen: Conspicuous consumption as a
sexual signaling system. Journal of Personality and Social Psychology, 100, 664680.
Sutter, M., Kocher, M., & Strauß, S. (2003). Bargaining under time pressure in an
experimental ultimatum game. Economics Letters, 81, 341-347.
Sylwester, K., Herrmann, B., & Bryson, J. J. (2013). Homo homini lupus? Explaining
antisocial punishment. Journal of Neuroscience, Psychology, and Economics, 6,
167-188.
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.
Tane, K., & Takezawa, M. (2011). Perception of human face does not induce
cooperation in darkness. Letters on Evolutionary Behavioral Science, 2, 24-27.
Tennie, C. (2012). Punishing for your own good: The case of reputation-based
cooperation. Behavioral and Brain Sciences, 35, 40-41.
Thomas, W. I., & Thomas, D. S. (1928). The child in America. Behavior problems and
programs. New York: Knopf.
Tops, M., & Boksem, M. A. (2011). Cortisol involvement in mechanisms of behavioral
inhibition. Psychophysiology, 48, 723-732.
Torrubia, R., Avila, C., Moltó, J., & Caseras, X. (2001). The Sensitivity to Punishment
and Sensitivity to Reward Questionnaire (SPSRQ) as a measure of Gray's anxiety
and impulsivity dimensions. Personality and Individual Differences, 31, 837-862.
Trivers, R. L. (1971). The evolution of reciprocal altruism. Quarterly Review of Biology,
46, 35-57.
REFERENCES 131
van Bommel, M., van Prooijen, J. W., Elffers, H., & van Lange, P. A. (in press).
Intervene to be seen: The power of a camera in attenuating the bystander effect.
Social Psychological and Personality Science.
van den Bos, W., Golka, P., Effelsberg, D., & McClure, S. (2013). Pyrrhic victories: The
need for social status drives costly competitive behavior. Frontiers in Neuroscience,
7, 1-11.
van Dijk, E., & Wilke, H. (2000). Decision-induced focusing in social dilemmas: Givesome, keep-some, take-some, and leave-some dilemmas. Journal of Personality
and Social Psychology, 78, 92-104.
van Goozen, S. H. (2005). Hormones and the developmental origins of aggression. New
York: Guilford Press.
van Goozen, S. H., Fairchild, G., & Harold, G. T. (2008). The role of neurobiological
deficits in childhood antisocial behavior. Current Directions in Psychological
Science, 17, 224-228.
van Goozen, S. H., Fairchild, G., Snoek, H., & Harold, G. T. (2007). The evidence for a
neurobiological model of childhood antisocial behavior. Psychological Bulletin,
133, 149-182.
van Honk, J., Harmon‐Jones, E., Morgan, B. E., & Schutter, D. J. (2010). Socially
explosive minds: The triple imbalance hypothesis of reactive aggression. Journal
of Personality, 78, 67-94.
van Honk, J., Tuiten, A., van den Hout, M., Koppeschaar, H., Thijssen, J., de Haan, E.,
& Verbaten, R. (1998). Baseline salivary cortisol levels and preconscious selective
attention for threat: A pilot study. Psychoneuroendocrinology, 23, 741-747.
van Honk, J., Tuiten, A., van den Hout, M., Koppeschaar, H., Thijssen, J., de Haan, E.,
& Verbaten, R. (2000). Conscious and preconscious selective attention to social
REFERENCES 132
threat: Different neuroendocrine response patterns. Psychoneuroendocrinology, 25,
577-591.
van Lange, P. A. M., Balliet, D. P., Parks, C. D., & van Vugt, M. (2014). Social
dilemmas: Understanding human cooperation. Oxford: Oxford University Press.
van Lange, P. A., De Bruin, E., Otten, W., & Joireman, J. A. (1997). Development of
prosocial, individualistic, and competitive orientations: Theory and preliminary
evidence. Journal of Personality and Social Psychology, 73, 733-746.
van Lange, P. A. M., De Cremer, D., van Dijk, E., & van Vugt, M. (2007). Self-interest
and beyond: Basic principles of social interaction. In A. W. Kruglanski & E. T.
Higgins (Eds.), Social psychology: Handbook of basic principles (pp. 540-561).
New York: Guilford Press.
van Lange, P. A. M., Joireman, J., Parks, C. D., & van Dijk, E. (2013). The psychology
of social dilemmas: A review. Organizational Behavior and Human Decision
Processes, 120, 125-141.
van Vugt, M., Griskevicius, V., & Schultz, P. (2014). Naturally green: Harnessing stone
age psychological biases to foster environmental behavior. Social Issues and
Policy Review, 8, 1-32.
van Vugt, M., Roberts, G., & Hardy, C. (2007). Competitive altruism: Development of
reputation-based cooperation in groups. In R. Dunbar & L. Barrett (Eds.),
Handbook of evolutionary psychology (pp. 531-540). Oxford: Oxford University
Press.
van’t Wout, M., Kahn, R. S., Sanfey, A. G., & Aleman, A. (2006). Affective state and
decision-making in the ultimatum game. Experimental Brain Research, 169, 564568.
Viau, V. (2002). Functional cross-talk between the hypothalamic-pituitary-gonadal and
-adrenal axes. Journal of Neuroendocrinology, 14, 506-513.
REFERENCES 133
Vinkers, C. H., Zorn, J. V., Cornelisse, S., Koot, S., Houtepen, L. C., Olivier, B., ... &
Joëls, M. (2013). Time-dependent changes in altruistic punishment following
stress. Psychoneuroendocrinology, 38, 1467-1475.
Vohs, K. D., & Baumeister, R. F. (2004). Unterstanding self-regulation. In R. F.
Baumeister & K. D. Vohs (Eds.), Handbook of self-regulation: Research, theory,
and application (pp. 1-9). New York: Guilford Press.
Vohs, K. D., & Schmeichel, B. J. (2003). Self-regulation and extended now: Controlling
the self alters the subjective experience of time. Journal of Personality and Social
Psychology, 85, 217-230.
von Dawans, B., Kirschbaum, C., & Heinrichs, M. (2011). The Trier Social Stress Test
for Groups (TSST-G): A new research tool for controlled simultaneous social
stress exposure in a group format. Psychoneuroendocrinology, 36, 514-522.
Vul, E., Harris, C., Winkielman, P., & Pashler, H. (2009). Puzzlingly high correlations in
fMRI studies of emotion, personality, and social cognition. Perspectives on
Psychological Science, 4, 274-290.
Wang, C. S., Sivanathan, N., Narayanan, J., Ganegoda, D. B., Bauer, M., Bodenhausen,
G. V., & Murnighan, K. (2011). Retribution and emotional regulation: The effects
of time delay in angry economic interactions. Organizational Behavior and Human
Decision Processes, 116, 46-54.
Weber, M. (1922/1972). Wirtschaft und Gesellschaft [Economy and society] (5th ed.).
Tübingen: Mohr.
Wedekind, C., & Braithwaite, V. A. (2002). The long-term benefits of human generosity
in indirect reciprocity. Current Biology, 12, 1012-1015.
Weesie, J. (1993). Asymmetry and timing in the volunteers dilemma. Journal of Conflict
Resolution, 37, 569-590.
REFERENCES 134
West, S. A., El Mouden, C., & Gardner, A. (2011). Sixteen common misconceptions
about the evolution of cooperation in humans. Evolution and Human Behavior, 32,
231-262.
Wilkowski, B. M., & Robinson, M. D. (2008). The cognitive basis of trait anger and
reactive aggression: An integrative analysis. Personality and Social Psychology
Review, 12, 3-21.
Williams, K. D. (1997). Social ostracism. In R. Kowalski (Ed.), Aversive interpersonal
behaviors (pp. 133-170). New York: Plenum Press.
Williams, K. D. (2007). Ostracism: The kiss of social death. Social and Personality
Psychology Compass, 1, 236-247.
Williams, K. D., Cheung, C. K., & Choi, W. (2000). Cyberostracism: Effects of being
ignored over the internet. Journal of Personality and Social Psychology, 79, 748762.
Windle, M. (1994). Temperamental inhibition and activation: Hormonal and
psychosocial correlates and associated psychiatric disorders. Personality and
Individual Differences, 17, 61-70.
Yamagishi, T. (1986). The provision of a sanctioning system as a public good. Journal of
Personality and Social Psychology, 51, 110-116.
Yurdusev, A. N. (1993). ‘Level of Analysis’ and ‘Unit of Analysis’: A case for
distinction. Millennium-Journal of International Studies, 22, 77-88.
Zahavi, A. (1975). Mate selection: Selection for a handicap. Journal of Theoretical
Biology, 53, 205-214.
Zahavi, A., & Zahavi, A. (1997). The handicap principle: The missing piece of Darwin’s
puzzle. Oxford: Oxford University Press.
REFERENCES 135
Zak, P. J., Kurzban, R., Ahmadi, S., Swerdloff, R. S., Park, J., Efremidze, L., ... &
Matzner, W. (2009). Testosterone administration decreases generosity in the
ultimatum game. PLOS ONE, 4, e8330.
Zaki, J., & Mitchell, J. P. (2013). Intuitive prosociality. Current Directions in
Psychological Science, 22, 466-470.
Zethraeus, N., Kocoska-Maras, L., Ellingsen, T., von Schoultz, B., Hirschberg, A. L., &
Johannesson, M. (2009). A randomized trial of the effect of estrogen and
testosterone on economic behavior. Proceedings of the National Academy of
Sciences, 106, 6535-6538.
Zillig, L. M. P., Hemenover, S. H., & Dienstbier, R. A. (2002). What do we assess when
we assess a Big 5 trait? A content analysis of the affective, behavioral, and
cognitive processes represented in Big 5 personality inventories. Personality and
Social Psychology Bulletin, 28, 847-858.
Zimbardo, P. G. (1969). The human choice: Individuation, reason, and order versus
deindividuation, impulse, and chaos. In W. J. Arnold & D. Levone (Eds.),
Nebraska symposium on motivation (pp. 237-308). Lincoln, NM: University of
Nebraska Press.
ZUSAMMENFASSUNG 136
Zusammenfassung
Negative
Sanktionen,
im
Konkreten
die
Bestrafung
anderer
Personen
im
Gruppenkontext, fördert menschliches Kooperationsverhalten. Diese Dissertation hat das
Ziel, negative Sanktionen im Kontext sozialer Dilemmasituationen zu verstehen und zu
erklären. Im Speziellen wird aus biopsychologischer Sicht eine Analyse hinsichtlich
kostspieliger Bestrafung im Kontext öffentlicher Güter vorgenommen. Dabei stellt diese
Dissertation eine theoretische Einordnung sowie die Zusammenfassung und Diskussion
fünf publizierter oder sich unter Begutachtung befindender Forschungsartikel dar.
Aufbauend auf Befunden aus der Aggressionsforschung wird in einem ersten Artikel
gezeigt, dass kostspieliges Bestrafen unkooperativer Personen positiv mit dem basalen
Kortisol-Level zusammenhängt. Ein zweiter Artikel dokumentiert, dass ein subtiler
sozialer Cue (stilisierte menschliche Augen) dazu führen kann, ein Gefühl des
Beobachtetwerdens herzustellen. Diese Erkenntnis wird in einem dritten Artikel
angewandt,
um
zu
Reputationsbedingungen
untersuchen,
(d.h.
unter
wie
sich
kostspieliges
Beobachtung
und
Bestrafen
unter
unter
öffentlicher
Selbstaufmerksamkeit) verändert. Dabei konnte eine Reduzierung kostspieligen
Bestrafens dokumentiert werden. In einer vierten Forschungsarbeit wird kostspieliges
Bestrafen unkooperativer Personen durch eine basale motivationale Orientierung, den
Prevention-Fokus, vorhergesagt. Ein fünfter Forschungsartikel beschäftigt sich mit der
Erklärung der Bestrafung kooperativer Personen. Es kann gezeigt werden, dass diese Art
der Bestrafung positiv mit dem basalen Testosteron-Level und psychologischem
Dominanzstreben zusammenhängt, wenn ein relativ niedriges Kortisol-Level gegeben
ist. Zusammenfassend tragen diese Dissertation und die einzelnen Forschungsarbeiten zu
einem besseren Verständnis kostspieligen Bestrafens bei.
APPENDIX 137
Appendix
Appendix A: Towards a biopsychological understanding of costly punishment: The role
of basal cortisol (Pfattheicher & Keller, in press)
Appendix B: The eye cue manipulation as subtle induction of a sense of being watched:
Empirical validation and theoretical integration (Pfattheicher & Keller, 2014)
Appendix C: Punishing for the sake of reputation? Effects of reputational concerns on
costly punishment (Pfattheicher, Landhäußer, & Keller, 2014)
Appendix D: Vigilant self-regulation and costly punishment in public goods situations
(Pfattheicher & Keller, 2013)
Appendix E: Individual differences in antisocial punishment in public goods situations:
The interplay of cortisol with testosterone and dominance (Pfattheicher, Landhäußer, &
Keller, in press)
Appendix F: Vigilant self-regulation, cues of being watched and cooperativeness
(Keller & Pfattheicher, 2011)
Appendix G: The Compassion-Hostility-Paradox: The interplay of vigilant, preventionfocused self-regulation, compassion and hostility (Keller & Pfattheicher, in press)
Appendix H: Feelings for the suffering of others and the environment: Compassion
fosters proenvironmental tendencies (Pfattheicher, Sassenrath, & Schindler, 2014)
Aus Urheberrechtsgründen sind die Artikel nicht dieser Fassung beigefügt.