A Conceptual Review of Decision Making in Social Dilemmas: Applying a Logic of Appropriateness
J. Mark Weber
Joseph L. Rotman School of Management
University of Toronto
University of Michigan Business School
David M. Messick
Kellogg School of Management
Despite decades of experimental social dilemma research, “theoretical integration has proven elusive” (Smithson & Foddy, 1999, p. 14). To advance a theory of decision making in social dilemmas, this article provides a conceptual review of the literature that applies a “logic of appropriateness” (March, 1994) framework. The appropriateness framework suggests that people making decisions ask themselves (explicitly or implicitly), “What does a person like me do in a situation like this?” This question identifies 3 significant factors: recognition and classification of the kind of situation encountered, the identity of the individual making the decision, and the application of rules or heuristics in guiding behavioral choice. In contrast with dominant rational choice models, the appropriateness framework proposed accommodates the inherently social nature of social dilemmas, and the role of rule and heuristic based processing. Implications for the interpretation of past findings and the direction of future research are discussed.
Social dilemmas are situations defined by two characteristics: (a) at any given decision point, individuals receive higher payoffs for making selfish choices than they do for making cooperative choices regardless of the choices made by those with whom they interact and (b) everyone involved receives lower payoffs if everyone makes selfish choices than if everyone makes cooperative choices (Dawes, 1980; D. Messick & Brewer, 1983). They are situations “in which individual rationality leads to collective irrationality. That is, individually reasonable behavior leads to a situation in which everyone is worse off than they might have been otherwise” (Kollock, 1998, p. 183).
Social dilemmas are everywhere. It is difficult to imagine a sphere of social life that is not dogged by one kind of social dilemma or another. When the city you live in asks its citizens to conserve water during a long dry summer, they are identifying a “common resource dilemma” (or “commons dilemma”). You get more of what you want (water), by ignoring the conservation request, regardless of what your neighbors do. If everyone ignores the conservation request the water supply may be exhausted, thereby denying everyone access to the resource. Compared with everyone acting selfishly, everyone is better off if everyone complies with the conservation request.
When the local hospital tries to raise funds for a new oncology center, a public goods dilemma arises for individuals. Those who choose not to contribute will likely still get access to the oncology center in the future if they need it—so they maximize their personal payoffs by not contributing. If nobody contributes, there will be no oncology center for anyone. Compared to the consequences of nobody contributing, everyone is better off if everyone makes a contribution.
Personality and Social Psychology Review 2004, Vol. 8, No. 3, 281–307 Copyright © 2004 by Lawrence Erlbaum Associates, Inc.
Earlier versions of these ideas were presented at the Ninth International Social Dilemmas Conference, Chicago, 2001, and the Culture and Negotiations Conference, Evanston, 2001. The comments, cautions, and suggestions of conference participants are gratefully acknowledged.
We are further indebted to Robyn Dawes, Mark Kennedy, Deepak Malhotra, Keith Murnighan, Paul Van Lange, Janine Webb, Arjaan Wit, Diane Mackie, and our two blind reviewers for their helpful comments and direction.
Requests for reprints should be sent to J. Mark Weber, Organizational Behaviour Department, Joseph L. Rotman School of Management, University of Toronto, 105 St. George St., Toronto, ON, Canada M5S 3E6. E-mail: firstname.lastname@example.org .
Encounter a situation that requires social coordination and you usually don’t have to dig too deeply before hitting a social dilemma of some kind. Social psychologists, anthropologists, economists, sociologists, and political scientists alike have demonstrated great interest in understanding when people make cooperative choices rather than selfish choices, why people make the choices they do, and the kinds of interventions that are effective in eliciting more socially advantageous behavior. Experimental social psychologists, economists, and sociologists have been particularly active in this enterprise, and their efforts have been well documented in a number of comprehensive empirical reviews (e.g., Agrawal, 2002; Dawes, 1980; Kollock, 1998; Komorita & Parks, 1996; Kopelman, Weber, & Messick, 2002; Ledyard, 1995; D. Messick & Brewer, 1983; P. Van Lange, Liebrand, Messick, & Wilke, 1992). Despite a large and growing body of empirical studies, “theoretical integration has proven elusive” (Smithson & Foddy, 1999, p. 14). This article offers a step in the development of a theoretical framework for understanding decision processes in social dilemmas.
What we offer here is a review of the experimental literature in light of a conceptual framework built on the foundation of March’s “logic of appropriateness” (March, 1994). We believe the appropriateness framework offers a useful way of understanding social dilemma decisions that directs attention to critical mechanisms common to most dilemmas, and that identifies predictable sources of variation that can account for the heterogeneous findings evident even across similar situations. Before considering what an appropriateness framework has to offer, we should outline the basic features of the prevailing alternative theoretical approach to decision making in social dilemmas.
Expected Utility/Rational Choice
The dominant theoretical framework applied to decision making in social dilemmas, and dilemma-like
situations, has been the expected utility, or “rational choice” model (cf. Ledyard, 1995; Luce & Raiffa,
1957; Pruitt & Kimmel, 1977). Expected utility (EU) and rational choice models presume vigilant, calculating decision makers who assess choice environments with care, determine the probable utility (i.e., payoff) associated with each possible choice, and then choose to maximize their EU. Rational choice models can offer social scientists helpful analytic frameworks (cf.
Murnighan, 1994; Murnighan & Ross, 1999), and they have the advantages of being parsimonious and highly precise in their predictions—two admirable qualities in any theory.
There are several characteristics of EU models that are worthy of note because they characterize the kinds of situations in which such models are most likely to be effective in predicting behavior, and because in so doing they specify circumstances in which they are less likely to be effective. First is the presumption of a relatively conscious and deliberate decision-making process. Second is the presumption that choice is preceded by evaluation and judgment (and not the other way around). Third is the relatively narrow (economic) definition of utility that generally characterizes applications of EU models. Fourth is the relatively “unsocialized” nature of the models; they tend to downplay social influence processes (when they are addressed at all), and when such processes are accounted for the models again presume hyper-rational others.
By noting these conditions, we intend to make it clear that we are not arguing for discarding rational choice theories or models. Indeed, they can be very helpful analytic frames. However, we contend that all four of these characteristics limit the explanatory power of rational choice models when applied to most social dilemmas. We return to these after summarizing the key features of the appropriateness framework.
The Appropriateness Framework
March (1994) argued that decisions are shaped by situational recognition, one’s identity, and the application of rules. Decisions result from people answering for themselves the question, “What does a person like me (identity) do (rules) in a situation like this (recognition)?” (cf. March, 1994; D. M. Messick, 1999). This logic of appropriateness contrasts with the dominant “logic of consequence,” or EU models. The logic of consequence sees decisions as “based on an evaluation of alternatives in terms of their consequences for preferences” (March, 1994, p. 57). D. M. Messick (1999) suggested that the logic of appropriateness may have greater explanatory power in social dilemmas than such expected utility models.
To act, people must answer for themselves the question, “What kind of situation is this?” (D. M. Messick, 1999, p. 13). Answering this question defines the situation. The appropriateness framework presented here suggests that answering this question hinges on recognition—on matching features of the situation encountered to features of other situations that are already (at least partly) understood. Recognition, therefore, is an act of categorization according to event prototypes—“coherent and inter-related sets of characteristics concerning the sort of person who typically features in the event, the typical explanation for the event and so on” (Lalljee, Lamb, & Abelson, 1992, p. 153). The more typical a new setting or experience is of an existing event prototype, the more likely it is to lead the perceiver to a confident conclusion regarding the nature of the situation. Impediments to easy and rapid
categorization—for example, unexpected actors, strange behaviors, uncertainty, attributional ambiguity, or novel contextual information—will make the process of recognition more difficult and costly in terms of cognitive and attentional resources.
Given the social nature of our world, we want to elaborate somewhat further and suggest that bound to the question of the kind of situation “this” is, is another question: “How do other people understand this kind of situation?” In other words, what is the normative context of this situation, and what would others expect me to do in a situation like this one? This is an important addition because, despite their universal influence, norms have received little attention in the dilemmas literature. We return to a more involved consideration of norms when we discuss rules for action. It is, nonetheless, important to note that others’ evident assumptions about a situation can be involved in activating a particular event prototype (e.g., others are running away; people do that when they are afraid, therefore this is a situation in which fear is appropriate). One of the powerful features of the event prototype concept is that explanations for a situation are integral parts of a prototype and, therefore, allow that understanding a situation within a social context—including its taken-for-granted antecedents and causal attributions—can occur on the level of recognition.
Identity is a complex, multifaceted factor in the appropriateness framework. People differ in many systematic ways, and we want to define identity in broad terms. Often social scientists associate identity only with personality factors, and clearly, people do differ along personality dimensions such as self-monitoring (Snyder & Gangestad, 1986), or locus of control (Lefcourt, 1982). However, they also differ in other ways such as their social value orientations (D. M. Messick & McClintock, 1968), and the nature of their personal histories, and personal experiences. Personal histories are significant because they influence the ease with which people facing a new situation find analogues and event prototypes in memory (recognition) that can direct their initial behavioral choices (Bettenhausen & Murnighan, 1985, 1991; Forgas, 1982). Identity also encompasses social identity (Brewer, 1991; D. M. Taylor & Modhaddam, 1994; Turner, Hogg, Oakes, Reicher, & Wetherell, 1987) and cultural influences (Moghaddam, Taylor, & Wright, 1993). Identity is, therefore, for our purposes, an umbrella concept that includes all the idiosyncratic factors that individuals bring with them into a social situation.
Individual differences of these sorts have already been demonstrated to have an impact on behavior in social dilemmas. For example, Kelley and Thibaut (1978) pointed out that the payoff matrix individuals react to in a dilemma experiment can be different from the objective matrix presented by experimenters.
Furthermore, the nature of individuals’ experiences with similar tasks has been found to quickly shape their understandings of what is normative for a new situation and the kinds of choices they make (Bettenhausen & Murnighan, 1985, 1991).
However, consistent with the Gestalt nature of event prototypes, we want to make special mention of how role prototypes (e.g., Buddhist, parent, Democrat, teacher)—normative constellations of qualities, status, behaviors, and values—may act as pivotal identity factors in social dilemmas. When March (1994) referred to identity’s impact on decision making, it is to such socially defined role prototypes that he referred. To our knowledge, there is no experimental work in the social dilemmas literature that takes such complex roles seriously as determinants of choice behavior. Such roles have sometimes been used as features of the context in dilemma experiments—for example, the role of another party in the dilemma (e.g., the identity of a
leader/allocator of rewards and punishments as in Wit & Wilke, 1988)—however they often have not been taken into account as features of the decision maker that might predict choice.
Our use of the termidentity, then, is intended to allow for consideration of socially defined roles and the
various idiosyncratic qualities, traits, and personal characteristics that are resident within individuals.
Rules offer boundedly rational (Simon, 1955) people a way to cope with the potentially overwhelming flow of stimuli to which they are constantly exposed.
Rules simplify behavioral choices by narrowing options. Within the category of rules we include not only explicit and codified guidelines for behavior (e.g., codes of ethics or laws) but also the less visible and explicit influence of social heuristics (e.g., “women and children first;” Allison & Messick, 1990) and habitual rituals (e.g., the equal division of resources; D. M. Messick & Allison, 1993). Gigerenzer and Todd (1999) presented a significant body of evidence that suggests the use of “fast and frugal” heuristics can be adaptive, and that, compared with more complex algorithms, such heuristics perform especially well when people generalize to new data. The very simplicity of such heuristics, they argued, leads to robustness. The appropriateness framework is particularly distinct from rational choice models in its emphasis on decision making as a rule-driven exercise. Indeed, within
the appropriateness framework, utility maximization (especially in narrow economic terms) is only one of many possible decision rules that may apply.
Within the growing social cognition literature, distinctions are often drawn betweenheuristicsandrules (cf. Chaiken & Trope, 1999; Smith & DeCoster, 1999).
Heuristic processing is characterized as a “fast, associative information-processing mode based on low-effort systematic reasoning” (Chaiken & Trope, 1999, p. ix).
Rule-based processing, on the other hand, is often characterized as deliberate, and demanding a higher level of effort. This dichotomous characterization of two approaches to processing information is common and has given rise to a broad collection of “dual-process” theories. What we propose here shares key features with dual-process theories; in particular, we embrace the notion that some circumstances elicit little conscious deliberation, whereas others elicit careful and effortful deliberation.(We speak to this distinction further under the heading Shallow Processing.) However, our use of the term rules has more to do with the content of the rule than the nature of cognitive processing. For example, “age before beauty” is a rule, as we use the term; however its activation and application can be either relatively automatic, or the consequence of careful thought.
In this article we generally refer to the dual levels of processing in terms of automaticity versus deliberation, shallow versus. deep, conscious versus. unconscious, and so forth. Regardless of the level of processing activated or required, rules are a categorizing cognitive shorthand that flow naturally from clear definitions of situations based on event prototypes.
Part of what makes the appropriateness framework more “social” than rational choice models is the assumption that the rules applied to choices will often be a consequence of perceived social norms. In a recent review of the norms literature, Cialdini and his colleagues defined social norms as “understood rules for accepted and expected behavior” (Cialdini, Bator, & Guadagno, 1999, p. 196). They further characterized norms as “rules for behavior” and “guidelines for socially appropriate behavior” (p.195).When people focus their attention on social norms, such norms have been found to be highly predictive of behavioral choice (e.g., Cialdini,Kallgren, & Reno, 1991; Kallgren, Reno, & Cialdini, 2000). Furthermore, long-standing psychological theories assume that, when people are uncertain about appropriate behavior in a social context, they will look to others for clues (e.g., Festinger, 1954).
Contrasting an Appropriateness
Framework With Expected
Utility/Rational Choice Models
We propose that expected utility and rational choice models are most likely to do a good job of predicting decision makers’ choices under three conditions: (a) when the choice environment is less, rather than more, social; (b) when the situation or context makes the economic structure of a decision task particularly salient; and (c) when the context calls for a deliberate, calculating approach to decision making.
Social psychologists have taken ample advantage of the undersocialized nature of rational choice models by chronicling a myriad of circumstances under which such models fail to account for empirical findings.
Generally these findings are of more cooperative behavior than rational choice models would predict. As we detail at length in our review, self-interested, economically rational choices are most likely to occur when social features of the context are downplayed or not present (e.g., when communication is not permitted, social distance is great, interactions are “one shot,” etc.). Furthermore, when people have reason to have a singular focus on economic payoffs either because of their own financial situations or because the amount of money involved is overwhelming, we would expect careful cost-benefit analyses and higher incidence of self-interested behavior.
The greatest single shortcoming of EU models when applied to social dilemmas is that they are insufficiently social to account for the heterodox factors and interactions that drive real, observed behavior. To accommodate the stylized facts of real behavior, rational choice theorists are forced to create new “utility functions” for each deviation from rationality.
The appropriateness framework, in contrast, is explicitly social and designed to accommodate behavior that occurs outside the conditions most favorable to rational choice models. We would expect, therefore, that when the social dimensions of a dilemma are apparent and/or salient, and especially when social features are combined with strong norms or evident event prototypes, the appropriateness framework will offer a more satisfactory account of choice data.
How an Appropriateness Framework
Applies to Social Dilemmas
Figure 1 offers a schematic representation of how an appropriateness framework might depict decision making in social dilemmas. The decision maker views the situation through a lens constituted by the interaction between identity and situational cues.
Figure 1. A schematic representation of the appropriateness framework.
The situation might offer a number of objective cues—who is in-volved, how situational characteristics are configured,where the event is taking place, how it has been labeled by a person in authority (e.g., an experimenter), and so forth. Personal history with similar situations (e.g., “this looks just like the game I played in an experiment last week”), personality traits, and values (e.g., a preference for cooperative behavior) may all affect how the situation is understood (e.g., as a cooperative task or a competitive task). Furthermore, the idiosyncratic motivations of decision makers may affect the situational cues to which they attend, thereby disposing them to situational construals that are favorable to some goal or other that is salient at the moment (e.g., Holmes, 1991;Kruglanski, 1996; Kunda, 1990; Weber, Malhotra, & Murnighan, 2001). For example, an undergraduate with macaroni and cheese in her cupboard and only lint in her wallet may be inclined to attend to characteristics of an experiment that pertain to monetary payoffs.
The figure also suggests that situational cues can influence the aspects of an individual’s identity (e.g., starving student vs. social activist) that affect situational construal and the accessibility of particular event prototypes. This complex interaction between identity and objective situational factors yields an initial definition of the situation.
The definition of the situation is the heart of the appropriateness framework. Is this a cooperative situation or a competitive situation? Is this a group task or an individual task? Is this a game or a problem to be solved? Is this a one-shot dilemma or an iterated dilemma? Is this a dilemma that demands an anonymous or a public choice? The definition of the situation should answer at least some of these questions. The definition of the situation informs the person about thenorms, expectations, rules, learned behaviors, skills, and possible strategies that are relevant. It should be, therefore, the proximal mediator of behavioral choice.
Some situational categorizations will yield a constrained list of possible behaviors, while others may be more ambiguous and consequently elicit a broad array of possible behaviors (cf. Forgas, 1982).
The definition of the situation suggests a choice set.
Choosing among the options, we contend, is a rule-directed exercise. However, many rules can be applied to the same choice. Imagine approaching a queue waiting to purchase tickets for a concert. You see the queue and understand the rules about waiting in lines. One rule is that you could “jump” the line if you have a friend closer to the head of the line. You may scrutinize the queue for friends. Failing to see anyone, you know that the appropriate action is to take your place and wait.
Others are expected to do the same thing. It is the fair thing to do. If someone jumps the line, others have the right to scold or reject the person. The ethics of the line are presumably understood by all. If you are in a great hurry (say the line is to check in for a flight in an airport and you are perilously close to missing your flight) you may ask permission of others to jump the line, and they may accommodate you. However, there is shared knowledge that it is wrong to race to the head of the line and to try to butt in.
Figure 1 introduces the impact of identity again between situational definition and rule selection because identity factors such as personal histories and value preferences are likely to inform the rule-selection process. A Quaker social activist is less likely than a professional boxing promoter, for example, to select “winners keepers losers weepers” as their guiding rule—even though they may note or concede that it is part of the choice set. Identity factors, therefore, help to sort through possible rules to apply to select the one or two that will drive behavioral choice.
Shallow Processing and Rules
Many social psychologists have suggested that human behavior in social settings is often more likely to
be driven by rules of thumb, heuristics, and habit than by deliberate utility-maximization (e.g., Bargh & Chartrand, 1999). Consistent with this argument, a growing body of evidence suggests that theories of human behavior that paint action and choice as considered, rational, deliberate, and strategic overemphasize conscious processes and capacities (e.g., Bargh & Chartrand, 1999; Gigerenzer & Todd, 1999; Gollwitzer, 1999). Indeed, there is a long history of social psychologists acknowledging low-energy, automatic processing as a regular determinant of behavior and behavioral choice (cf. S. E. Taylor, 1998).
Bargh and Chartrand (1999) have characterized the “lion’s share” of human action as nonconscious and automatic. Similarly, Gigerenzer and Todd (1999) provided evidence for the utility of using “fast and frugal heuristics” that do most of the heavy lifting accomplished by much more sophisticated decision-making algorithms. This is an important factor to consider when applying the appropriateness framework to social dilemmas. It suggests that the question “what does a person like me do in a situation like this?” is only consciously articulated in circumstances that evoke attentional and cognitive resources.
If, for example, you are a person who leaves your lights on allthe time and runs your dishwasher daily, you may continue to do so even during an energy crisis without giving your behavior a second thought. This is selfish behavior in a common resource dilemma. You are more likely to articulate the question or one similar to it if something happens to intervene in the schematic represented in Figure 1, thereby making the step problematic rather than habitualor automatic.This could happen in a number of different ways, and the appropriateness framework helps define them. Something could prime your understanding of yourself as an environmentally conscious person (first identity step). You could experi-ence rolling blackouts in your community (situational cues). A pamphlet on strategies for reducing power consumption could be delivered to your door (increasing the array of plausible, “appropriate” behaviors and rules to choose from).Your socially conscious teenage daughter could arrive home from school and brand you a selfish wastrel who thinks of nothing but your own creature comforts while ignoring the plight of others and the good of the community (introducing an identity crisis that may constrain rule selection).
We believe the step in the appropriateness framework that is most likely to involve shallow processing
on average is the rule-selection step. This intuition is built on the logic of pattern-matching theories in social psychology (e.g., Forgas, 1982; Lalljee et al., 1992).
We would expect most clearly categorized patterns to elicit a single “best” (or at least “most accessible”) rule—thereby minimizing the attention demanded by the rule-selection process. This kind of process is also conceptually consistent with the work of cognitive psychologists on behavioral scripts (e.g., Abelson, 1981), and social psychologists working in the area of mindful versus mindless processing (cf. Langer, 1989).
However, there are at least three circumstances under which we think rule selection rises to a level of conscious deliberation.
First, conscious processing of rule selection might be necessary when two or more values within an individual are in conflict and both are salient. In times of resource shortage, for example, an individual may be attuned to the need to conserve the resource for the community and the competing desire to ensure one’s family members are well cared for. In parts of the world where food or potable water can be scarce, one can imagine such intrapersonal value conflicts.
Second, rule selection may not be automatic when people know there are clear social expectations in a given situation, however they do not know what they are. Such situations are likely in cross-cultural contexts or when boundaries of social class are spanned. For example, if you find yourself unexpectedly meeting the Queen or some other head of state, you may well know that there are clear rules and protocols that apply but be uncertain as to what they are.
Finally, if the definition of the situation is vague, ambiguous, or unpleasant, we would expect the rule-selection process to be a deliberate one. In such circumstances, actors may move on to considering the rules that might apply to a number of possible definitions of the situation in the hopes that considering the situation by rule interaction will make the appropriate (or at least comfortable) choice clearer. Research in the area of bystander behavior offers a good example of this. When people in experiments were alone and encountered situational stimuli that suggested their assistance might be needed by another, a high proportion offered assistance, whereas those in the experiments with other participants were significantly less likely to help (cf. Latane & Darley, 1970). When participants were alone, the primary stimulus was another’s apparent need for help. When participants were in groups, the situation added the apparent concern, action, or inaction of others to the mix of stimuli, which, in turn, complicated the processing of what constituted “appropriate” behavior; the situation was more ambiguous.
As noted earlier, by assuming relatively shallow processing under some conditions and more deliberate processing under others, the appropriateness framework presented here is consistent with a growing body of dual-process theories in social psychology (cf. Smith & DeCoster, 1999). We assume that decisionmaking within the appropriateness framework has dual-process characteristics.
Reviewing the Experimental Literature in Light of an Appropriateness
In our selective review of the experimental commons dilemma literature (Kopelman et al., 2002), we
offered a classification of recent findings. Kopelman et al. first identified nine classes of independent variables that covered the majority of the experimental literature—social motives, gender, payoff structure, uncertainty, power and status, group size, communication, causes, and frames. We further categorized each class of independent variables as being either an individual difference variable or a situational factor. Situational factors were subdivided into task structure and perceptual factors. And finally, task structure was subdivided into independent variables related to the decision structure or the social structure of the dilemma. In this article, we use a modification of this classification framework (Figure 2) to guide our review. The primary alterations are the inclusion of personal history and personality under the identity (individual differences) category, the inclusion of “protocol of play” and “response options” under the decision structure category, and the inclusion of “group dynamics” under the social structure category.1
In each section we first review the literature and then address how the appropriateness framework might apply.
1 .One category we have not included under the heading of identity is culture. We acknowledge that this is an important identity-based consideration and that some new work is being done in the area (e.g., Kopelman, 2003; Wade-Benzoni et al., 2002), however we do not review it here because of a relative dearth of empirical research examining this variable in social dilemma settings. However, we hope researchers will take a greater interest in such research in the future. (For a preliminary consideration of possible cultural dynamics in social dilemmas, see Brett & Kopelman, 2004.)
Identity (Individual Differences)
Ordinary personology is inclined to attribute behavior to the enduring characteristics of others (Gilbert,
1998)—be they personality traits, knowledge, skills, or motives. Indeed, there is a history of researchers and theorists explaining social dilemma outcomes in terms of the interaction between parties with cooperative versus competitive dispositions (e.g., Kelley & Stahelski, 1970). Whereas the field of social psychology has placed empirically justified emphasis on the relativeimportance of situational factors over idiosyncratic personal factors (e.g., Ross & Nisbett, 1991), there is a growing appreciation of how individual differences can shape situational construal, and therefore behaviorin social dilemmas (e.g., Kurzban & Houser, 2001; Liebrand, Jansen, Rijken, & Suhre, 1986). In the appropriateness framework, individual differences are identity factors that influence decision making.
Social motives are sometimes referred to as social value orientations or social values. They are people’s relatively stable preferences with respect to their own and others’ outcomes in social dilemmas. Four motives have received greatest attention in the social dilemma literature (McClintock, 1972): individualism, competition, cooperation, and altruism. Individualism is the motivation to maximize one’s own outcomes. Competition is the motivation to maximize one’s outcomes relative to the outcomes of others. Cooperation is the motivation to maximize joint outcomes. And finally, altruism is defined as the motivation to maximize the outcomes of others. Those with individualistic and competitive motives are referred to as being “proself,” whereas those with cooperative and altruistic motives are referred to as being prosocial.
It is not surprising that those with prosocial motives have been found to be more cooperative in social dilemmas, whereas those with proself motives have been found to behave in a more selfish, competitive fashion.
In common resource dilemmas, for example, when there are conditions of scarcity those with proself motives harvest more of a shared resource for themselves than do those with prosocial motives (e.g., Kramer, McClintock, & Messick, 1986; Roch & Samuelson, 1997). Similarly, prosocial individuals are more inclined to commute using public transportation and express concern regarding their impact on the environment than are proself individuals (Van Vugt, Meertens, & Van Lange, 1995; Van Vugt, Van Lange, & Meertens, 1996).
As the appropriateness framework suggests, people’s social motives (identity factor) can play a role in construing what is appropriate in a given social context. For example, researchers have found that although dependence on another party is predictive of a willingness to make sacrifices in relation to that party, it is more predictive of sacrifices by proself individuals than it is of sacrifices by prosocial individuals (Van Lange, Agnew, Harinck, & Steemers, 1997). Prosocial individuals seem more inclined to make such sacrifices regardless of dependence and relational commitments.
In other words, prosocial individuals are less likely to qualify their willingness to behave in other-regarding ways according to their investment or interest in a particular relationship. Similarly, Samuelson (1993) reported that competitive (proself) individuals tend to reject the appointment of superordinate authorities to help manage common resources regardless of the state of the resource, whereas cooperative (prosocial) individuals are willing to appoint such authorities in cases of extreme overuse.
A striking example of how social motives influence the definition of dilemma situations is the “might versus morality” effect. Liebrand and his colleagues
(Liebrand et al., 1986) studied how individuals with different social motives interpreted cooperative and competitive behavior. People with prosocial motives tended to define behaviors in moral terms (right or wrong), whereas those with proself motives defined behaviors in terms of “might,” or “what works.”
Prosocial individuals define rational action from the Figure 2. Classification of the social dilemma literature.
perspective of the collective, that is, whatever is in the interest of the group constitutes rational action. Proself individuals, on the other hand, define rational action more egocentrically, that is, if it improves personal outcomes it constitutes rational action.
Liebrand et al.’s (1986) initial finding provoked a wave of further investigation. Prosocial individuals are more likely to attribute cooperative behavior to intelligence, whereas proself individuals are more likely to attribute such behavior to a lack of intelligence (Van Lange, Liebrand, & Kuhlman, 1990). Furthermore, prosocial individuals were found to expect more cooperation from honest others than dishonest others, whereas proself individuals made less qualified predictions based on this characteristic (Van Lange & Kuhlman, 1994). Meanwhile, proself individuals anticipated greater cooperation from unintelligent others than from intelligent others, and prosocial individuals did not distinguish their predictions based on others’ intelligence.
The original might versus morality study (Liebrand et al., 1986) offered a strong endorsement of the need to consider identity factors such as social motives. Similar to Kelley and Stahelski (1970), they found that the behavior of cooperatively motivated individuals was behaviorally assimilated by those who behaved competitively. The desire to avoid being taken advantage of over whelmed prosocial motives. This demonstrates how important it can be to take account of identity dynamics; if researchers were to focus exclusively on ultimate outcomes, they might miss important identity factors affecting how those outcomes come about—and how different parties might experience them.
There is some evidence to suggest that prosocial behavior may be partly explained by individuals’ tendencies to self-monitor. Strong cooperators, for example, have been found to score highly on measures of self-monitoring (Kurzban & Houser, in press). Furthermore, a number of studies have found that high self-monitors cooperate more in social dilemmas than low self-monitors (Boone, De Brabander, & van Witteloostuijn, 1999; Danheiser & Graziano, 1982; De Cremer, Snyder, & Dewitte, 2001). The presence of other people and the nature of experimental interactions among participants seem to affect high self-monitors’ choices. The prospect of future interaction, for example, tends to increase how much high self-monitors choose to cooperate while leaving the choices of low self-monitors largely unaffected (e.g., Boone et al., 1999; De Cremer et al., 2001). We are not aware of any studies that seek to identify how much overlap there is between distributions of self-monitoring and social motives. Such research might help us better understand the distinctive populations encountered in the study of social dilemmas and perhaps illuminate the contribution that relevant personality factors such as self-monitoring might contribute to understanding social dilemma dynamics.
People seem to be naturally curious about the impact of gender on decision making. Folk theories
abound, usually predicting that women are uniformly and predictably more cooperative than men. There have been a number of studies examining the relationship between gender and behavioral choice in social dilemmas. However, the observed effects have not been strong, and at times they have appeared contradictory—suggesting a complex relationship between gender and behavioral choice.
More women than men exhibit prosocial motives, whereas more men than women exhibit proself motives (Van Lange, De Bruin, Otten, & Joireman, 1997). In a related field, a meta-analysis of gender and competitiveness in negotiations revealed a tendency for women
to be more cooperative than men (Walters, Stuhlmacher, & Meyer, 1998). However, this latter finding was qualified in important ways. The difference between the genders was reduced by constraints on the negotiations (e.g., limiting communication between parties), and women actually exhibited more competitive behaviors when facing a counterpart using a tit-for-tat strategy.
A number of studies have documented qualifications of the oversimplified assumption that women are more cooperative than men. For example, initially significant differences in the amount men and women contribute to the public good have been found to diminish over repeated rounds (Cadsby & Maynes,
1998). Stockard and her colleagues (Stockard, Van de Kragt, & Dodge, 1988) found that gender differences were overwhelmed by other situational factors when it came to predicting behavior, but that women were nonetheless more inclined than men to explain their behavior in cooperative, altruistic, and harmonious terms. In one study of single sex, four-person groups, male groups were found to make contributions to public goods at higher rates than female groups (Brown Kruse & Hummels, 1993), whereas another study found that all-female groups were more cooperative than all-male groups or mixed-gender groups (Nowell & Tinkler, 1994).
Using a prisoner’s dilemma paradigm, Orbell, Dawes, and Schwartz-Shea (1994) discovered a pattern of results that provides evidence of people’s general, naïve assumptions about the gender effect, and the complexity of determining gender’s real impact. First,women as a group were trusted to cooperate more than men as a group. Yet when given the option of playing or not playing with particular others, this group-level assessment of the genders’ respective trustworthiness was not reflected in individuals’ judgments of the likelihood that particular men or women would cooperate.
And finally, perhaps ironically, there were no differences in cooperation rates between the men and women in their study.
If there is indeed a gender effect, it would seem to be rather weak in most social dilemma settings, and a good deal less predictive of behavior than other identity factors such as social motives and personal histories and experiences. Perhaps the most important thing to note about gender is that although it may not predict behavior in clear and consistent ways across situations, there is a dominant assumption about how it can be expected to affect behavior, and in certain contexts that may affect others’ choices.
Although social learning has been offered as an explanation for various dilemma behaviors (e.g.,
Ledyard, 1995), relatively little attention has been focused on individuals’ idiosyncratic personal histories and experiences as possible determinants of behavior in social dilemmas. Nonetheless, personal history and experiences are clearly among the identity factors that influence how dilemma situations are perceived and understood.
Silverstein and his colleagues (Silverstein, Cross, Brown, & Rachlin, 1998) ran a pair of prisoner’s dilemma experiments in which experimenters “trained” participants in an initial phase. The training first involved uniform instruction regarding payoffs and procedures across conditions. Then, based on random assignment, the experimenter ran participants through practice rounds in which the experimenter played one of four strategies (always defect, always cooperate, random, and tit-for-tat). With the differential training experiences behind them, participants then played a real iterated prisoner’s dilemma with other participants who had been trained using a tit-for-tat strategy. Although the influence of the experimenter’s strategy during training had only a small effect, it is worth noting. Those who had encountered a tit-for-tat strategy in training cooperated slightly more during the real dilemma than those who had encountered the other strategies. Furthermore, when the tit-for-tat participants began playing the real dilemma, they started out cooperating more than their non-tit-for-tat counterparts. Although other structural factors ultimately overwhelmed the training/prior experience effect, this study nonetheless points to how personal history can shape the initial understanding of appropriate behavior in a new situation.
The effect of personal histories and experience in social dilemmas was more compellingly demonstrated in a study of the effects of intragroup challenges to established operating norms (Bettenhausen & Murnighan, 1991). In this study, groups of participants played a disguised iterated prisoner’s dilemma called The Truffle Purchasing Exercise:
Participants played the role of corporate purchasing agents seeking a limited supply of exclusive French truffles that were sold periodically during a short, unpredictable growing season. Only two firms bid for each supply. While each company needed the truffles, their products were different: the companies competed only for the critical truffle supply, not for retail sales. Each offered the single supplier a high or low price in each market period. If both bid high, they split the supply evenly and each made a small profit. If both bid low, they again split the supply but made a larger profit. The largest possible profit went to high bidders when the others bid low; these low bidders earned the lowest possible profit. (Bettenhausen & Murnighan, 1991, pp. 27–28)
Over two separate sessions, the researchers used confederates to help establish, with uniform success, either cooperative or competitive norms. In the third session, new two-person groups were formed without confederates. These groups were either homogeneous (both with cooperative experiences or both with competitive experience) or heterogeneous (one member with cooperative experience and the other with competitive experience). Pairs in which both participants had cooperative experiences behaved more cooperatively than those with heterogeneous experiences, or those in which both participants had competitive experiences. In heterogeneous pairings, a cooperative norm emerged more often than a competitive norm. Analyses indicated that prior experience shaped not only the nature of subsequent behavior but also the nature of expectations about how other parties would behave. The results of this study exemplify the potential of personal histories and experiences to shape understandings of a situation and consequent behavior; prior experience created “separate, opposite norms in groups facing the same task” (Bettenhausen & Murnighan, 1991, p. 33).
Identity and the Definition of the
The appropriateness framework suggests that identity factors play a pivotal role in how the situation is defined. The results presented here suggest that this is particularly true when individuals first encounter a new social dilemma. Individuals’ social motives, personality dispositions, and experiences with similar situations all shape the cues to which they attend, and how they come to understand the situations they face (Kelley & Thibaut, 1978). Rational choice models tend to account for such differences in terms of differential levels of understanding (e.g., cooperative people would behave more competitively if they really under-stood the true nature of the task), or alternatively in terms of differing utility functions. Short of characterizing prosocial people as stupid, it is difficult to account for sustained levels of cooperation by prosocial individuals over time in terms of understanding. Somewhat more plausible is the notion that cooperative people value actions and outcomes differently (utility function) than competitive people.
The appropriateness framework emphasizes the
interaction between identity and situational cues in
the definition of a situation and what, therefore, constitutes the set of possible appropriate actions. The
Bettenhausen and Murnighan truffle study (1991) offers evidence of this interaction and a transition to the
consideration of situational factors. Participants faced
a fourth round during which groups were exposed to
changes in the payoff structure of the exercise that either represented a strong challenge, a moderate challenge, or no challenge to the groups’ prevailing norms
(i.e., a group with cooperative norms might face a
competitive challenge, and vice versa). Whereas the
researchers found that cooperative interpersonal challenges were more persuasive than appeals to competition in norm formation, strong competitive structural
challenges yielded the greatest effects in changing established norms. They concluded that the best account for their data was a joint-effects model with
roughly equal effects of experience and task structure.
The categorical framework applied here (Figure 2)
identifies task structure as a subcategory of situational
factors that is, itself, further subdivided into decision
structure and social structure. The decision structure of
a social dilemma relates to the payoffs associated with
the situation, the number and nature of the choices to
be made, and the level of certainty or uncertainty that
applies. Decision structure is about the task itself. The
social structure, on the other hand, pertains to the social context of the task. There are far too many conceivable social structure factors to enumerate completely.
Instead, we focus on four that have been the subject of
considerable experimental investigation to date: power
and status, communication, group size, and group dynamics. The appropriateness framework views task
structure factors as situational cues that are used by individuals to define the nature of the social dilemma or
situation that they face.
Payoff structure. There is a long history of research regarding the effects of adjustments in the material payoff structure of a social dilemma (see Van
Lange, Liebrand, Messick, & Wilke, 1992, pp. 14–15
for a concise review). Not surprisingly, increasing relative incentives for cooperation generally increases cooperation, whereas increasing relative incentives for
competition generally increases competition (e.g.,
Bell, Petersen, & Hautaluoma, 1989). This is in keeping with rational choice models and popular and psychological understandings of human behavior. That
which is rewarded is more likely to be done; that which
is punished is less likely to be done. Rewards and punishments, however, need not be material in nature, and
this is at least somewhat at odds with the general spirit
of the EU paradigm. For example, in an attempt to understand behavior that seems contrary to game theoretic predictions, Andreoni (1990/1995) suggested that
some people experience a “warm glow” when they cooperate, and that this warm glow has an incentive effect. For this review, we focus primarily on nonmaterial/nonmonetary payoffs and some of the interactions
that occur with structural features of a social dilemma.
Using a public goods dilemma, Gächter and Fehr
(1999) tested whether social rewards alone could overcome free-rider problems. Pretesting revealed that, as
predicted, their participants expected to receive approval from others in proportion to the amount of their
endowments that they contributed to the public good.
In the study proper, participants were assigned to one
of four conditions: (a) an anonymous condition, (b) a
social exchange condition in which participants knew
they would interact after the experimental exercise began, (c) a group identity condition in which participants interacted before the exercise but knew they
would not encounter one another after the game, and
(d) a condition in which participants met before the
game and met after the exercise. Neither the social exchange nor the group identity conditions had cooperation levels greater than the control group (anonymous
condition). However, those who interacted before and
after the exercise (Condition 4) did have significantly
higher levels of contribution than the control group. “If
the social distance between subjects is somewhat reduced by allowing the creation of a group identity and
of forming weak social ties, approval incentives give
rise to a large and significant reduction in free-riding”
Yamagishi’s (1988) study of sanctioning systems in
a public goods dilemma faced by American and Japanese participants offers a revealing set of results from a
logic of appropriateness perspective. When there was
no sanctioning system in a public goods experiment,
American participants cooperated more than their Japanese counterparts. This finding runs counter to commonplace intuitions that collectivist cultures are made
up of individuals who are more motivated to cooperate
than those from individualistic cultures. Yamagishi,
borrowing from Taylor (1976), suggested that the presence of a strong sanctioning system can undermine the
basis for voluntary cooperation. He argued that the reality of pervasive, intensive mutual monitoring and
clear social sanctions in Japan make the Japanese trust
the sanctioning system, rather than the individuals with
whom they interact. This point was bolstered by questionnaire data that revealed a higher level of interpersonal trust among American participants than among
Sometimes the nature of the payoffs themselves can
invoke differential responses. Pillutla and Chen (1999)
found that their participants behaved more competitively when making contribution decisions related to
an economic public good (i.e., a joint investment field)
and more cooperatively when making contribution decisions related to a noneconomic public good (i.e., a
social event). They explain their results in terms of the
implicit norms that each context invokes.
How does an appropriateness framework account
for these payoff results? First, different decision structures appear to invoke different situational definitions.
In the case of Gächter and Fehr (1999), for example,
the procedure of participants meeting before and after
the task seemed to invoke a definition of the situation
as a group task rather than an individual task and lead
to the selection of different, more cooperative, behavioral rules. Their results, however, also support the role
of identity in situational definition; even in the condition in which participants met before and after the task,
a minority of participants behaved competitively and
exploited the endgame.
Yamagishi’s (1988) study also supports the significant role of identity’s interaction with situational cues
to define the situation. In this study, the behavior of
Japanese participants might be well described by a rational choice model; however, the contrasting findings
between Japanese and American participants are intractable using a rational choice framework. The appropriateness framework, however, can potentially account for the observed interaction. An appropriateness
account would suggest that Japanese participants bring
to the situation long histories of relying on sanctioning
systems to create a sense of security in cooperation
(identity) that interact with the lack of such a sanctioning system (situational cue) to define the situation as a
risky one. Seeing the situation as risky, they invoke a
defensive heuristic that, in a public goods dilemma, is
to hold onto their endowments.
Uncertainty. The last 10 years have seen a wave
of research into the major and subtle effects of uncertainty on behavior in social dilemmas. This research has
been conducted not only in psychology but across the
social sciences (e.g., Agrawal, 2002). Even such
long-standing conclusions as the basic efficacy of tit for
tat have been shown to be compromised or qualified by
the presence of uncertainty (e.g., Bendor, Kramer, &
Stout, 1991; Kramer, Wei, & Bendor, 2001).
In commons dilemmas, uncertainty about the size
of a resource has been found to increase (a) harvesting
by individuals, (b) how much others are expected to
harvest, (c) estimates of the size of the common resource pool, and (d) variability in the amount people
harvest (e.g., Budescu, Rapoport, & Suleiman, 1990;
Budescu, Suleiman, & Rapoport, 1995; Gustafsson,
Biel, & Gaerling, 1999). Similarly, when there is uncertainty about the rate at which a common resource
regenerates, harvesting rates go up (Hine & Gifford,
There have been several hypotheses as to why resource uncertainty results in lower levels of cooperation. A substantial program of research by Gustafsson,
Biel, and Gaerling concludes that the most tenable explanation offered to date is a tendency of individuals to
simply be optimistic about the size of resources that are
of value to them (see Gaerling, Gustafsson, & Biel,
1999 for an excellent summary).
However, research by Roch and Samuelson (1997)
points to the role of identity interactions with situational cues. They found that social motives moderated
the effect of uncertainty on harvesting behavior. Specifically, they found that, under conditions of high uncertainty, cooperators harvested less than
noncooperators and held their harvests constant,
whereas noncooperators increased their harvests from
a common resource. The researchers posited that the
effect of social motives was a consequence of “influencing the scanning and processing of goal-relevant
[situational] cues in [the] decision environment” (p.
Van Dijk and his colleagues found that the general
view of uncertainty leading to competitive behavior
was in need of additional important qualification (van
Dijk, Wilke, Wilke, & Metman, 1999). This understanding was based primarily on research regarding behavior in symmetric common resource dilemmas.
Drawing data from two experiments, the researchers
concluded that the effects of uncertainty depend on the
kind of dilemma (in this case common resource vs.
public goods), the nature of asymmetries, and the kind
of uncertainty encountered. In general, people used
certain rather than uncertain information to select their
coordination rule (i.e., equal division, equal proportions, or equal division of outcomes). Consequently,
the authors advise “that in order to predict how uncertainty may affect choice behavior, one should first analyze what information group members use under conditions of environmental certainty” (p. 130).
In related work, Wit and Wilke (1998) found that
environmental uncertainty (varying the size of the
range from which the provision point for a step-level
public good would be selected) led to decreased cooperation under circumstances of high social uncertainty
(the range of most frequently occurring contributions
by earlier participants) but not under circumstances of
low social uncertainty (narrower range). This latter
study highlights how normative social contexts (i.e.,
highly variable or highly constrained histories of
choice by others) can influence decisions in social dilemmas. When people’s previous choices were narrowly defined, the tendency of environmental uncertainty to lead to self-interested behavior was
moderated. This is one example of the kind of finding
for which the appropriateness framework offers a more
compelling account than a traditional rational choice
or EU model.
In light of the appropriateness framework, these
collected findings raise a pair of additional interesting
empirical questions. First, does uncertainty, on its own,
invoke a more conscious and deliberate selection of
rules, or do people simply apply an optimistic estimate
heuristic when the resource is a valued one? Research
by Hsee (1995) suggests that, in general, self-interest
may dominate in circumstances of uncertainty with respect to task-relevant criteria. Furthermore, Gaerling et
al. (1999) noted that it is not possible to infer from their
program of research the degree of intentionality in the
“outcome-desirability bias.” Second, in light of Roch
and Samuelson (1997), is it possible that those with
competitive social motives apply such a rule automatically under conditions of uncertainty, whereas those
with cooperative social motives respond to uncertainty
by becoming deliberate in their processing? Studies involving response-time may help to tease this puzzle
apart in the future.
Protocols of play. Another interesting experimental finding that has attracted research attention has
to do with the temporal order in which people harvest
from a common resource pool. Budescu, Au, and Chen
(1997) called this aspect of the decision problem the
“protocol of play.” One protocol—the standard one in
many early studies —is thesimultaneous protocol.In
the simultaneous protocol, all players make their harvest decisions simultaneously with no knowledge of
the requests of the other players. What the players
know is the pool size, possibly with some uncertainty,
and the number of people with whom the pool will be
shared. A second protocol of play is that the players are
assigned (in some way) sequential positions—first,
second, and so forth—and they make their decisions
sequentially, knowing what position they occupy and
also knowing what the size of the remaining pool is
when they make their harvest decisions. This is thesequential protocol. In the sequential protocol, there is a
clear position effect. The requests of those who come
earlier in the sequence tend to be larger than those of
the players who come later (Budescu, Rapoport, &
Suleiman, 1992; Rapoport, Budescu, & Suleiman,
1993). It is as if there is an advantage associated with
being one of the earlier players to withdraw resources
from the pool. The interpretation of this effect is that
those who play first feel entitled to take more than they
would if they came later. Players who come later must
make allowances for the decisions that were made by
those who came earlier.
An interesting variant of this phenomenon comes
from thepositional protocol. In this protocol, players
are assigned sequential positions in which they make
their decisions, however they have no knowledge of the
size of the remaining pool. In this case, first movers
cannot depend on those who come later to adapt to
larger initial harvests because the magnitude of the
early harvests is not known. This protocol permits
three hypotheses about decision making. First, because
sequential pool-size information is unavailable, there
should be no position effect—the results should look
similar to the simultaneous protocol. Second, if players
all expect the position effect to exist, then they will act
in accordance with it and create the effect, and the results should look similar to the sequential protocol.
There may be ambiguity and uncertainty about the
early players, even for the early players, with some
thinking that the appropriate model is the simultaneous
protocol and others thinking that the appropriate model
is the sequential protocol. Thus, a third hypothesis is
that the results should fall somewhere between the two
pure benchmarks. Budescu and his colleagues
(Budescu et al., 1995; Budescu et al., 1997) confirmed
this third hypothesis.
Finally, Budescu et al. (1997) described acumulative protocolin which a player knows only how much
of the resource remains but does not know his or her serial position. (Because players know the total group
size and the average pool size they can make an inference about their serial position from the pool size.)
With the methodology used in this study, there was little difference between this protocol and the sequential
The interesting empirical riddle posed by research
on protocols of play is this: Why does the positional
protocol, which is formally identical to the simultaneous protocol, show a position effect similar to the sequential protocol? Two suggestions have been offered
(e.g., Budescu et al., 1997). First, the knowledge of position may provide a “coordinating device” that allows
players to share expectations and deal with the dilemma effectively. The second suggestion is that the
effect reflects the operation of a “social decision heuristic” of the sort discussed by Allison and Messick
(1990). A heuristic of this sort is a cognitive rule that is
evoked by a social situation. In the simultaneous protocol, where there is no way to differentiate among players, the heuristic that is likely to be evoked is an “equal-292
WEBER, KOPELMAN, MESSICK
ity” heuristic (Messick & Allison, 1993). This
heuristic requires that one divide the likely pool size by
the number of participants to calculate an equal share.
The sequential protocol may evoke a different heuristic
such as the first-come, first-served rule that governs social behavior in queues. People early in the queue get
better options than people later in the queue. The positional protocol is ambiguous in that it could evoke either or both of these heuristics. If this interpretation is
correct, variables that accentuate one interpretation
over the other should also have an impact on the size of
the position effect with the positional protocol. Some
of the factors that have been shown to influence (at
least the first mover’s) behavior are group size
(Allison, McQueen, & Schaerfl, 1992; Budescu et al.,
1995), the divisibility of the resource (Allison &
Messick, 1990), the reason behind the position assignment (Samuelson & Allison, 1994), the title (leader,
for instance, or supervisor) associated with the first
mover (Samuelson & Allison, 1994), and the degree of
uncertainty about the size of the pool (Budescu et al.,
There is no incompatibility between viewing the
position effect as a coordinating device, on one hand,
and as a social decision heuristic on the other. Many
heuristics do coordinate to the extent that they are
shared. Traffic on the right has priority. Take turns
passing through intersections. These are commonly
held rules or heuristics that effectively coordinate action. Within the appropriateness framework, such
heuristics are the rules that are applied as a result of
the process of defining the situation. The suggestion
that the positional protocol is ambiguous and may
evoke more than one heuristic raises the possibility
that the positional protocol evokes more conscious
and deliberate processing than the sequential or simultaneous protocols. Investigation of this possibility
might help to better understand how automatic or
shallow situational processing and rule selection are
in social dilemmas.
Response options. Other important aspects of a
dilemma’s decision structure are the behavioral
choices available. What options are open to the people facing the dilemma? The vast majority of experimental social dilemma research has been done in
“forced play” contexts. In other words, participants in
social dilemma experiments generally have no choice
but to play with the counterparts they are assigned.
Although such constraints are reflective of the majority of experimental dilemma situations, they represent
the minority of real world dilemmas. As Orbell and
Dawes (1993) noted, “outside of prisons and other total institutions (e.g., mental hospitals, prep schools,
ghettos, and the military), humans usually don’t have
to interact with each other…” (p. 787). Furthermore,
“even the most long-lasting relationships such as
marriage and friendship … entail possibilities of termination by voluntary moves of the people involved”
(Hayashi & Yamagishi, 1998, p. 227). These quotations identify two typical features of most real-world
social dilemmas: People can generally choose
whether to play, and if they play, they usually have
some influence over the selection of counterparts
with whom they interact.
Although a great deal more remains to be done in
this area, recent research offers some insight into the
impact of offering parties to a dilemma a larger palette
of behavioral choices. Drawing on earlier experimental
work using the prisoners’ dilemma paradigm, Orbell
and Dawes (1991) offered a theory-driven account of
the “cooperators’ advantage” in situations in which
parties can choose to play or not play. Because cooperators have higher estimates of the likelihood that others
will cooperate than do noncooperators, cooperators
should be more likely to play in such situations, to encounter other cooperators playing, and therefore reap
superior returns. Noncooperators, on the other hand,
should opt not to play given their pessimistic estimates
of the likelihood that they will encounter cooperation.
Building on their initial model, Orbell and Dawes
(1993) later reported that the freedom to choose
whether to play improves aggregate outcomes for
groups (social welfare) by increasing the frequency of
beneficial cooperate-cooperate pairings. The probability of cooperators encountering one another in voluntary play also improved their outcomes relative to
In the spirit of Axelrod’s (1984) groundbreaking
computer tournaments, Hayashi and Yamagishi (1998)
competitively tested a collection of programmed strategies in a prisoners’ dilemma setting in which parties
could choose with whom they would play. More interesting, strategies that performed well in such an environment demonstrated considerable trust in strangers.
In fact, the winning strategy was unconditionally cooperative—with a selective eye for new partners that had
yielded positive outcomes for it in the past. The winning strategy would never betray a partner, however it
coolly remained focused on its own outcomes without
becoming enmeshed in a single relationship. The general trust demonstrated by the best performing strategies seemed to play “the role of emancipating people
from the confinement of existing relationships by providing booster power for launching them from the security of old relationships into the unknown world of
opportunities” (p. 287).
The research reviewed in this section emphasizes
the role of factors that may predispose people to take
risks in social dilemmas. Orbell and Dawes’ work
(1993) suggests the importance of social motives in
such risk taking. Hayashi and Yamagishi (1998)
pointed to general trust. However, it seems that the
probability of acting on such identity-based predispo-293
DECISION MAKING IN SOCIAL DILEMMAS
sitions to take cooperative risks (e.g., prosocial motives
or general trust) is significantly increased (or in the
case of computer simulations, rewarded) by situational
parameters that minimize the long-term risks. The response options available to participants (or programmed strategies) made it possible to sanction an offending party by withdrawing, thereby minimizing
losses, and possibly even going in search of a new,
more cooperative partner.
Power and status. Under the heading of power
and status, we highlight a handful of interesting findings related to leadership, power asymmetries, and justifications of deviant behavior.
Since the early days of formal social dilemma research, appointing a leader has been offered as a possible solution to social dilemmas (Hardin, 1968). A
significant body of experimental research has since
investigated the circumstances under which leaders
are appointed. For example, people are more likely to
appoint leaders when a common resource is being
overused (D. M. Messick et al., 1983; Rutte & Wilke,
1984), and when managing a common resource is
perceived to be a difficult task (Samuelson, 1991).
This suggests that the appointment of a leader is a solution to a problem that people apply in particular
kinds of situations—when there is a crisis, or when
the difficulty of a challenge makes the possibility of a
crisis seem imminent.
Leaders themselves, however, also constitute situational cues—aspects of the environment worthy of
evaluation and likely to help define the nature of the
situation. Experimental evidence tells us, perhaps not
surprisingly, that followers endorse their leaders when
their leaders are successful (e.g., Wit & Wilke, 1988;
Wit, Wilke, & Van Dijk, 1989). Leaders tend to get
more credit for their ability when they face a predictable environment than when they face an unpredictable
environment, even though their failures are equally
likely to be attributed to their ability (or lack thereof)
regardless of whether the environment is predictable or
unpredictable (Wit et al., 1989).
Wit and Wilke (1990) offerred a provocative set of
findings from the perspective of how identity factors
shape the response to situational cues where leadership
is concerned. In Wit and Wilke’s study, participants
were given the role of chemical company managers
who had to make a choice between storing waste and
treating waste. Storing waste was in the company’s
short-term financial interests, whereas treating waste
was in the long-term interests of the community. As in
the real world, leaders were the source of incentives
and sanctions, and in this particular study, one half the
participants encountered an incentive or sanction system administered by the parent company whereas the
other half encountered a system administered by the
government. For 124 undergraduate students, rewards
and punishments were equally effective regardless of
whether they were administered by the parent company or the government. For 239 real-world managers,
on the other hand, parent company rewards were very
effective in focusing them on long-term considerations, whereas rewards offered by a government
source were actually counterproductive. The experiences people bring to new dilemmas even affect how
they assess equivalent actions coming from different
sources of power or leadership. Findings such as this
one are difficult to accommodate within a rational
choice framework, in which a reward is a reward, regardless of who offers it.
Sometimes the decision people face in a social dilemma is what kind of leader to select. Van Vugt and
De Cremer (1999) demonstrated that the level of group
identification affects the kind of leader deemed most
appropriate, and the differential effectiveness of leader
types. All participants in their study preferred to select
leaders who were legitimate—for example, elected, internal—however this was particularly true when individuals identified strongly with their group. Strong
identification made the democratic selection from
within the appropriate choice. In a second study, the researchers found that when group identification was
low an instrumental leader who would punish free riders was more effective than a leader who instead focused on building positive relationships within the
group. When group identification was high, both kinds
of leaders were equally effective.
In most settings, and especially in organizational
settings, one can assume that there are frequent power
asymmetries in working groups. Building on a stream
of research regarding coalition formation in groups
(Mannix, 1991; Mannix & White, 1992), Mannix
(1993) used an organizational resource distribution
task (a commons dilemma) to examine the impact of
power imbalance versus power balance on coalition
formation and group outcomes. In her study, groups
had five members, the option of forming coalitions,
and were either composed of individuals with balanced
power or individuals with unequal power.
Mannix (1993) found that groups with power imbalances were more likely to start the exercise by forming
coalitions, made less efficient use of the available resources, included fewer people in their resource distributions, and had to expend more effort to reach agreements. In addition, those in groups with a power
imbalance were, understandably, more likely to see the
group as competitive, and more likely to seek to retaliate
against other group members. Mannix explained her results in terms of the difficulties group members have focusing on mutual gains in groups with power imbalances; they instead focus on protecting their own
interests. Relative power imbalance, then, seems to cue
WEBER, KOPELMAN, MESSICK
individuals to define the socialdilemma as a competitive
situation, rather than a group task.
2People may, in general, have experienced such situations of imbalance as
competitive in their pasts and therefore readily invoke
competitive rules based on their personal histories. Indeed, four decades ago Emerson (1962) noted that people experience power imbalances as fundamentally
aversive, and he identified coalition formation as one of
a limited number of strategies available to redress the experience of imbalance.Because coalitions appear inefficient in social dilemmas, Mannix offerred another solution. She suggested that organizations can try to balance
power in groups by selecting people fromthe same organizational or social stratum, thereby reducing the likelihood of individuals feeling threatened by one another
and forming coalitions defensively.
Massey, Freeman, and Zelditch (1997) reported a series of interesting findings concerning the effects of different kinds of justifications in social dilemmas. Justifications are defined as taking responsibility for an act but
denying that the action was wrong or inappropriate. As
one would predict using an appropriateness framework,
justifications can be of pivotalsignificance in howsocial
dilemmas play out. A successful justification involves
accepting that something that seemed a violation of social norms or reasonable expectations was, in fact, acceptable. A justification asks others to redefine their understandings about the norms that govern a given
situation and the rules that should be applied.
The researchers found that offenders who had
higher status than other group members (a Ph.D. in resource management in this case) were judged to have
acted less egregiously when they offered valid justifications, or justifications that were at least ambiguous
in terms of their validity. Status had its greatest positive
impact for those offering ambiguously valid justifications. However, high-status individuals were judged
more harshly than anyone if the justifications of their
offending actions were invalid.
Furthermore, offending individuals with higher
power than their group members were judged publicly
to have acted more appropriately than those with
equivalent power to their group members. However,
power had no salutary effect on private judgments of
behavioral appropriateness. Finally, when offending
individuals were high in status and in power, this had a
salutary effect on public and private judgments of their
behavior’s appropriateness. These findings suggest
that the power and status of actors—even deviant actors—can be powerful situational cues in the definition
of the norms that should apply in a social dilemma.
Furthermore, it is worth noting that power and status
were operationalized relative to the power and status of
those making the judgments, reinforcing the significance of the identity/situation interaction.
To summarize, the power and status literature offers
evidence that is supportive of an appropriateness
framework being applied to decision making in social
dilemmas. The specifics of a situation elicit different
kinds of leadership solutions as a function of social
motives (e.g., Roch & Samuelson, 1997). Situational
factors shape how leadership skills are assessed (e.g.,
Wit et al., 1989), and the kind of leader that people favor (Van Vugt & De Cremer, 1999). The source of
leadership can also shape how different people understand the same situation (e.g., Wit & Wilke, 1990). In
two structurally identical situations, rational choice
models would anticipate identical choices. The appropriateness framework, on the other hand, anticipates
that superficial features of the situation can lead to fundamentally different understandings, and therefore
markedly different choices.
Group size. One of the longest standing assumptions in social dilemma research is that small groups
are better able to establish and sustain cooperation than
larger groups (e.g., Agrawal, 2002; Dawes, 1980).
Only in the past 10 years, however, has compelling
work been done on the mechanisms that might explain
the small group effect. Kerr (1989) hypothesized that
work on self-efficacy could provide a key insight.
Self-efficacy is an individual’s belief in her or his capacity to act effectively to achieve a particular outcome
(Bandura, 1986). The experience of self-efficacy is not
necessarily the same as the objective capacity of an individual to achieve an outcome. In a step-level public
goods dilemma in which contributions to the public
good must reach a certain level for the public good to
be achieved, for example, there is an objectively
discernable capacity for each individual to make a difference. Kerr demonstrated that even when group size
had no objective impact on the ability of individuals to
make a difference, participants still felt more self-efficacious—more able to personally make a difference in
the group’s outcome—when they were in smaller
groups than when they were in larger groups.
In the final experiment in a series of three, Kerr
(1989) measured individuals’ assessments of collective
efficacy, or people’s beliefs that their groups were capable of succeeding at their task. Group size had no effect on such assessments when the provision point
(proportion of group members that needed to contribute to achieve the public good) was high (67%). When
the provision point was lower (33%), however, participants perceived smaller groups to be more efficacious
than large groups. More striking, even when the opposite was objectively true, (i.e., members of larger
DECISION MAKING IN SOCIAL DILEMMAS
An alternative, or complementary interpretation in light of our
discussion of rules is that the application of more cooperative rules
and heuristics (e.g., equal division of resources) is more difficult in
situations with power and status differences. There is a set of potentially contentious contingencies, for example, about how one would
determine equality in such situations.
groups could be more efficacious) participants per-sisted in their beliefs that smaller groups were better.
Consistent with Kerr’s (1989) data, Seijts & Latham
(2000) reported that members of seven-person groups
were more likely to set greedy personal goals than
were members of three-person groups, and similarly
had lower estimates of collective efficacy and lower
outcome expectancies. Seijts, Latham, and Whyte
(2000) found that smaller groups had higher collective
efficacy than larger groups.
More interesting, it was only in his third study that
Kerr (1989) actually found group-size effects on cooperative behavior. In an attempt to make sense of such
complexities, Kerr proposed that smaller groups not
only increase participants’ sense of the potential efficacy of their own contributions but also increase assessments of the potential efficacy of others’ contributions—thereby encouraging free riding. The
temptation to free ride may have been particularly
great in Kerr’s study because the paradigm minimized
the interaction and identifiability of group members.
Kerr’s (1989) explanation for these “illusions of efficacy” is consistentwith the dual-process assumptions of
the appropriateness framework as it is presented here.
He suggested thatthe illusions are attributable to “familiar judgmental heuristics, involving an
overgeneralization of experience in groups of varying
sizes” (p. 287). People, the argumentgoes, generally experience small groups as more effective, and in the absence of an attention-grabbing reason to believe otherwise, they use these experiences to define the situation
they face.In smallgroups,they are generally more likely
to define cooperative goals as attainable. However, the
lack of significant findings in the first two studies, and
Kerr’s proposed explanation, begs an appropriateness
framework–driven question of whether proself individuals understand small groups as prime opportunities for
free riding, whereas prosocial individuals see it as a
more opportune contextin which to make contributions.
It may be that people generally experience members
of small groups to be more accountable to their fellow
group members than those of larger groups. This could
flow from the norms that are assumed to guide people’s
behaviors in such situations. Allison and his colleagues, for example, found that members of small
groups are more motivated to use the equal division of
resources as a distribution rule than are members of
large groups (Allison et al., 1992).
Some experimental economists have disputed the
general assertion that a group’s capacity to achieve the
optimal level of a public good is “inversely related to
group size” (Isaac et al., 1994, p. 1). In fact, in their
studies they found that under certain conditions,
groups of 40 and 100 were able to more efficiently provide a public good than were groups of 4 and 10. How
large and small groups differ appears to be a consequence, at least in part, of “marginal per capita returns”
(MPCR) from the public good (or group account).
“The marginal per capita return from the group account
… is defined as the ratio of benefits to costs for moving
a single token from the individual to the group account” (p. 3). When the MPCR was .30, the larger
groups were more efficient in providing public goods
(higher contributions to the group account) than the
smaller groups. However, when the MPCR was .75,
there was no significant difference in efficiency attributable to group size. Specifically, with a higher MPCR
the contributions made in the smaller groups rose to the
level attained by larger groups under conditions with a
In the face of the failure of standard economic predictions, Isaac et al. (1994) offered “an asymmetric,
forward-looking, non-binary approach” (p. 23) to the
data that focuses on individuals’ hypothesized interests
This approach is composed of three principal components: (1) the assumption that individualibelieves his
decisions have signaling content to others; (2) a
benchmark earnings level for measuring the success
of signaling; and (3) the formulation of a subjective
probability function for evaluating the likelihood of
success. (p. 23)
Assume, for example, that people are interested in
ensuring that at the end of each round they earn as
much as they would have if they had not contributed to
the public good. Isaac et al. (1994) argued that people
then assess the probability that their actions will have
sufficient signaling strength to elicit per capita contributions from others that will result in achievement of
their benchmark for success. If people act in this way,
MPCR has an important and intuitive role to play in
predicting outcomes. Holding MPCR constant, the
larger the group the lower the probability one must assign to each other group member responding to cooperative signals to achieve the personal success benchmark. Furthermore, the lower the MPCR, the higher
the probability one must assign to others’ contributions
to achieve a personal success benchmark. This might
explain why people in large groups would contribute
more with a low MPCR, and why differences between
large and small groups would disappear as MPCR
There is, however, another plausible explanation
worth considering based on the same mathematical
derivations: that people establish a benchmark—or experience a threshold—for reciprocity, rather than for
assessing the success or failure of cooperative signaling. In other words, people may experience a psychological need to reciprocate when they reach a certain
level of enrichment as the result of others’ contributions. Additional research is required to determine
which of these explanations provides a more accurate
WEBER, KOPELMAN, MESSICK
description of people’s decision making. Furthermore,
as the appropriateness framework suggests, people
may differ systematically in the way they arrive at the
same decisions; some may focus on the success of signaling whereas others are more attuned to issues of reciprocity. In either case, the findings demonstrate that
interactions among key situational cues can be as important in the definition of a situation as interactions
between situational and identity factors.
Contrasting Kerr’s (1989) findings with those of
Isaac et al. (1994) highlights a potentially pivotal contingency in assessing the effect of group size on cooperation: the nature of the public good. Kerr’s research
used a step-level public good; everyone was rewarded
if the “provision point” was reached. Isaac and his colleagues, on the other hand, utilized a continuous payoff
paradigm. Given the opposing findings, this distinction
seems significant. It is not clear, for example, how one
would operationalize efficacy (except in the most subjective of ways) in a continuous payoff environment.
Group size is clearly a salient situational cue that
has noteworthy effects on the conclusions people reach
about appropriate behavior in social dilemmas. However, as the appropriateness framework anticipates, it is
also clear that it is a cue that is open to many possible
construals, and that more research into the mechanisms
that drive group size effects is merited.
Communication. Another well-replicated finding in the social dilemmas literature is that communication among participants results in higher levels of cooperation (e.g., Dawes, McTavish, & Shaklee, 1977).
Systematic investigations by many researchers pointed
to the conclusion that all but two explanations for the
communication effect were insufficient: (a) communication enhances group identity and/or solidarity and
(b) communication elicits commitments to cooperate
(Dawes, van de Kragt, & Orbell, 1990).
Kerr and Kaufman-Gilliland (1994) compared
these two explanations. Although groups that communicated with one another about a public good did express a greater sense of group identity, and group identity was found to explain some variation in the results,
the identity explanation was found to be insufficient.
The commitment explanation proved more powerful.
The researchers found that groups that communicated
with one another generated behavioral commitments,
and that most participants followed through with such
Bouas and Komorita (1996) extended these findings. Their results led them to agree that group identity
enhancement was an insufficient explanation, and that
commitments offered a more satisfactory explanatory
mechanism. However, whereas Kerr and
Kaufman-Gilliland (1994) investigated the effect of
universal consensus commitments, Bouas and
Komorita (1996) found that a less restrictive or extreme degree of consensus was also sufficient to elicit
similar cooperative effects.
One of the provocative aspects of Kerr and
Kaufman-Gilliland’s (1994) results was that anonymity regarding individual participants’ contribution decisions had no effect on their decisions. In other words,
the participants whose decisions were hidden from
their fellow group members made the same kinds of
contributions as those whose choices were transparent
to their fellow group members. Did participants in their
anonymous condition really feel anonymous? In the
original study, participants might have believed the experimenters would know whether they had cheated on
their commitments—thereby introducing the experience of social monitoring. If so, then the cooperative
behavior in the anonymous condition might have been
attributable to dominant social norms regarding follow-through on commitments.
In a follow-up study, however, Kerr and his colleagues (Kerr, Garst, Lewandowski, & Harris, 1997)
ensured that those in the anonymous condition felt no
fear of effective monitoring by anyone else; the experimenter’s videotape was publicly mangled beyond repair after communication but before decision making.
Even under such circumstances, people still honored
their commitments. The results of two experiments
converge on the conclusion that those who enter into
cooperative commitments during periods of discussion
seem driven by internalized personal norms—“that
still, small voice”—rather than the fear of reprisal attached to external social norms. (See Murnighan,
Oesch, & Pillutla, 2001, for a theory of self-impression
management based on similar data.)
However, once again there is evidence that identity
factors interact with situational cues in defining appropriate behavior. Whereas the vast majority of participants honored their commitments under conditions of
anonymity, 32% did not. For some individuals, the
anonymous conditions were more likely to help define
the situation as an opportunity for exploitation than the
making of a commitment was to define it as a situation
to be guided by a promise.
Communication studies, by their nature, are highly
social. They are, therefore, precisely the kinds of situations in which we would expect the appropriateness
framework to offer a more compelling account of observed behavior than an EU or rational choice model.
That is clearly the case. How would a rational choice
model account for cooperative behavior in an anonymous interaction with a fixed endpoint? To make sense
of such behavior, a situational definition that invokes
guiding social norms is necessary.
Group dynamics. An increasingly important
stream of experimental research acknowledges that
group associations and identifications (or the absence
of them) are important social features of many social
DECISION MAKING IN SOCIAL DILEMMAS
dilemmas, and that individuals behave differently in
groups than they do alone.
Group identification can be powerful, as demonstrated in an experiment by Kramer and Brewer
(1986). Two groups of three participants were harvesting a common resource. Indication that the resource
was being rapidly depleted by members of an individual’s own group prompted the individual to compensate for the fellow group member by harvesting less. If
the offending individual(s) were members of the other
group, participants increased their harvests.
As Campbell (1965), Sherif (1966), and others suggested decades ago, intergroup competition can improve within-group cooperation. Employing a
within-subjects design, Erev, Bornstein, and Galili
(1993) conducted a field experiment involving high
school boys picking produce (an orange grove dilemma). When the boys were rewarded based on their
personal performance they picked 30% more fruit than
they did when they were rewarded based on their
group’s performance. In conditions of collective rewards, there was a greater free-rider problem. However, the 30% loss of productivity disappeared when
there was between-group competition. Also worth noting was the fact that the more similar the other group
was, the more effective the competition in motivating
In the Erev et al. study (1993), the public good was
conceptualized as endogenous to the group; the behavior of interest was within-group cooperation. What
happens when groups are discrete parties to the same
social dilemma? Insko, Schopler, and their colleagues
invested over a decade in careful exploration of an important and provocative finding using the prisoners’ dilemma paradigm: Groups interacting with other
groups are more competitive than individuals interacting with other individuals (see Schopler & Insko, 1999,
for a concise account of this stream of research). This
oft-replicated finding has been labeled the
“interindividual-intergroup discontinuity effect.”
Other researchers have found similar effects when they
contrast how groups and individuals play “chicken”—
a variant on the Prisoner’s Dilemma Game (Bornstein,
Budescu, & Zamir, 1997).
There are at least three different factors that have
been found to drive the discontinuity effect. The first
two explanations are fear and greed (e.g., Insko,
Schopler, Hoyle, Dardis, & Graetz, 1990). Fear arises
from the assumption that groups are competitive by nature (a rational conclusion given the robustness of the
experimental evidence). Therefore, when people face
groups, they should act defensively or competitively,
which is the same thing in a prisoners’ dilemma (i.e.,
defect). Greed also seems a potent factor. Groups offer
social support/facilitation of individuals’ short-term,
selfish inclinations. Individuals acting alone may feel
more bound to social norms such as reciprocity, in the
absence of the counternormative encouragement and
support of others.
The third explanation for the discontinuity is that individuals feel less identifiable when they act in
groups—and are therefore liberated to act selfishly.
This explanation was supported by the finding that,
when individuals in groups knew their votes regarding
the action the group should take would be identifiable
by the members of the other group, individuals were
more likely to vote for cooperative action (Schopler et
Groups add an important dynamic to social dilemma settings. From the perspective of the appropriateness framework, intergroup activity seems to cue
mechanisms, such as fear and greed, that drive competitive behavior. Furthermore, participation in a group
may also define the situation as one in which self-serving behavior is unlikely to be sanctioned or punished
socially. As Dawes and Messick (2000) noted, the
“vigorous tendency” (p. 114) to support one’s own
group can, in social dilemma situations, result in negative outcomes for not only other groups but also for
one’s own group in the long (or even short) run. Given
the long-term risks, it is important to consider that
Insko and his colleagues (Insko et al., 1998) found that
one reliable way to reduce the discontinuity between
intergroup and interindividual situations was to induce
a collective focus on long-term outcomes, or as
Axelrod (1984) described it, the “shadow of the future”
(p. 126). Within the appropriateness framework, this is
shaping the situational cues that individuals use to define the situation. The shadow of the future may evoke
uncertainties other than the immediate choice of the
External agents often provide people with explanations and labels for the situational cues they encounter
in social dilemmas. People may be told why something
is the way it is (e.g., the salmon stock is close to nil because of gross overfishing). This is a statement regarding causes. People may also simply be given a label for
an aspect of the dilemma (e.g., “this is the altruistic
choice, and that is the selfish choice”). Such labels invoke cognitive frames. There is a growing literature
addressing the impact of manipulating causes and
frames in social dilemmas. This literature is interesting, in part, because it points out how subtle manipulations that have nothing to do with the underlying structure of a social dilemma can affect decision making.
The appropriateness framework suggests that such
subtle manipulations are effective because they tap into
people’s associations with surface features of social
situations, their value orientations, and a pervasive use
of efficient and shallow processing of social stimuli.
WEBER, KOPELMAN, MESSICK
Causes. The reasons given for why things are the
way they are matter in determining choices and outcomes. In a commons dilemma in which group members made serial (rather than simultaneous) harvests,
those who were told they had earned the right to make
the first harvest took significantly more than those who
were simply designated the priority position (Hoffman
& Spitzer, 1985). In similar work, Samuelson and
Allison (1994) varied the reasons participants were
given for being assigned the first position for harvesting from a common resource. Those who were told
they had been assigned the position as a consequence
of a fair-selection mechanism (i.e., a coin toss or the
best score on an achievement test) took almost 50%
more of the common resource than those whose position was assigned as a result of a mechanism that was
deemed to be a poor prototype of a fair mechanism
(i.e., distance of birth date from a randomly selected
date and receiving the easiest version of a selection
Participants can also be expected to infer causes
based on information provided to them. Rutte, Wilke,
and Messick (1987) reported an experiment in which
all participants were told they were the fifth person in a
six-person group to harvest from a common resource.
Each person was presented with the harvesting decisions of the (fictitious) first four group members. Together, the first four took 20 units. One half of the participants were told that there were 35 units initially, and
one half were told that there had been 25 units. Furthermore, in each of these two conditions one half were
told that their fellow group members knew how big the
pool was, and one half were told that they had not
known how big the pool was.
When participants thought the pool size was known
by the previous four harvesters, the available number
of units could be attributed to the greedy (5 points left)
or generous (15 points left) behavior of their group
members. Under these conditions, the cause of the
state of the common resource was attributable to the
other people involved. Where people were seen as the
cause, participants were more greedy when there had
been an apparent norm of greed, and more generous
when there had been an apparent norm of generosity,
than those participants who could only attribute the
state of the resource to forces beyond anyone’s control.
The greedy behavior of others appeared to liberate participants to act greedily (though this meant punishing
an innocent—the sixth person in the group), whereas
the generous behavior of others appeared to constrain
such greedy inclinations and elicit more cooperative
behavior (cf. the consistent contributor effect, Weber,
Solution preferences (rule selection, in appropriateness terms) may also be driven by causal understandings in a social dilemma. As noted in the section on
leadership, Samuelson (1991) found that people who
believed their group had performed poorly in managing a common resource because of the greedy behavior
of group members were more likely to favor appointing a leader than those who believed that the poor performance was due to environmental conditions.
Frames. Framing is a matter of description and
labeling. The interest in framing originated with
Kahneman and Tversky’s (1979) prospect theory.
Prospect theory proposes that people respond differently to outcomes depending on whether they are described as gains or losses. Specifically, people experience the threat of a loss as more serious than an
equivalent gain (hence the idea of loss aversion). The
frame can shift with the reference point used. For example, a $17,000 living stipend for an organizational
behavior graduate student might be seen either as
$2,000 more than students receive in other similar
graduate programs (positive frame) or as $2,000 less
than marketing students at the same institution (negative frame).
To many, the social dilemma literature offered an
obvious example of structurally similar tasks with different labels that correspond with prospect theory’s notions of gain and loss framing. In public goods dilemmas, people must make decisions about making
contributions to the public good (loss of a personal resource). In common resource dilemmas, on the other
hand, people must make decisions about harvesting
from the common resource (gain of personal resources). However, early attempts to study gain/loss
framing effects in social dilemmas failed to yield a
clear and consistent story (cf. Aquino, Steisel, & Kay,
1992; Brewer & Kramer, 1986; De Dreu, Emans, &
Van de Vliert, 1992; McDaniel & Sistrunk, 1991).
More recent research points out that social dilemmas
have many possible reference points, any one of which
may form the basis for individuals’ decision making
(Sonnemans, Schram, & Offerman, 1998). This makes
the single clear reference point required by a test of
prospect theory difficult to identify. The appropriateness framework presented here highlights the fact that
the labels and descriptions used to frame dilemmas
will evoke different interpretations and, hence, different behaviors in otherwise identical situations. A number of studies support this generalization.
For example, De Dreu and McCusker (1997) found
that framing outcomes as gains or losses had different
effects for prosocial individuals and individualists.
Prosocial individuals, motivated to maximize the sum
of outcomes for both parties, were more likely to cooperate in loss frames than in gain frames. Conversely, individualists, motivated to maximize their own outcomes, were more likely to defect in loss frames than in
gain frames. Their respective motivations interacted
with the gain-loss frame.
DECISION MAKING IN SOCIAL DILEMMAS
Allison, Beggan, and Midgely (1996) found that
people perceive social dilemmas quite differently
when they are described using different metaphors.
The metaphors create connotations that lead to different construals of the situation. More concretely, Batson
and Moran (1999) found that cooperation in a prisoner’s dilemma task was greater when the task was labeled as a social exchange study than when the structurally identical task was labeled as a business
transaction study. Batson and Moran referred to this effect as a “business exemption on moral motivation” (p.
912). However, they also found that inducing empathy
for the other party induced increased cooperative
choices in the business and social frames. In fact, in the
high empathy condition, there was no difference in the
proportion of participants opting for cooperative versus competitive choices as a function of task framing.
Thus, broad normative associations with classes of human endeavor can also make framing a powerful determinant of individual decision making.
Elliott and her colleagues (Elliott, Hayward, &
Canon, 1998) demonstrated how recent exposure to differentinstitutionalunderstandings can alter the effectof
frames by simply priming the kinds of normative associations noted in the Batson and Moran (1999) study. Participants read materials thatemphasized either entrepreneurial or cooperative business strategies and were
asked to create examples of the kinds of strategies they
had read about. Then, under the auspices of participating in a separate experiment, the same participants engaged in an iterated public goods dilemma. Those
primed with thinking about business in entrepreneurial
terms cooperated significantly less than those primed
with thinking about business in cooperative terms.
Larrick and Blount (1997) reported that the way an
action is labeled also affects people’s decisions. They
pointed out that the structure of a sequential social dilemma (in this case a sequential commons dilemma)
and an ultimatum bargaining game is identical. However, people typically cooperate more in the social dilemmas than in the ultimatum bargaining games. The
researchers demonstrated that the differences could be
attributed to how the actions were described. In the social dilemma setting, the second participant may claim
what is left after the first participant makes a decision;
in the ultimatum game, the second participant may accept or reject the first mover’s offer. Not only are second movers in social dilemmas more likely to claim
what is left than second movers in ultimatum games are
to accept what is offered, but the claiming language
also elicits higher offers from the first participant in
such social dilemmas.
Another key frame is one of property rights; is one
dealing with one’s own private resource, or a communal resource? van Dijk and Wilke (1997) compared
behavior in a common resource dilemma framework
and a public goods dilemma framework. In the common resource dilemma setting, people drew more
from their personal, partitioned piece of the resource
pool than they did from a common resource pool. The
authors attributed the greater care with a common resource to concern about the others who had similarly
been yoked to the common resource. When the resource was characterized as personal property, there
was no need to be concerned with the impact of one’s
behavior on others. In the case of the public goods
setting, however, the contribution—whether from a
personal or collective source—is directed to a shared
outcome, so awareness of the interests of others is inescapable. Consequently, as the researchers predicted,
personal versus private frames on resources in a public goods dilemma made no difference. An appropriateness account would argue that when framing
draws attention to the interests of others, the situation
is more likely to be defined as one in which norms of
other-regarding concern are experienced as appropriate, and greater cooperation results.
In fact, van Dijk and Wilke (2000) articulated a
similar account of the mechanism that drives framing
effects in social dilemmas and provided empirical evidence that backs it up. They argued that framing manipulations focus people on particular aspects of a
complex problem. Cooperation in resource dilemmas
involves deciding how much to take and how much to
leave, whereas cooperation in public goods dilemmas
involves how much to give and how much to keep.
These verbs may focus people on some heuristic or
rule that determines choices, rather than on the numerical outcomes of the choices. For example, in
public goods dilemmas in which people have different endowments of the relevant resource (i.e., an
asymmetric dilemma), people tend to contribute
equal proportions of their personal resources. They
do not like to feel that they were suckers relative to
other members of their cohorts (cf, Kurzban,
McCabe, Smith, & Wilson, 2001). The notions of
giving and keeping may keep people focused on
tweaking their choices to achieve the status of equivalent contributions. People in resource dilemmas, on
the other hand, seem more focused on achieving
equal final outcomes, that is, having equal sums in
their personal coffers at the end of the day. In the
common resource dilemma, then, the framing of relevant action as taking seems to evoke a focus on what
they get rather than what gets left behind. The researchers tested their hypotheses by creating both
common resource and public goods dilemmas that focused either on what participants got to have themselves (take or keep) and the benefits the collective
enjoyed (leave or give). Calculations were made to
determine whether individuals attempted to achieve
proportionality in contribution, or equality of outcomes. As predicted by van Dijk and Wilke (2000),
the focus induced by the action frames accounted for
WEBER, KOPELMAN, MESSICK
a great deal of the variation between the two kinds of
To this point we have characterized framing as a description of situational characteristics (e.g., outcomes,
actions, etc.) that focuses attention in systematic ways
and, in turn, affects how people facing a dilemma think
about the dilemma and the nature of the problem they
face. Frames can also be indirectly invoked by situational characteristics. Batson and Moran (1999) explicitly labeled their experimental situations as having
to do with business or social exchange—however characteristics of the situation can, of course, invoke the
same frames of reference and attendant heuristics
without explicit labels. Tenbrunsel and Messick (1999)
reported on a series of studies that did just that. The researchers were interested in the effects of economic
sanctions on decision makers facing an environmental
dilemma. Their research suggests that sanctions
changed the way decision makers understood their
problem. For many participants, the presence of sanctions changed the problem from an ethical concern
(e.g., what is our responsibility here?) to a business
concern (e.g., what are the economic costs and benefits
of reducing emissions?).
Participants in this research were given the role of
decision makers at a moderately sized manufacturing
plant that produced toxic gas as a byproduct of its operation. Environmentalists had grown concerned about
the impact of the gas on the ecosystem. The industry
agreed to install scrubbers and run them 80% of the
time to assuage the environmentalists’ concerns and
avoid the probable compliance costs of imposed legislation. However, running the scrubbers had a significant financial cost. Participants in the no-sanction condition were told that there would be no inspections to
ensure compliance. Participants in the weak-sanction
condition were told that there was a 5% chance of being caught if they failed to comply, and the fine would
be negligible ($50,000). Participants in the
strong-sanction condition were told that there was a
greater than 50% chance of being caught if they failed
to comply, and the fine would be significant ($2M).
Complying with the 80% agreement was a cooperative
choice; the costs for the whole industry would increase
dramatically if the majority of plants failed to comply
and the threatened legislation was imposed.
The presence of sanctions evoked a business problem framing. In the absence of sanctions, the problem
was understood in ethical terms, and most participants
cooperated. When the business frame was evoked, the
authors noted that sanctions had to be sufficiently
strong (size of fine), and monitoring sufficiently certain (high probability of being caught), for the economics to come down on the side of the angels. If sanctioning was weak and monitoring uncertain, those viewing
the problem through a business frame were more likely
The framing of social dilemmas can clearly have
significant effects—whether it is a framing of the task
(e.g., business or social exchange), the action (e.g.,
take or keep), or the outcomes (e.g., gain or loss). Furthermore, cognitive frames can be evoked explicitly
with labels and descriptions, however they can also be
primed (e.g., Elliott et al., 1998) or prompted indirectly
by other situational cues (e.g., Tenbrunsel & Messick,
1999). Compared to many other ways to increase cooperation in social dilemmas (e.g., adjusting group size,
controlling the values of people in a group, etc.), framing seems more strategically useful, if only because it
is easier to accomplish. For social scientists, framing
effects offer insight into how the decision process
works. That insight is consistent with the appropriateness framework. The framing effects documented here
demonstrate how subtle changes in the perception of a
situation and its appropriate norms change the way the
situation is understood. They demonstrate how similar
situations can evoke different responses for people
with different social motives (identity). And finally,
they demonstrate how situational definitions impact
The phenomena that we described in this section are
perhaps the most important in highlighting the differences between traditional rational choice models and
the appropriateness framework. The former focuses attention on the underlying economic structure of decision situations; the latter acknowledges that this structure is important but also accepts the importance of
surface features that deal with interpretations, connotations, assumed causal processes, and linguistic variations—variables that create an interpretational stretch
for rational choice theories. A psychology of decision
making that rests on the appropriateness framework
can handle these phenomena in stride whereas an economics of decision making that privileges risks and
consequences does so less gracefully.
Our purpose in writing this article was twofold. First, we have tried to provide an up-to-date, selective review of the experimental literature on social dilemmas. The framework presented in Figure 2 may assist researchers in thinking about the factors that influence people’s behavior in social dilemmas. Second, and more important, we have adapted March’s (1994) logic of appropriateness into a process framework and applied it to the literature to outline a theory of decision making in social dilemmas. We believe the appropriateness framework presented in Figure 1 may integrate some of the mechanisms that underlie decisions in social dilemmas. March (1994) clearly intended his logic of appropriateness ideas to be broadly applicable to decision making in social situations—not only in the context of social dilemmas. The sheer size of the social di-lemma literature offers a unique opportunity to focus on how an appropriateness framework can help explain and advance a particular domain of inquiry. Our hope, therefore, is that by effectively making the case for its efficacy in this domain we can simultaneously make a contribution to our own literature (social dilemmas) and offer an opportunity for others in the broader field of social psychology to adopt and adapt the framework to advance their subfields as well.
It is not our purpose in this article to attack rational choice or expected utility models per se. The point that we want to make about these approaches to the explanation of social behavior is that they do a good job in some contexts and a less good job in others. Our approach parallels that of Fiske (1992) who argued that
there are four elemental forms of social life and that these four forms characterize different domains of our social worlds. One of these forms, which Fiske refers to as “market pricing,” is similar to the consequentialist focus of rational choice theories. Our claim, similar to that of Fiske, is not that market pricing such as rational choice models of social behavior are wrong, just that they have been, in the case of social dilemmas, overgeneralized. The logic of appropriateness that we are advocating is a conceptual perspective that allows us to ask how else social dilemmas are perceived and how these perceptions influence decision behavior.
One of the strengths of the appropriateness framework as it is presented here is that it offers a process model for decision making in social dilemmas and therefore directs attention to opportunities to intervene in the choice process. This opportunity for intervention is important in a field such as social dilemma research in which people are motivated by an interest in encouraging a particular kind of behavior—in this case cooperation and better joint or social outcomes.
The appropriateness framework specifies several key relationships and interactions. It clarifies the interaction of identity factors (broadly defined) and situational cues in jointly determining how individuals define the situations they face. The suggestion that behavior is jointly determined by intra- andextraindividual factors is as old as the field of social psychology itself. However, documenting such interactions is a relatively recent effort among experimental social dilemma researchers, and it is our intention to put that interaction back at the forefront of researchers’ consciousness. As we have argued, and as classic studies have demonstrated (e.g., Liebrand et al., 1986), consideration of situational factors without considering relevant identity factors is likely to miss important, and sometimes critical, dynamics. Furthermore, the field has invested energy in understanding a narrow band of identity variables. The majority of this research has focused on social motives. Additional research investigating identity factors such as self-efficacy (e.g., Kerr & Kaufman-Gilliland, 1997), self-monitoring (e.g., Kurzban & Houser, 2001), and personal histories (e.g., Bettenhausen & Murnighan, 1991) may prove fruitful, as might research into new areas like locus of control or the need to manage self-impressions (Murnighan et al., 2001).
The appropriateness framework (Figure 1) not only proposes that identity factors interact with situational cues to determine an individual’s definition of the situation butalso proposes thatthe situationalcues play a role in eliciting the very identity factors that then interact with the situational cues. For example, features of the situation may elicitmemories of personalexperiences in similar situations thatin turn evoke information relevant to the role played by the individual in those previous situations (e.g., professional or personal).
Considering how the situation elicits identity factors opens the door to another promising way to think about advancing identity research in social dilemmas.
In experimental social dilemma research, identity factors have generally been conceptualized as discrete individual difference variables. Such variables have been documented to be relevant and important. However, when March (1994) wrote about identity, he argued that people have messier, multifaceted, multiple “incompletely integrated” (p. 68) identities. As a close colleague recently pointed out, it seems unlikely that anyone facing a social dilemma and asking the question “what does a person like me do in a situation like this” is thinking “what does a high self-monitoring proself individual like me do in a situation like this?” More plausible is the notion that people ask themselves what a father, or Christian, or lawyer does in “a situation like this.” As March put it, “A decision maker is a parent as well as a police officer, a friend as well as a physician, a lover as well as a woman” (p. 68). Research to date tells us very little about the phenomenology of people in social dilemma experiments. Systematically collected data tends to be limited to reports of the probabilities participants apply to the decision making of their counterparts (e.g., “how likely do you think it is that player 2 will make a contribution in the next round?”). Richer data, when collected, tends to be reported as anecdotal explanations offered or elicited during debriefing. Understanding the complexities of what people actually think, and when they think it in relation to social dilemmas is a wide open arena for future investigation that the ppropriateness framework calls for.
We concur with others (e.g., Bargh & Chartrand, 1999; Gigerenzer & Todd, 1999; Gollwitzer, 1999) that much decision making in everyday social situations involves shallow cognitive processing. For this reason, subtle manipulations such as framing can guide people to different behavioral choices in structurally similarsituations. One of the themes that emerged from our conceptual review is the pivotal role of construal in de termining decisions. Framing highlights particular as-pects of the dilemma. Reduced social distance seems to increase the likelihood that people see a dilemma as a group task rather than an individual task. Social motives research suggests that people’s personal characteristics influence the aspects of a dilemma on which they focus. Influences of this kind can fly under the radar of consciousness, however they can also evoke deliberate processing. One of the underresearched areas
in the social dilemma literature is the degree of energy, attention, and consciousness of processing involved in different situations and for different people—and the implications of different modes on outcomes and the kinds of interventions that successfully evoke conscious consideration.
Situations in which there are nested or competing social dilemmas also merit further consideration. For
example, when a fisherman takes a large harvest from a fish stock, on one hand he may be acting selfishly relative to the large-scale common resource dilemma, while at the same time he sees himself making a contribution to the public good that is his family’s welfare.
Very little has been done to understand the dynamics of such situations (for two noteworthy exceptions see Polzer, Stewart, & Simmons, 1999; and Wit & Kerr, 2002). The appropriateness framework suggests that identity factors, and the particular situational cues that evoke them, should play critical roles in directing attention to the different levels of nested dilemmas.
We want to stress that the appropriateness framework we are offering in this article is more than just a way to view the results of experiments that have been done in the past. We are suggesting this framework as a first approximation to a theory of how decisions are made in social dilemmas. As D. M. Messick (1999) argued, this approach not only allows for predictions and independent variables that would not be obvious from a more traditional decision making standpoint but also has some important implications for the nature of the data that should be observed in appropriately designed experimental studies. Most of the studies we have reviewed in this article do not report data in a way that allows us to evaluate the properties that D. M. Messick (1999) and Tenbrunsel and Messick (1999) outlined.
We mention these properties in the following paragraph to urge researchers to take the appropriateness framework seriously as a description of the decision-making process and to design studies that can assess its validity.
The basic ideas are these. If a situational construal or recognition is causally central (the proximal mediator)to judgments about a dilemma situation and the decisions that are made in it, and if different construals are available to decision makers, then the decision variable and the related judgments should be bi- or tri-modal, depending on the numbers of available interpretations there are. This “clumping” should characterize not only the decision variable,for example,how much to contribute to a public good, but also related variables, for example, how much one expects others to contribute.
Furthermore, the variables should be correlated, an implication that has been confirmed many times, for instance, with choice and expectation in social dilemmas.
Finally, in the simplest version of the appropriateness approach, the situational interpretation causes the decision and the related judgments, implying that there should be complete mediation of the decision with any of the related judgments.In other words,the only impact of independent variables is to change the likelihood of one situational interpretation relative to another.
Tenbrunseland Messick (1999) found supportfor such anotion in their experiment. Ultimately, the value of theframework that we are suggesting will be measured in terms of clustering,correlation,and mediation as wellas the creativity of the predictions that it can support.
Our approach, similar to Fiske’s (1992) and many theories that strive to broaden the understanding of an empirical domain, is vulnerable to the charge that it cannot be disproven, that it is invulnerable to falsifiability.
In addition, if a theory cannot be embarrassed by data, then it cannot be taken seriously as an explanation. We do notbelieve thatthe appropriateness framework is immune to embarrassment by data. D. M. Messick (1999) spelled out some empirical implications of the processes that we have been describing more fully here and summarize briefly in the preceding paragraph. If subsequent research that carefully looks for evidence of multimodality, clustering, and mediation fails to find such evidence, then the conclusion is that our proposal in this article is incorrect. However, we will then be left with the need to find alternative explanations for the results reported by Tenbrunsel & Messick (1999), or by Weber (2003) who has data indicating that consistent contributors in multiparty social dilemmas can earn more, on average, than people in groups without consistent contributors because they encourage others in their own groups to cooperate more (i.e., define the situation differently) than they would otherwise.Findings such as these, and many others, derail rational choice theories and yetare relatively simple to explain within the appropriateness framework. There are many details of how these decision processes work that are not yet understood, however the framework that we are offering here is a step in the direction of achieving a more complete understanding.
Abelson, R. P. (1981). Psychological status of the script concept. American Psychologist, 36, 715–729.
Agrawal, A. (2002). Common resources and institutional sustainability. In E. Ostrom, T. Dietz, N. Dolsak, P. C. Stern, S.
Stonich, & E. U. Weber (Eds.),The drama of the commons(pp. 41-86). Washington, DC: National Academy Press.
Allison, S. T., Beggan, J. K., & Midgley, E. H. (1996). The quest for “similar instances” and “simultaneous possibilities”: Metaphors in social dilemma research.Journal of Personality and Social Psychology, 71(3), 479–497.
Allison, S. T., McQueen, L. R., & Schaerfl, L. M. (1992). Social decision making processes and the equal partitionment of shared resources. Journal of Experimental Social Psychology, 28(1), 23–42.
Allison, S. T., & Messick, D. M. (1990). Social decision heuristics in the use of shared resources. Journal of Behavioral Decision Making, 3, 195–204.
Andreoni, J. (1990). Impure altruismand donations to public goods:A theory of warmglowgiving.Economic Journal,100,464–477.
Andreoni, J. (1995). Warm glow versus cold pickle: The effect of positive and negative framing on cooperation in experiments. Quarterly Journal of Economics,110, 1–21.
Aquino, K., Steisel, V., & Kay, A. (1992). The effects of resource distribution, voice, and decision framing on the provision of public goods.Journal of Conflict Resolution, 36(4), 665–687.
Axelrod, R. (1984).The evolution of cooperation. New York: Basic Books.
Bandura, A. (1986).Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice Hall.
Bargh, J. A., & Chartrand, T. L. (1999). The unbearable automaticity of being.American Psychologist, 54(7), 462–479.
Batson, C. D., & Moran, T. (1999). Empathy-induced altruism in a prisoner’s dilemma.European Journal of Social Psychology, 29(7), 909–924.
Bell, P. A., Petersen, T. R., & Hautaluoma, J. E. (1989). The effect of punishment probability on overconsumption and stealing in a simulated commons. Journal of Applied Social Psychology, 19(17, Pt 1), 1483–1495.
Bendor, J., Kramer, R. M., & Stout, S. (1991). When in doubt . . .: Cooperation in a noisy prisoner’s dilemma.Journal of Conflict Resolution, 35(4), 691–719.
Bettenhausen, K., & Murnighan, J. K. (1985). The emergence of norms in competitive decision-making groups.Administrative Science Quarterly, 30, 350–372.
Bettenhausen, K. L., & Murnighan, J. K. (1991). The development of an intragroup norm and the effects of interpersonal and structural challenges.Administrative Science Quarterly, 36, 20–35.
Boone, C., De Brabander, B., & van Witteloostuijn, A. (1999). The impact of personality on behavior in five prisoner’s dilemma games.Journal of Economic Psychology, 20(3), 343–377.
Bornstein, G., Budescu, D., & Zamir, S. (1997). Cooperation in intergroup, N-person, and two-person games of chicken.Journal of Conflict Resolution, 41(3), 384–406.
Bouas, K. S., & Komorita, S. S. (1996). Group discussion and cooperation in social dilemmas.Personality and Social Psychology Bulletin, 22(11), 1144–1150.
Brett, J. M., & Kopelman, S. (2004). Cross-cultural perspectives on cooperation in social dilemmas. In M. J. Gelfand & J. M. Brett (Eds.),The handbook of negotiation and culture(pp. 395–411). Stanford, CA: Stanford University Press.
Brewer, M. B. (1991). The social self: On being the same and different at the same time.Personality and Social Psychology Bulletin, 17, 475–482.
Brewer, M. B., & Kramer, R. M. (1986). Choice behavior in social dilemmas: Effects of social identity, group size, and decision framing.Journal of Personality and Social Psychology, 50(3), 543–549.
Brown Kruse, J., & Hummels, D. (1993). Gender effects in laboratory public goods contribution: Do individuals put their money where their mouth is?Journal of Economic Behavior and Organization, 22(3), 255–267.
Budescu, D. V., Au, W. T., & Chen, X.-P. (1997). Effects of protocol of play and social orientation on behavior in sequential resource dilemmas.Organizational Behavior and Human Decision Processes, 69(3), 179–193.
Budescu, D. V., Rapoport, A., & Suleiman, R. (1990). Resource dilemmas with environmental uncertainty and asymmetric players.European Journal of Social Psychology, 20(6), 475–487.
Budescu, D. V., Rapoport, A., & Suleiman, R. (1992). Simultaneous vs. sequential requests in resource dilemmas with incomplete information.Acta Psychologica, 80(1-3), 297–310.
Budescu, D. V., Suleiman, R., & Rapoport, A. (1995). Positional and group size effects in resource dilemmas with uncertain resources.Organizational Behavior and Human Decision Processes, 61(3), 225–238.
Cadsby, C. B., & Maynes, E. (1998). Gender and free riding in a threshold public goods game: Experimental evidence. Journal of Economic Behavior and Organization, 34(4), 603–620.
Campbell, D. T. (1965). Ethnocentrism and other altruistic motives. In D. Levine (Ed.), The Nebraska Symposium on Motivation (pp. 283–311). Lincoln: University of Nebraska Press.
Chaiken, S., & Trope, Y. (1999).Dual-process theories in social psychology. New York: Guilford.
Cialdini, R. B., Bator, R. J., & Guadagno, R. E. (1999). Normative influences in organizations. In L. L. Thompson, J. M. Levine, & D. M. Messick (Eds.),Shared cognition in organizations: The management of knowledge(pp. 195–211). Mahwah, NJ: Lawrence Erlbaum Associates, Inc.
Cialdini, R. B., Kallgren, C. A., & Reno, R. R. (1991). A focus theory of normative conduct: A theoretical refinement and reevaluation of the role of norms in human behavior.Advances in Experimental Social Psychology, 24, 201–234.
Danheiser, P. R., & Graziano, W. G. (1982). Self-monitoring and cooperation as a self-presentational strategy.Journal of Personality and Social Psychology, 42(3), 497–505.
Dawes, R. M. (1980). Social dilemmas.Annual Review of Psychology, 31, 169–193.
Dawes, R. M., McTavish, J., & Shaklee, H. (1977). Behavior, communication, and assumptions about other people’s behavior in a commons dilemma situation.Journal of Personality and Social Psychology, 35, 1–11.
Dawes, R. M., & Messick, D. M. (2000). Social dilemmas.International Journal of Psychology, 35(2), 111–116.
Dawes, R. M., van de Kragt, A. J. C., & Orbell, J. M. (1990). Cooperation for the benefit of us—Not me, or my conscience. In J. J. Mansbridge (Ed.),Beyond self-interest(pp. 97–110). Chicago: University of Chicago Press.
De Cremer, D., Snyder, M., & Dewitte, S. (2001). “The less I trust, the less I contribute (or not)?” The effects of trust, accountability and self-monitoring in social dilemmas. European Journal of Social Psychology, 31(1), 93–107.
De Dreu, C. K., Emans, B. J., & Van de Vliert, E. (1992). Frames of reference and cooperative social decision-making. European Journal of Social Psychology, 22(3), 297–302.
De Dreu, C. K. W., & McCusker, C. (1997). Gain-loss frames and cooperation in two-person social dilemmas: A transformational analysis.Journal of Personality and Social Psychology, 72(5), 1093–1106.
Elliott, C. S., Hayward, D. M., & Canon, S. (1998). Institutional framing: Some experimental evidence. Journal of Economic Behavior and Organization, 35(4), 455–464.
Emerson, R. M. (1962). Power-dependence relations.American Sociological Review, 27(1), 31–41.
Erev, I., Bornstein, G., & Galili, R. (1993). Constructive intergroup competition as a solution to the free rider problem: A field experiment. Journal of Experimental Social Psychology, 29(6), 463–478.
Fehr, E., & Gaechter, S. (2000). Fairness and retaliation: The economics of reciprocity.Journal of Economic Perspectives, 14, 159–181.
Festinger, L. (1954). A theory of social comparison processes.Human Relations, 7, 117–140.
Fiske, A. P. (1992). The four elementary forms of sociality: Framework for a unified theory of social relations.Psychological Review, 99(4), 689–723.
Forgas, J. P. (1982). Episode cognition: Internal representations of interaction routines. In L. Berkowitz (Ed.),Advances in experimental social psychology 15(pp. 59–103). New York: Academic.
Gächter, S., & Fehr, E. (1999). Collective action as a social exchange. Journal of Economic Behavior and Organization, 39(4), 341–369.
Gaerling, T., Gustafsson, M., & Biel, A. (1999). Managing uncertain common resources. In M. S. M. Foddy (Ed.),Resolving social dilemmas: Dynamic, structural, and intergroup aspects(pp. 219-225). Philadelphia: Psychology Press/Taylor & Francis.
Gigerenzer, G., & Todd, P. M. (Eds.). (1999).Simple heuristics that make us smart.New York: Oxford University Press.
Gilbert, D. T. (1998). Ordinary personology. In D. T. Gilbert, S. T. Fiske, & G. Lindzey (Eds.),The handbook of social psychology (4th ed., Vol. 1, pp. 680–740). New York: McGraw-Hill.
Gollwitzer, P. M. (1999). Implementation intentions: Strong effects of simple plans.American Psychologist, 54(7), 493–503.
Gustafsson, M., Biel, A., & Gaerling, T. (1999). Overharvesting of resources of unknown size.Acta Psychologica,103(1–2),47–64.
Hardin, G. (1968). The tragedy of the commons.Science, 162, 1243–1248.
Hayashi, N., & Yamagishi, T. (1998). Selective play: Choosing partners in an uncertain world.Personality and Social Psychology Review, 2(4), 276–289.
Hine, D. W., & Gifford, R. (1996). Individual restraint and group efficiency in commons dilemmas: The effects of two types of environmental uncertainty.Journal of Applied Social Psychology, 26(11), 993–1009.
Hoffman, E., & Spitzer, M. L. (1985). Entitlements, rights, and fairness: An experimental examination of subjects’ concepts of distributive justice.Journal of Legal Studies, 14, 259–297.
Holmes, J. G. (1991). Trust and the appraisal process in close relationships. In W. H. P. D. Jones (Ed.),Advances in personal relationships: A research annual (Vol. 2, pp. 57–104). London: Jessica Kingsley Publishers.
Hsee, C. K. (1995). Elastic justification: How tempting but task-irrelevant factors influence decisions. Organizational Behavior and Human Decision Processes, 62(3), 330–337.
Insko, C. A., Schopler, J., Hoyle, R. H., Dardis, G. J., & Graetz, K. A. (1990). Individual-group discontinuity as a function of fear and greed.Journal of Personality and Social Psychology, 58(1), 68–79.
Insko, C. A., Schopler, J., Pemberton, M. B., Wieselquist, J., McIlraith, S. A., Currey, D. P., et al. (1998). Long-term outcome maximization and the reduction of interindividual-intergroup discontinuity.Journal of Personality and Social Psychology, 75(3), 695–711.
Isaac, R. M., Walker, J. M., & Williams, A. W. (1994). Group size and the voluntary provision of public goods.Journal of Public Economics, 54, 1–36.
Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decision under risk.Econometrica, 47, 263–291.
Kallgren, C. A., Reno, R. R., & Cialdini, R. B. (2000). A focus theory of normative conduct:When norms do and do notaffectbehavior. Personality and SocialPsychology Bulletin,26(8),1002–1012.
Kelley, H. H., & Stahelski, A. J. (1970). Social interaction basis of cooperators’ and competitors’ beliefs about others.Journal of Personality and Social Psychology, 16(1), 66–91.
Kelley, H. H., & Thibaut, J. W. (1978).Interpersonal relations: A theory of interdependence. New York: Wiley.
Kerr, N. L. (1989). Illusions of efficacy: The effects of group size on perceived efficacy in social dilemmas.Journal of Experimental Social Psychology, 25(4), 287–313.
Kerr, N. L., Garst, J., Lewandowski, D. A., & Harris, S. E. (1997).
That still, small voice: Commitment to cooperate as an internalized versus a social norm. Personality and Social Psychology Bulletin, 23(12), 1300–1311.
Kerr, N. L., & Kaufman-Gilliland, C. M. (1994). Communication, commitment, and cooperation in social dilemma.Journal of Personality and Social Psychology, 66(3), 513–529.
Kerr, N. L., & Kaufman-Gilliland, C. M. (1997). “. . . and besides, I probably couldn’t have made a difference anyway”: Justification of social dilemma defection via perceived self-inefficacy. Journal of Experimental Social Psychology, 33(3), 211–230.
Kollock, P. (1998). Social dilemmas: The anatomy of cooperation. Annual Review of Sociology, 24, 183-214.
Komorita, S. S., & Parks, C. D. (1996).Social dilemmas. Boulder, CO: Westview.
Kopelman, S. (2003).Cultural heterogeneity and economic power asymmetry in resource dilemmas: Implications for self-interested behavior in negotiations. Unpublished doctoral dissertation, Northwestern University, Evanston, IL.
Kopelman, S., Weber, J. M., & Messick, D. M. (2002). Factors influencing cooperation in commons dilemmas: A review of experimental psychological research. In E. Ostrom, T. Dietz, N. Dolsak, P. C. Stern, S. Stonich, & E. U. Weber (Eds.),The drama of the commons(pp. 113-156). Washington, DC: National Academy Press.
Kramer, R. M., & Brewer, M. B. (1986). Social group identity and the emergence of cooperation in resource conservation dilemmas. In H. Wilke, D. M. Messick, & C. Rutte (Eds.),Experimental social dilemmas(pp. 177-203). Frankfurt am Main, Germany: Verlag Peter Lang.
Kramer, R. M., McClintock, C. G., & Messick, D. M. (1986). Social values and cooperative response to a simulated resource conservation crisis.Journal of Personality, 54(3), 576-592.
Kramer, R. M., Wei, J., & Bendor, J. (2001). Golden rules and leaden worlds: Exploring the limitations of tit-for-tat as a social decision rule. In J. M. Darley, D. M. Messick, & T. R. Tyler (Eds.), Social influences on ethical behavior in organizations (pp. 177-200). Mahwah, NJ: Lawrence Erlbaum Associates, Inc.
Kruglanski, A. W. (1996). Motivated social cognition: Principles of the interface. In E. T. Higgins & A. W. Kruglanski (Eds.),Social psychology: Handbook of basic principles(pp. 493-520). New York: Guilford.
Kunda, Z. (1990). The case for motivated reasoning.Psychological Bulletin, 108, 480-498.
Kurzban, R., & Houser, D. (2001). Individual differences in cooperation in a circular public goods game.European Journal of Personality,15(1), S37–S52
Kurzban, R., McCabe, K., Smith, V. L., & Wilson, B. J. (2001). Incremental commitment and reciprocity in a real time public goods game.Personality and Social Psychology Bulletin, 27, 1662–1673.
Lalljee, M., Lamb, R., & Abelson, R. P. (1992). The role of event prototypes in categorization and explanation. In W. Stroebe & M. Hewston (Eds.),European review of social psychology(Vol. 3, pp. 153–181). New York: John Wiley & Sons, Ltd.
Langer, E. J. (1989). Minding matters: The consequences of mindlessness-mindfulness. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 22, pp. 137–173). San Diego: Academic Press.
Larrick, R. P., & Blount, S. (1997). The claiming effect: Why players are more generous in social dilemmas than in ultimatum games. Journal of Personality and Social Psychology, 72(4), 810–825.
Latane, B., & Darley, J. M. (1970).The unresponsive bystander: Why doesn’t he help?New York: Appleton-Century-Crofts.
Ledyard, J. O. (1995). Public goods: A survey of experimental research. In J. H. Kagel & A. E. Roth (Eds.),The handbook of experimental economics(pp. 111–194). Princeton, NJ: Princeton University Press.
Lefcourt, H. M. (1982).Locus of control: Current trends in theory and research(2nd ed.). Hillsdale, NJ: Lawrence Erlbaum Associate, Inc.
Liebrand, W. B., Jansen, R. W., Rijken, V. M., & Suhre, C. J. (1986). Might over morality: Social values and the perception of other players in experimental games.Journal of Experimental Social Psychology, 22(3), 203–215.
Luce, R. D., & Raiffa, H. (1957).Games and decisions: Introduction and critical survey. New York: Wiley.
Mannix, E. A. (1991). Resource dilemmas and discount rates in decision making groups.Journal of Experimental Social Psychology, 27(4), 379–391.
Mannix, E. A. (1993). Organizations as resource dilemmas: The effects of power balance on coalition formation in small groups.Organizational Behavior and Human Decision Processes, 55(1), 1–22.
Mannix, E. A., & White, S. B. (1992). The impact of distributive uncertainty on coalition formation in organizations.Organizational Behavior and Human Decision Processes, 51(2), 198–219.
March, J. (1994).A primer on decision-making: How decisions happen. New York: Free Press.
Massey, K., Freeman, S., & Zelditch, M. (1997). Status, power, and accounts.Social Psychology Quarterly, 60(3), 238–251.
McClintock, C. G. (1972). Social motivation: A set of propositions. Behavioral Science, 17(5), 438–455.
McDaniel, W. C., & Sistrunk, F. (1991). Management dilemmas and decisions: Impact of framing and anticipated responses.Journal of Conflict Resolution, 35(1), 21–42.
Messick, D. M. (1999). Alternative logics for decision making in social settings.Journal of Economic Behavior and Organization, 39(1), 11–28.
Messick, D. M., & Allison, S. (1993). Equality as a decision heuristic. In B. A. Mellers & J. Baron (Eds.),Psychological perspectives on justice(pp. 11–31). New York: Cambridge University
Messick, D. M., & Brewer, M. B. (1983). Solving social dilemmas. In L. Wheeler & P. Shaver (Eds.),Review of personality and social psychology: Vol 4(pp. 11–44). Beverly Hills, CA: Sage
Messick, D. M., & McClintock, C. G. (1968). Motivational bases of choice in experimental games.Journal of Experimental Social Psychology, 4(1), 1–25.
Messick, D. M., Wilke, H., Brewer, M. B., Kramer, R. M., Zemke, P. E., & Lui, L. (1983). Individual adaptations and structural change as solutions to social dilemmas.Journal of Personality and Social Psychology, 44(2), 294–309.
Moghaddam, F. M., Taylor, D. M., & Wright, S. C. (1993).Social psychology in cross-cultural perspective. New York: Freeman. Murnighan, J. K. (1994). Game theory and organizational behavior.
In B. M. Staw & L. L. Cummings (Ed.), Research in organizational behavior(pp. 83–123). Greenwich, CT: JAI Press.
Murnighan, J. K., Oesch, J. M., & Pillutla, M. (in press). Player types and self impression management in dictator games: Two experiments.Games and Economic Behavior.
Murnighan, J. K., & Ross, T. (1999). On the collaborative potential of psychology and economics.Journal of Economic Behavior and Organization, 39, 1–10.
Nowell, C., & Tinkler, S. (1994). The influence of gender on the provision of a public good.Journal of Economic Behavior and Organization, 25(1), 25–36.
Orbell, J., Dawes, R., & Schwartz-Shea, P. (1994). Trust, social categories, and individuals: The case of gender.Motivation and Emotion, 18(2), 109–128.
Orbell, J., & Dawes, R. M. (1991). A cognitive miser theory of cooperators’ advantage. American Political Science Review, 85, 515–528.
Orbell, J. M., & Dawes, R. M. (1993). Social welfare, cooperators’ advantage, and the option of not playing the game.American Sociological Review, 58(6), 787–800.
Pillutla, M. M., & Chen, X.-P. (1999). Social norms and cooperation in social dilemmas: The effects of context and feedback.Organizational Behavior and Human Decision Processes, 78(2), 81–103.
Polzer, J. T., Stewart, K. J., & Simmons, J. L. (1999). A social categorization explanation for framing effects in nested social dilemmas. Organizational Behavior and Human Decision Processes, 79(2), 154–178.
Pruitt, D. G., & Kimmel, M. J.. (1977). Twenty years of experimental gaming: Critique, synthesis, and suggestions for the future.Annual Review of Psychology, 28, 363–392.
Rapoport, A., Budescu, D. V., & Suleiman, R. (1993). Sequential requests from randomly distributed shared resources.Journal of Mathematical Psychology, 37(2), 241–265.
Roch, S. G., & Samuelson, C. D. (1997). Effects of environmental uncertainty and social value orientation in resource dilemmas. Organizational Behavior and Human Decision Processes, 70(3), 221–235.
Ross, L., & Nisbett, R. E. (1991).The person and the situation: Perspectives of social psychology. New York: McGraw-Hill.
Rutte, C. G., & Wilke, H. A. (1984). Social dilemmas and leadership. European Journal of Social Psychology, 14(1), 105–121.
Rutte, C. G., Wilke, H. A., & Messick, D. M. (1987). Scarcity or abundance caused by people or the environment as determinants of behavior in the resource dilemma.Journal of Experimental Social Psychology, 23(3), 208–216.
Samuelson, C. D. (1991). Perceived task difficulty, causal attributions, and preferences for structural change in resource dilemmas.Personality and Social Psychology Bulletin, 17(2), 181–187.
Samuelson, C. D. (1993). A multiattribute evaluation approach to
structural change in resource dilemmas.Organizational Behavior and Human Decision Processes, 55(2), 298–324.
Samuelson, C. D., & Allison, S. T. (1994). Cognitive factors affecting the use of social decision heuristics in resource-sharing tasks. Organizational Behavior and Human Decision Processes, 58(1), 1–27.
Schopler, J., & Insko, C. A. (1999). The reduction of the interindividual-intergroup discontinuity effect: The role of future consequences. In M. Foddy, M. Smithson, S. Schneider, &
M. Hogg (Eds.),Resolving social dilemmas: Dynamic, structural, and intergroup aspects(pp. 281–294). Philadelphia: Psychology Press.
Schopler, J., Insko, C. A., Drigotas, S. M., Wieselquist, J., Pemberton, M. B., & Cox, C. (1995). The role of identifiability in the reduction of interindividual-intergroup discontinuity. Journal of Experimental Social Psychology, 31(6), 553–574.
Seijts, G. H., & Latham, G. P. (2000). The effects of goal setting and group size on performance in a social dilemma.Canadian Journal of Behavioural Science, 32(2), 104–116.
Seijts, G. H., Latham, G. P., & Whyte, G. (2000). Effect of self- and group efficacy on group performance in a mixed-motive situation.Human Performance, 13(3), 279–298.
Sherif, M. (1966).Group conflict and co-operation: Their social psychology. London: Routledge & Kegan Paul.
Silverstein, A., Cross, D., Brown, J., & Rachlin, H. (1998). Prior experience and patterning in a prisoner’s dilemma game.Journal of Behavioral Decision Making, 11(2), 123–138.
Simon, H. A. (1955). A behavioral model of rational choice.Quarterly Journal of Economics, 69, 118.
Smith, E. R., & DeCoster, J. (1999). Associative and rule-based processing. In S. Chaiken & Y. Trope (Eds.),Dual-process theories in social psychology(pp. 323–336). New York: Guilford.
Smithson, M., & Foddy, M. (1999). Theories and strategies for the study of social dilemmas. In M. Foddy, M. Smithson, S. Schneider, & M. Hogg (Eds.),Resolving social dilemmas(pp. 1–14). Philadelphia: Psychology Press.
Snyder, M., & Gangestad, S. (1986). On the nature of self-monitoring: Matters of assessment, matters of validity.Journal of Personality and Social Psychology, 51, 125–139.
Sonnemans, J., Schram, A., & Offerman, T. (1998). Public good provision and public bad prevention: The effect of framing.Journal of Economic Behavior and Organization, 34(1), 143–161.
Stockard, J., Van de Kragt, A. J., & Dodge, P. J. (1988). Gender roles and behavior in social dilemmas: Are there sex differences in cooperation and in its justification?Social Psychology Quarterly, 51(2), 154–163.
Taylor, D. M., & Modhaddam, F. M. (1994).Theories of intergroup relations(2nd ed.). London: Praeger.
Taylor, M. (1976).Anarchy and cooperation. New York: Wiley.
Taylor, S. E. (1998). The social being in social psychology. In D. T.
Gilbert, S. T. Fiske, & G. Lindzey (Eds.),The handbook of social psychology(Vol. 1, pp. 58–95). Boston: McGraw-Hill.
Tenbrunsel, A. E., & Messick, D. M. (1999). Sanctioning systems, decision frames, and cooperation. Administrative Science Quarterly, 44(4), 684–707.
Turner, J. C., Hogg, M. A., Oakes, P. J., Reicher, S. D., & Wetherell, M. S. (1987).Rediscovering the social group: A self-categorization theory.Oxford, England: Blackwell.
van Dijk, E., & Wilke, H. (1997). Is it mine or is it ours? Framing property rights and decision making in social dilemmas.Organizational Behavior and Human Decision Processes, 71(2), 195–209.
van Dijk, E., & Wilke, H. (2000). Decision-induced focusing in social dilemmas: Give-some, keep-some, take-some, and leave-some dilemmas. Journal of Personality and Social Psychology, 78(1), 92–104.
van Dijk, E., Wilke, H., Wilke, M., & Metman, L. (1999). What information do we use in social dilemmas? Environmental uncertainty and the employment of coordination rules.Journal of Experimental Social Psychology, 35(2), 109–135.
Van Lange, P. A. M., Agnew, C. R., Harinck, F., & Steemers, G. E. M. (1997). From game theory to real life: How social value orientation affects willingness to sacrifice in ongoing close relationships.Journal of Personality and Social Psychology, 73(6), 1330–1344.
Van Lange, P. A. M., De Bruin, E. M. N., Otten, W., & Joireman, J. A. (1997). Development of prosocial, individualistic, and competitive orientations: Theory and preliminary evidence.Journal of Personality and Social Psychology, 73(4), 733–746.
Van Lange, P. A. M., & Kuhlman, D. M. (1994). Social value orientations and impressions of partner’s honesty and intelligence: A test of the might versus morality effect.Journal of Personality and Social Psychology, 67(1), 126–141.
Van Lange, P. A. M., Liebrand, W. B., & Kuhlman, D. M. (1990). Causal attribution of choice behavior in three N-person prisoner’s dilemmas.Journal of Experimental Social Psychology, 26(1), 34–48.
Van Lange, P. A. M., Liebrand, W. B. G., Messick, D. M., & Wilke, H. A. M. (1992). Social dilemmas: The state of the art. In W. B.
G. Liebrand, D. M. Messick, & H. A. M. Wilke (Eds.),Social dilemmas: Theoretical issues and research findings(pp. 3–28). Oxford, England: Pergamon.
Van Vugt, M., & De Cremer, D. (1999). Leadership in social dilemmas: The effects of group identification on collective actions to provide public goods.Journal of Personality and Social Psychology, 76(4), 587–599.
Van Vugt, M., Meertens, R. M., & Van Lange, P. A. M. (1995). Car versus public transportation? The role of social value orientations in a real-life social dilemma. Journal of Applied Social Psychology, 25(3), 258–278.
Van Vugt, M., Van Lange, P. A. M., & Meertens, R. M. (1996). Commuting by car or public transportation? A social dilemma analysis of travel mode judgments.European Journal of Social Psychology, 26(3), 373–395.
Wade-Benzoni, K. A., Okumura, T., Brett, J. M., Moore, D. A., Tenbrunsel, A. E., & Bazerman, M. H. (2002). Cognition andbehavior in asymmetric social dilemmas: A comparison of two cultures.Journal of Applied Psychology,87(1), 87–95.
Walters, A. E., Stuhlmacher, A. F., & Meyer, L. L. (1998). Gender and negotiator competitiveness: A meta-analysis.Organizational Behavior and Human Decision Processes, 76(1), 1–29.
Weber, J. M. (2003).Catalysts for cooperation: Consistent contributors in public good dilemmas. Unpublished doctoral dissertation, Northwestern University, Evanston, IL.
Weber, J. M., Malhotra, D., & Murnighan, J. K. (2004).Normal acts of irrational trust: Motivated attributions and the trust development process.Manuscript submitted for publication.
Wit, A., & Wilke, H. (1988). Subordinates’ endorsement of an allocating leader in a commons dilemma: An equity theoretical approach.Journal of Economic Psychology, 9(2), 151–168.
Wit, A., & Wilke, H. (1998). Public good provision under environmental social uncertainty.European Journal of Social Psychology, 28(2), 249–256.
Wit, A., & Wilke, H. A. (1990). The presentation of rewards and punishments in a simulated social dilemma.Social Behavior, 5(4), 231–245.
Wit, A. P., & Kerr, N. L. (2002). “Me versus just us versus us all”: Categorization and cooperation in nested social dilemmas. Journal of Personality and Social Psychology, 83(3), 616–637.
Wit, A. P., Wilke, H. A., & Van Dijk, E. (1989). Attribution of leadership in a resource management situation.European Journal of Social Psychology, 19(4), 327–338.
Yamagishi, T. (1988). The provision of a sanctioning system in the United States and Japan.Social Psychology Quarterly, 51(3), 265–271.