MANAGEMENT & ORGANIZATIONS
Harold H. Hines Jr. Professor of Risk Management
Professor Murnighan has received numerous awards including the Top Faculty Member Award from the Kellogg-Hong Kong Executive MBA program, Teaching Excellence Awards from the University of Illinois and the University of British Columbia, and the Professor of the Year Award from the MBA Association at the University of Illinois. In 2006 he received the Distinguished Educator Award from the Academy of Management, a Career Achievement award.
His research has been published in many different journals, in organizational behavior, psychology, and economics. His recent books include The Art of High-Stakes Decision-Making: Tough Calls in a Speed-Driven World (with John Mowen; John Wiley & Sons, 2002) and Social Psychology and Economics (with David De Cremer and Marcel Zeelenberg: Lawrence Erlbaum, 2006).
Professor Murnighan teaches regularly in executive programs in the US, Canada, the Middle East, Asia, and Europe. He is an active consultant and trainer and has worked with several major corporations, including the American Dental Association, Aon, Caterpillar, CDW, Jefferson Wells, Kraft, the Metro Achievement Center, Motorola, National Wildlife Federation, Pfizer, and the Young Presidents Organization, among others.
Ethics
Group Decision-Making
Group Dynamics
Leadership
Leading High-Impact Teams
Negotiations
Psychology
Risk Management
Teams
- Recent Media Coverage
The Boston Globe: The lying game - 10/18/2009
South China Morning Post: Educator revels in his role as provocateur - 6/7/2009
The Globe and Mail: Thanks for sharing: You set a good example - 4/3/2009
The Economist: What's in the journals, January 2009: One born every day - 1/28/2009
See all Kellogg in the Media
- Recent Kellogg News
Full speed ahead - 10/2/2009
Change leadership - 1/19/2009
See all Kellogg News
An experiment tested whether groups lie more than individuals. Groups lied more than individuals when deception was guaranteed to maximize economic outcomes, but lied relatively less than individuals when honesty could be used strategically. These results suggest that groups are more strategic than individuals in that they will adopt whatever course of action best serves their economic interest.
The long history of experimental research on the prisoner's dilemma (PD) has primarily used a methodology that eliminates cues to participants. Researchers, however, have interpreted participants' choices as cooperative or competitive. The authors' research shows that giving participants researchers' interpretive labels of the game, the choices, and the outcomes, compared to no labels, led to significantly more cooperation; labels such as trust and cooperate/defect augmented cooperation even more. A second experiment found that independent evaluations of the labels led to perceptions that were similar to individuals' choices in the first experiment. These results suggest that we might need to rethink the import of many of our previous findings and their applicability to everyday interactions.
We investigated the effects of intragroup and cross-subgroup communications in an experimental field study on demographic faultlines. The results indicated that faultlines explained more variance in perceptions of team learning, psychological safety, satisfaction, and expected performance than single-attribute heterogeneity indexes. In addition, cross-subgroup work communications were effective for groups with weak faultlines but not for groups with strong faultlines. Overall, this study extends the original faultline model, documents the utility of the concept of faultlines, and identifies some of their effects on work group outcomes.
In most models of bargaining, costless and unverifiable lies about private information and incredible threats about future actions are considered cheap talk and do not impact outcomes. In practice, however, this type of talk is often an integral part of bargaining. This experiment examines the impact of cheap talk in an ultimatum bargaining setting with two-sided imperfect information. In contrast to previous work, the experiment provides an opportunity for deceptions to be revealed and punished. Results show that lies about private information and (incredible) threats of future actions do influence bargaining outcomes (offers and responses) in both the short- and long-term.
This article presents the true story of an exercise conducted in an executive MBA class. The Dollar Auction has been used to great benefit in many classes on negotiations and organizational behavior. The case described here represents the most extreme outcome that the author or his colleagues have ever observed. It leads to a number of observations about emotions, rationality, and strategy in the heat of battle and provides an opportunity to discuss the potent effects of time pressure and competition in individual and organizational decision making.
This paper uses two laboratory experiments to investigate the effects of contracts on interpersonal trust. We predict that the use of binding contracts to promote or mandate cooperation will lead interacting parties to attribute others' cooperation to the constraints imposed by the contract rather than to the individuals themselves, thus reducing the likelihood of trust developing. We also predict that, although non-binding contracts may not generate as much initial cooperation as binding contracts, they will generate personal rather than situational attributions for any cooperation that results and will therefore not interfere with trust development. Two experiments investigated the effects of the use and removal of binding and non-binding contracts. When binding contracts that were previously allowed were no longer allowed or no longer chosen, trust dropped significantly. In contrast, non-binding contracts led to considerable cooperation, and their removal reduced trust less than removing binding contracts. Behavioral and perceptual data suggest that non-binding contracts lead to personal attributions for cooperation and thus may provide an optimal basis for building interpersonal trust in a variety of situations.
Interpersonal relationships can be fragile. The mere perception of opportunistic behavior can lead to a breakdown in cooperation. Once damaged, the question then arises as to whether and how cooperation might be restored. Noncooperative game theory raises serious doubts about the possibilities, although interactional justice and impression management research have shown that verbal explanations can dampen reactions to aversive behavior. Philosophical, anthropological, and ethological research all suggest that genuine forgiveness may require something more tangible and substantive than an explanation. Thus, the current experiment investigated the effects of explanations and varying forms of substantive amends on the restoration of mutual cooperation. The results confirm that rebuilding cooperation is feasible. Apologies and simple explanations can be effective to a degree, though substantive amends have significantly more positive effects than explanations alone. In contrast to prior findings on interactional justice, acknowledgments were more effective than denials in repairing short interactions. This research demonstrates that, once breached, cooperation can be reestablished and that actions as well as explanations and apologies can augment the process in important and sometimes subtle ways.
Recent experimental research on dictatorship games shows that many dictators share their outcomes with unknown, anonymous others. The data suggests that dictators can be “typed” as rational (taking the maximum), equal (splitting outcomes equally), or “other.” This paper experimentally tests the self-impression management model, which predicts that individuals act to show themselves in a positive light, even when they are the only observer of their own behavior. The model predicts that the “other” type of dictator will avoid being greedy by taking more only as their choices are increasingly restricted. Results from two experiments support the model's predictions. The conclusion advocates differentiating basic psychological motivations in modeling individual behavior.
This paper presents a model of the cognitive processes that precede decisions to help another person. The empathy-prospect model predicts that potential helpers make decisions in much the same way as decision makers in other contexts do (i.e., they evaluate prospects) and that perceptions of need and the empathic reactions and intentions to help that they generate will be stronger for people observing losses rather than gains. The model also predicts that intentions to help should increase when (a) the predicament is serious, (b) money is not involved, or (c) help entails few costs for the potential altruist. The results from 2 experiments provide clear support for these predictions. The findings suggest that (a) the gains or losses of another person contribute to perceptions of that person's needs and feelings of empathy, (b) empathy is the primary proximal determinant of prosocial motivations, and (c) potential losses that are serious accentuate altruistic reactions.
This paper investigates the dynamics of deception and retribution in repeated ultimatum bargaining. Anonymous dyads exchanged messages and offers in a series of four ultimatum bargaining games that had prospects for relatively large monetary outcomes. Variations in each party's knowledge of the other's resources and alternatives created opportunities for deception. Revelation of prior unknowns exposed deceptions and created opportunities for retribution in subsequent interactions. Results showed that although proposers and responders chose deceptive strategies almost equally, proposers told more outright lies. Both were more deceptive when their private information was never revealed, and proposers were most deceptive when their potential profits were largest. Revelation of proposers' lies had little effect on their subsequent behavior even though responders rejected their offers more than similar offers from truthful proposers or proposers whose prior deceit was never revealed. The discussion and conclusions address the dynamics of deception and retribution in repeated bargaining interactions.
Reviewed work(s): Debating Rationality: Non-Rational Aspects of Organizational Decision Making by Jennifer Halpern; Robert Stern
In January of 1999, leading negotiation scholars and practitioners gatheredat Northwestern University for a conference sponsored by the HewlettFoundation. Here, we chronicle a conversation that occurred regardingnegotiation research. Our diverse group consisted of people with backgroundsin social psychology, anthropology, sociology, and clinical practice.We discussed issues surrounding where research takes place (in the lab orin the field), experimental participants (students or practicing professionals),and the ultimate value of research.
This essay introduces the papers of this special issue by framing them within a discussion of the past and present state of relations between social psychologists and economists working on models of human decision making.We reviewthe different approaches scholars have taken in the two fields, as well as some experimental work conducted in each, and argue that we have much to learn from each other. While psychologists and economists have typically attended to different issues in their research, the papers in this collection ably illustrate the potential for collaborations drawing on the strengths of each field.
Review essay of "Fair Division: From Cake-Cutting to Dispute Resolution," by Stephen J. Brams and Alan D. Taylor. Cambridge: Cambridge University Press, 1996.
Business negotiations often involve cooperative arrangements. Sometimes one party will renege on a cooperative enterprise for short-term opportunistic gain. There is a common assumption that such behavior necessarily leads to a spiral of mutual antagonism. We use some of the philosophical literature to frame general research questions and identify relevant variables in dealing with defection. We then describe an experimental approach for examining the possibility of reconciliation and discuss the results of one such experiment where participants were the victims of defection. In contrast to the initial assumptions we found that many participants were willing to reconcile, and that penance conditions, when demanded, were less stringent than expected. We suggest that these findings warrant further study and have implications for business dealings.
This paper investigates the information dilemma in negotiations: if negotiators reveal information about their priorities and preferences, more efficient agreements may be reached but the shared information may be used strategically by the other negotiator, to the revealers' disadvantage. We present a theoretical model that focuses on the characteristics of the negotiators, the structure of the negotiation, and the available incentives; it predicts that experienced negotiators will outperform naive negotiators on distributive (competitive) tasks, especially when they have information about their counterpart's preferences and the incentives are high--unless the task is primarily integrative, in which case information will contribute to the negotiators maximizing joint gain. Two experiments (one small, one large) showed that tire revelation of one's preferences was costly and that experienced negotiators outperformed their naive counterparts by a wide margin, particularly when the task and issues were distributive and incentives were large. Our results help to identify the underlying dynamics of the information dilemma and lead to a discussion of the connections between information and social dilemmas and the potential for avoiding inefficiencies.
In this article we address issues of diversity within organizational groups by discussing and summarizing previous approaches and by introducing a new variable--faultlines--which depends on the alignment of individual member characteristics. By analyzing a group's faultlines, we focus attention on the underlying patterns of group member characteristics, which can be an important determinant of subgroup conflict, particularly when the group's task is related to one of its faultlines. We discuss the dynamics of faultlines from the early to later stages of a group's development and show how they may be strongest and most likely when diversity of individual member characteristics is moderate.
This paper investigates the impact of some of the underlying dynamics of volunteering choices in organizational contexts, focusing on individual, group, and organizational level causes. Three scenario-based experiments manipulate individuals' standing within their organization (i.e., whether they are doing well or poorly) in combination with variables such as the expected efficacy of one's team and positive or negative organizational performance. In comparison to other recent volunteering studies, all three current experiments focused on an explicit organizational context and found much stronger intentions to volunteer, particularly when a person's standing was good. The combination of poor standing with expectations of poor performance by one's group or one's organization led to reductions in these otherwise strong intentions to volunteer. The results also show that feelings of obligation, expectations of extrinsic rewards, and identifying with one's organization are all significantly related to volunteering choices.
The winner's curse is frequently observed in auctions: The person who bids the most and wins the auction may ultimately regret the bid since it often exceeds the value of the object being auctioned. Ball, Bazerman, and Carroll (1991) have shown that repeated experience as a bidder leads to little learning and repeated winner's curses. The current study investigated whether feedback and relevant experience could contribute to bidders learning to avoid the winner's curse in the two tasks that have dominated winners' curse research. The results showed that experience had little impact and that additional feedback reduced overbidding but never extinguished the winners' curse; instead, it was pervasive and highly resistant to learning. High endowments and limited feedback, in particular, led to the most extreme curses. The results are discussed with respect to the emotions associated with winning and the emotional bases of risky decision making.
The winner's curse is frequently observed in auctions: The person who bids the most and wins the auction may ultimately regret the bid since it often exceeds the value of the object being auctioned. Ball, Bazerman, and Carroll (1991) have shown that repeated experience as a bidder leads to little learning and repeated winner's curses. The current study investigated whether feedback and relevant experience could contribute to bidders learning to avoid the winner's curse in the two tasks that have dominated winners' curse research. The results showed that experience had little impact and that additional feedback reduced overbidding but never extinguished the winners' curse; instead, it was pervasive and highly resistant to learning. High endowments and limited feedback, in particular, led to the most extreme curses. The results are discussed with respect to the emotions associated with winning and the emotional bases of risky decision making.
This paper addresses an anomaly in experimental economics, the rejection of ultimatum offers, and uses a psychological explanation for this essentially economic event. The wounded pride/spite model predicts that informed, knowledgeable respondents may react to small ultimatum offers by perceiving them as unfair, feeling anger, and acting spitefully. Results of a large scale experiment support the model, showing that rejections were most frequent when respondents could evaluate the fairness of their offers and attribute responsibility to offerers. In addition, anger was a better explanation of the rejections than perceptions that the offers were unfair. The discussion addresses the rarely studied but frequently observed emotions that negotiations provoke.
We designed two laboratory experiments to test popular hypotheses explaining the failure of subgame-perfect equilibrium models to explain behavior in ultimatum games. The first experiment varied information available to respondents. When respondents did not know the amount being divided, offerers offered (and respondents accepted) significantly lower offers than when the respondents knew the amount being divided. The second experiment replicated this result and also showed that people occasionally reject “free” money (i.e., offers with no strings attached). This evidence does not support earlier explanations for ultimatum anomalies and identifies conditions where subgame-perfect models apply.
This article presents the results of two experiments that investigated fairness and apparent fairness in ultimatum negotiations. Results suggest that offerers--those presenting offers--were both strategic and exploitative: they made offers that appeared fair only when respondents (potential recipients) had full information about the amounts to be divided or when third parties labeled offers as to their fairness. In contrast, respondents ignored fairness claims but reacted to the sizes of offers and to third party's evaluations.
This study examined the dispute-resolution behavior of the "intravenor," a distinct third-party role in organizational dispute resolution. Unlike a mediator, whose involvement in the dispute is at the whim of the disputants, the intravenor can control the outcome of the dispute. Unlike an arbitrator, who is compelled to dictate the outcome of the dispute, the intravenor may or may not impose an outcome. The experiment reported here examined the impact of four variables on third party behavior: The third party′s role (intravenor versus mediator), the third party′s beliefs about the disputants reaching agreement (cooperative versus uncooperative disputants), the third party′s self-interest in the outcome, and the third party′s concern about the disputants′ outcome (interest in the disputant′s mutual welfare). The results suggest that intravention spawns a distinctive pattern of third-party behavior: Intravenors imposed outcomes in 66% of the cases, but more when they viewed the disputants as uncooperative than cooperative. Only 44% of the imposed outcomes reflected the disputants′ underlying interests, but this was greater when the intravenor had high compared to low concern for the disputants′ aspirations. Intravenors were more likely than mediators to use forceful, pressure tactics, and were more confident and saw themselves as more influential. Taken together, the results provide the basis for an integrated model of third-party intervention in organizational dispute resolution.
This study examined the dispute-resolution behavior of the "intravenor," a distinct third-party role in organizational dispute resolution. Unlike a mediator, whose involvement in the dispute is at the whim of the disputants, the intravenor can control the outcome of the dispute. Unlike an arbitrator, who is compelled to dictate the outcome of the dispute, the intravenor may or may not impose an outcome. The experiment reported here examined the impact of four variables on third party behavior: The third party′s role (intravenor versus mediator), the third party′s beliefs about the disputants reaching agreement (cooperative versus uncooperative disputants), the third party′s self-interest in the outcome, and the third party′s concern about the disputants′ outcome (interest in the disputant′s mutual welfare). The results suggest that intravention spawns a distinctive pattern of third-party behavior: Intravenors imposed outcomes in 66% of the cases, but more when they viewed the disputants as uncooperative than cooperative. Only 44% of the imposed outcomes reflected the disputants′ underlying interests, but this was greater when the intravenor had high compared to low concern for the disputants′ aspirations. Intravenors were more likely than mediators to use forceful, pressure tactics, and were more confident and saw themselves as more influential. Taken together, the results provide the basis for an integrated model of third-party intervention in organizational dispute resolution.
This experiment factorially combined the major independent variables from previous demand-game experiments (discount factors, outside options, termination probability, and first mover). Game-theoretic predictions were largely refuted by the data, and outcomes were often inefficient. Players without an outside option demanded more than predicted, and those with an option appeared to anticipate this behavior. Nonetheless, there was a positive relationship between differences in equilibrium predictions and differences in behavior. Bargainers appeared to focus on a minimally acceptable offer in making their demands and in considering the likelihood that the other party would accept their offer.
This paper presents experimental tests of two models of cooperation in finitely-repeated prisoner's dilemma games (Kreps, Milgrom, Roberts, and Wilson, 1982). The models suggest that either a perception that the other party may use the tit-for-tat strategy or mutual uncertainty concerning dominant noncooperative strategies can lead to rational cooperation. The experiment independently manipulated both types of uncertainty and allowed for inferences concerning the players' prior, ‘homemade’ preferences for cooperation. Only in relatively restricted situations did either type of uncertainty promote cooperation. Instead, players cooperated much more than was predicted; they also cooperated more when they were certain of their opponents' payoffs.
This paper uses evolutionary, social psychological, organizational, and game theoretic literatures as frames to explore the dilemma of the choice to volunteer. Although one's group may benefit from voluntary action, volunteers typically incur more costs than others, even when their actions are successful. The four experiments presented here addressed when and why people volunteer in fictitious financial scenarios that varied group size, the number of volunteers required, the payoffs to volunteers and nonvolunteers, and the need to work more or for smaller outcomes than others. The results suggest that, in general, undergraduates, M.B.A.s, and executives volunteered most often when personal benefits were high. Executives also focused on issues related to self-interest when they were questioned about voluntary action in the workplace. The discussion raises issues about voluntary action in situations in which people interact at a distance and speculates about implications for the dynamics and antecedents of altruistic action.
This article presents a new form of prisoner's dilemma (PD) game called asymmetric dilemmas (ADs). The underlying structure of ADs conforms to the requirements of a PD game, but mutually cooperative choices by the two parties give them known, different outcomes. Similarly, mutually noncooperative choices also give them different outcomes. This article summarizes the results of a series of experiments on ADs and presents the theoretical rationale derived from those findings. The results question the generality of Axelrod's confidence in the general utility of the tit-for-tat strategy and show that asymmetries are best solved via systems of complex alternation. Finally, the games and the theory can be easily adapted for classroom or training exercises on complex strategic planning in negotiations.
This paper proposes that interactions in accounting and auditing can be viewed as a large negotiation system. We define negotiations broadly to include any context in which two or more parties with differing preferences jointly make decisions that affect the welfare of both (all) parties. After a brief overview of negotiation research, the paper: (1) shows how recent research in accounting and auditing has implicitly addressed various aspects of negotiations; (2) comments on the Elias (1990) and Chalos and Haka (1990) studies which appear in this issue; (3) discusses departures from decision maker rationality which may seriously affect negotiations; and (4) uses three scenarios to highlight the importance of three negotiation issues (third party processes, the coordination of work groups, and social dilemmas) in accounting. We conclude by suggesting that a multifaceted research strategy, combining theory, description, and prescription, provides the greatest potential for progress in the study of accounting and auditing negotiations.
This paper focuses on the relationship between the internal dynamics and success of a population of intense work groups, professional string quartets in Great Britain. We observed three basic paradoxes: leadership versus democracy, the paradox of the second violinist, and confrontation versus compromise. The central findings indicate that the more successful quartets recognized but did not openly discuss the paradoxes. Instead, they managed these inherent contradictions implicitly and did not try to resolve them. The discussion addresses the study of intense work groups, the forces that drive these paradoxes, and potential applications to other organizational groups.
This paper proposes that interactions in accounting and auditing can be viewed as a large negotiation system. We define negotiations broadly to include any context in which two or more parties with differing preferences jointly make decisions that affect the welfare of both (all) parties. After a brief overview of negotiation research, the paper: (1) shows how recent research in accounting and auditing has implicitly addressed various aspects of negotiations; (2) comments on the Elias (1990) and Chalos and Haka (1990) studies which appear in this issue; (3) discusses departures from decision maker rationality which may seriously affect negotiations; and (4) uses three scenarios to highlight the importance of three negotiation issues (third party processes, the coordination of work groups, and social dilemmas) in accounting. We conclude by suggesting that a multifaceted research strategy, combining theory, description, and prescription, provides the greatest potential for progress in the study of accounting and auditing negotiations.
This paper reports the results of three experiments designed to test the predictions of the principal game-theoretic models of bargaining concerning the influence of risk aversion on bargaining outcomes. These models predict that risk aversion will be disadvantageous in bargaining except in situations in which potential agreements are lotteries with a positive probability of being worse than disagreement. The experimental results support the models' predictions. However, in the range of payoffs studied here, the effects due to risk aversion may be smaller than some of the focal point effects observed in previous experiments. Implications for further theoretical and experimental work are considered.
Presents a study on the effect of risk aversion on bargaining. Classes of bargaining model; Hypothesis on the impact of risk aversion on bargaining; Conclusion.
The literature on organizations has long focused on the concept of coalition behavior. An experiment on coalition formation was designed as a simulation of organizational decision-making groups. Previous research and a variety of game theoretic and social psychological models of coalition behavior provided strong empirical and theoretical bases for the study. Groups of masters students interacted in four sessions that manipulated the power of the five actors in each group. Results were compared to previous findings and were used to evaluate the theories. The findings suggest that (1) revenge was not often used by the players in these groups; (2) social pressure and the weak players' ability to communicate depressed the strongest players' outcomes; and (3) S.S. Komorita and J. Chertkoff's (1973 Psychological Review, 80, 149–162) Bargaining theory and the Roth-Shapley value (Roth, 1977 and Roth, 1977 Journal of Mathematical Psychology, 16, 153–160; Econometrica, 45, 654–657; L. S. Shapley, 1953 in H. W. Kuhn & A. W. Tucker (Eds.), Contributions to thefs theory of games (Vol. 2), Princeton, NJ: Princeton University Press) predicted coalition outcomes better than the other models tested. The implications for organizations and their members are discussed.
This paper examines the development of norms in newly formed groups. The behavior of 19 decision-making groups provided the basis for a model of norm development, in which uncertainty over appropriate behavior leads members to use their past experiences in similar social settings as scripts for choosing behaviors in the current situation. Depending on the similarity of the members' scripts, a common basis for action is either taken for granted or negotiated within the group. As the members interact they either tacitly revise their beliefs about appropriate action, implicitly agreeing with the direction being taken by the group, or overtly attempt to pull the group toward their own interpretation through challenges to the implied norm. Data from the decision-making groups is used to illustrate the model, and implications for related research domains are discussed.
Any group that includes three or more people provides the opportunity for its members to form coalitions. This study used two three-person games to investigate the formation of blocs, the joint action of two people against a third. The process as well as the outcomes of negotiations were observed: in addition to the two games, the secrecy of negotiations and the experience of the bargainers were studied. The results did not support either the applicable game theoretic predictions, or several hypotheses concerning the independent variables. Equal splits of the potential payoffs and altruistic blocs, who offered relatively high payoffs to the excluded third players, predominated. The results are discussed with respect to the political process or organizational groups.
Practitioners often apply theories and techniques, but the results of their applications rarely inform the original theory. Consequently, theories and theory-based techniques are seldom altered on the basis of practice, even when such attention is appropriate. This article describes an application of the Nominal Group Technique (NGT) and suggestions for altering the technique on the basis of learnings from the application. The NGT was used in two decision-making meetings. On the basis of problems experienced with NGT during the first meeting, it was modified in ways that increased its ability to (1) achieve decision acceptance and (2) address ill-structured problems. The modified technique was used successfully during the second meeting. After describing the use of NGT, the article presents an expanded model of the technique and its underlying assumptions.
Several models of prisoner's dilemma interactions were tested in a series of twelve games whose termination point was determined probabilistically. A new model was introduced to discriminate among equilibrium and nonequilibrium situations on the basis of a player's expected benefits or losses for cooperating. The experiment included twelve payoff matrices, three probabilities for continuing, two opponent strategies, and the player's sex as independent variables. Results showed that both the game payoffs and the probability that the game would continue interacted to affect the rates of cooperation observed, and that the equilibrium model predicted this outcome most accurately. While the predictions of each of the models were supported, the equilibrium models appeared to be superior to the others. The discussion highlights the importance of considering the likelihood of a game terminating as a major determinant of the cooperation that can be expected in mixed—motive interactions.
This study investigated the effects of two group decision making techniques on a set of four problems in groups of sizes 3, 7, and 11. Participants included 192 male and 132 female undergraduates. Estimates that could be evaluated for accuracy for each of the problems were collected for a series of three trials. One set of groups received face-to-face verbal feedback from each other, while the other set received written feedback. These data were compared to mean estimates obtained from randomly selected, pooled individual estimates. The results suggested that the pooled individual estimates were somewhat more accurate than those obtained from either of the interacting groups. At the same time, all individuals became more confident of their answers, suggesting the possibility of groupthink. No effects for different group sizes were found, possibly due to the constraints imposed by the structured nature of the two techniques.
A fundamental assumption in much of game theory and economics is that all the relevant information for determining the rational play of a game is contained in its structural description. Recent experimental studies of bargaining have demonstrated effects due to information not included in the classical models of games of complete information. The goal of the experiment reported here is to separate these effects into components that can be attributed to the possession of specific information by specific bargainers, and to assess the extent to which the observed behavior can be characterized as equilibrium behavior. The results of the experiment permit us to identify such component effects, in equilibrium, including effects that depend on whether certain information is common knowledge or not. The paper closes with some speculation on the causes of these effects.
This study explored the causes and effects of defecting from a stable coalition. Vulnerability and the relative power of the defectors were the primary independent variables. Results showed that defectors were obtaining higher payoffs than nondefectors prior to their defection. Defection from stable coalitions in the four coalition games studied here led to reduced benefits for both the defectors and the nondefectors. At the same time defectors fared better than nondefectors, with vulnerability reducing the defectors' benefits. Relative power, over all conditions, led to significantly more positive outcomes than relative equality or weakness. The data also suggested that vulnerable defectors may fare more poorly than their nondefecting partners. The use of coalitions as models of a variety of interactions and the social contexts that facilitate or block defections are also discussed.
The best strategy depends largely on the problem to be solved. Here's a four-step approach that will help clear the way
Most game-theoretic models of strategic interaction, indeed most economic models of any sort, specify potential outcomes entirely in terms of the preferences of the agents, as captured in their (possibly cardinal) utility functions. The underlying assumption is that the outcome of such interactions is determined entirely by these preferences, together with the strategic possibilities available to the agents. The purpose of this paper is to challenge the adequacy of this assumption in general, by investigating it in the specific context of two-person bargaining. In particular, we consider whether certain experimental results reported earlier can be accounted for strictly in terms of players' preferences and strategic possibilities, and we report a new experimental study designed to answer this question. The results strongly support the conclusion that sociological factors, unrelated to what we normally consider to be the ‘economic’ parameters of a game, can decisively influence the outcome of bargaining, in a systematic manner.
Examined the effects of manipulating the size of groups of 250 male undergraduate negotiators and their ability to communicate with one another in a game in which 1 player held veto power (i.e., had to be included in any winning coalition). The predictions of 3 models (core, value, and weighted probability) were tested. The game, in which the veto player could form a winning coalition with any other single player, was repeated for a series of 10 winning coalitions (trials). The effects for group size indicate significant differences between 3-person groups and 4-, 5-, and 6-person groups, and between all of these groups and 7-person groups. The increasing payoffs over trials were significant in the no-communication conditions, but no significant increases occurred in the conditions in which communication was available. The increasing payoffs for the veto player when communication opportunities were not available yielded the only support for the predictions of the core model; the overall payoffs consistently supported the predictions of the value and the weighted probability models.
Tested the predictions of 3 models of coalition behavior. 120 graduate students played each of 4 games, rotating among the 5 player positions (including a veto player) between games. The games were played under 1 of 3 time pressure/default conditions: (a) no time pressure, (b) a condition such that the constant payoff to coalitions was lost if an agreement was not reached in 3 attempts, and (c) a condition such that the payoff for no agreement was fixed at 60 points for the veto player and 10 for the other players. The veto players' payoffs varied over games and tended to increase as play continued, at times approaching the entire payoff. Thus, the weighted probability (S. S. Komorita, 1974) and Roth-Shapley (A. E. Roth, 1977; L. S. Shapley, 1953) models were not supported; the core model received some support. The default conditions had little effect. The likelihood of socially beneficial behavior in competitively motivating situations is discussed.
The effects of three factors on the reactions of group members toward a group decision and the support they subsequently accorded such decisions were examined. These factors were: (a) the decision process used by the group (majority rule with formal voting, discussion to unanimity, discussion to majority agreement, or choice by an experimenter appointed executor); (b) the level of prediscussion agreement among group members (concordance); and (c) the relative extent to which an individual's preferences were congruent with his group's decision (relative individual goal attainment, RIGA). Individual ratings of satisfaction with decisions, anticipated commitment to decisions, and the support accorded to decisions were highest in groups with initially high concordance and among individuals higher on RIGA. Members of groups required to reach unanimity and those run by executive choice exhibited comparable support for their groups' decisions across all levels of concordance. In groups required to discuss to majority agreement, support of the group decisions dropped markedly under low levels of concordance. Individual ratings of the difficulty experienced in reaching a decision were highest in groups required to reach unanimous agreement. However, members of these groups showed the greatest degree of postdecision change in the direction of their groups' decisions.
This paper examines a model of the prisoner's dilemma in which the nature of the Nash equilibria of the game can be varied. Two equilibrium indices are derived and are compared with two cooperation indices proposed by Rapoport and Chammah (1965). Preliminary experimental results indicate that the nature of the equilibria of a game affect the amount of cooperation.
This paper presents the results of an n-person characteristic function game played by between seven and and twelve players, one of whom was a monopolist. A factorial design allowed for analysis of the effects of group size, the availability of information, and communication opportunities for a series of seven trials. The data were compared to the game theoretic concepts of the core and Shapley value, (Shapley, 1953, Roth, 1977a), and to the predictions of the Weighted Probability model (Komorita, 1974). The findings indicated that the monopolist held a great deal of power, especially when communication among the players was not allowed. His payoffs increased over trials and approached the core in all of the conditions except when communicaion was available in seven and eight-person groups. The overall results were very close to the Shapley value and the predictions of the Weighted Probability model. The results were compared to an earlier study on a similar three-person game; increasing the group size seemed to be the primary case of the increase in the monopolist's payoffs.
Reviews 3 classes of game theoretic solution concepts (solutions, subsolutions, and the core; bargaining set models; and the Shapley value), 4 social psychological models of coalition formation (minimum resource theory, minimum power theory, bargaining theory, and the weighted probability model), and 3 sets of political coalition models (minimum size, minimum range, and policy distance minimization). The research on characteristic function games, experimental coalition situations involving more than 3 players, and on coalition governments is summarized, and the models are evaluated. The advantages of collaboration among the 3 areas are discussed
This study tested four models of coalition behavior in decision making on systems at the group level, in four different games, each of which was played by five individuals. Each game established a different distribution of power among the players. Data for coalition frequencies and for the payoffs of the players when they were included in a particular coalition supported Komorita & Chertkoff s (1973) bargaining theory over Komorita's (1974) weighted probability model and Gamson's (1961) minimum resource theory. When the overall payoffs received by each of the players in each game were used as a measure of the player's success in bargaining, the predictions of the Roth-Shipley and the weighted probability models received mixed support. Finally, the results suggested when the strength is weakness phenomenon might be expected to occur. Players with equal Shapley values but different resources within a particular game supported strength is weakness. Players with different Shapley values supported a strength is strength conclusion. Reports of the players provided a possible explanation of the underlying causes of this phenomenon.
This study investigated the effects of six communication/information conditions on the outcomes reached by three-person groups playing a characteristic function game. The game was played by a monopolist and two weaker players. The conditions consisted of six combinations which varied the amount of information available to the players and their ability to communicate with one another. The investigation focused on the effects of the independent variables and the relationship between the data and several game theoretic solution concepts. The results indicated that the monopolist's payoffs depended to a large extent on the communication/information conditions. Announcement of the payoff division and the availability of messages tended to reduce his payoffs. In conditions where no messages were allowed, the monopolist's payoffs increased over time. Although the data diverged significantly from the core, the situations which contributed to greater competition resulted in outcomes closer to the core. A comparison between von Neumann-Morgenstern solutions and the more general class of subsolutions indicated that subsolutions were more reflective of the behavior observed. In addition, the results over the entire set of conditions closely approximated the Shapley value, which has recently been shown to be a risk neutral player's expected utility for playing the game. Directions for future research were suggested.
The predictions of four theories of coalition behavior were compared to the results obtained from three coalition games conducted under two reference group conditions. Participants were given a set to use a reference group composed either of the other players in their experimental session or of players in other groups in the same position as themselves (“similar others”). While the different games had an impact on the accuracy of the theoretical predictions, the data as a whole tended to support Bargaining theory (Komorita & Chertkoff, 1973) and the Weighted Probability model (Komorita, 1974) over Minimum Resource theory (Gamson, 1961) and Minimum Power theory (Shapley & Shubik, 1954). The results also indicated that a reference group of “similar others” led to more accurate theoretical predictions and to higher payoffs for the powerful player in each of the games, even though his demands were higher in these conditions. The use of four-person coalition games in coalition research was also discussed.
A team theoretic model that establishes a criterion (decision rule) for a financial institution branch to report exceptional loan requests to headquarters for action was compared to such choices made by graduate industrial management students acting as financial vice-presidents. Results showed that the loan size criterion specified by subjects was typically greater than the optimal criterion when the optimal criterion was small relative to the maximum loan size, and less than the optimal criterion when the optimal criterion was large relative to the maximum loan size. That is, subjects specified criteria that would result in the reporting of too few exceptions (a case of informational overdecentralization) and too many exceptions (a case of informational overcentralization) when the optimal criterion was relatively small and large respectively. The behavior exhibited was attributed in part to a subject's utility function, which was inferred to have a Friedman-Savage double inflected structure.
The present study manipulated two variables, leadership involvement in subordinates' discussion of a problem and the importance of the subordinate's task as perceived by the subordinate, and assessed their impact on two measures (quantity and quality) of productivity and a set of affective questionnaire items. Undergraduate students led by graduate student leaders (whose involvement varied over four levels) discussed problems in their curriculum and individually suggested solutions to these problems. The results showed that: (1) individuals led by a more involved leader produced more than individuals led by a less involved leader; (2) a task rated as very important resulted in increases in both the quantity and the quality of performance; (3) a significant interaction between the two variables indicated that the poorest quality performance was evidenced when the leader was involved and the task was of low importance; and (4) both variables had an impact on affective responses (e.g., self-reports of satisfaction) in the expected direction. The findings were discussed with respect to the appropriate strategies for optimal performance.
An experiment involving a simulated financial lending organization was used to compare subjects' decisions regarding the degree to which the organization should be informationally centralized, with that of an optimal decision rule based on a normative model. Typically, subjects overdecentralized the information structure to process small loans, and overcentralized the information structure to process large loans. Both the theoretical and practical implications of these results are discussed.
The present experiment pitted three choice shift hypotheses against one another in an attempt to eliminate one or more of the hypotheses and find support for those remaining. Subjects responded three times to the twelve CDQ items, once as a pretest, and twice following presentation of homogenous sets of three arguments which advocated either a risky or a cautious position. The risk-as-value, relevant arguments, and conformity-attitude change hypotheses generated three separate predictions for the subjects' responses. Results mirrored the prediction of the relevant arguments hypothesis: New information, whether it is contained in cautious or risky arguments, caused a shift toward the type of argument presented. The risk-as-value and the conformity-attitude change hypotheses could not explain the present data.
This paper presents a new, motivated attributions model of trust development. The model builds on two simple insights: that the parties in a potentially trusting relationship are likely to view their interaction differently and that their attributions of each other’s behavior will be self-servingly motivated. The model specifically focuses on the role of dependence in motivating attributions of trustworthiness, suggesting, for instance, that people ameliorate the anxiety associated with dependence by perceiving others as trustworthy. The model explains why trustors, contrary to the prescriptions of the dominant, rational choice approach, often engage in large, seemingly irrational acts of trust and when and why these acts, despite being tremendously risky, can be crucial to trust development. The paper explores the consequences of these insights for interpersonal interactions as well as touching on the potential for extensions to inter-organizational and international interactions.
This paper reports the results of a study investigating how a sample of successful real estate agents reacted to two ethical dilemmas; the findings formed the basis and a first evaluation of a model of bounded personal ethics. The agents faced dilemmas of whether to (1) reveal to purchasers their primary, legal responsibilities to sellers, increasing their chances of losing the purchasers as clients and (2) comply with regulations and reveal to sellers the information they had about purchasers, violating any personal connections they had established with them. Interviews indicated that agents tended to hide their responsibilities to sellers but still protected purchasers’ interests; in essence, they acted illegally on both issues. These findings led to the formation of a model of bounded personal ethics which predicts that agents (and people in general) will act ethically until their ethical values conflict with their self-interest; then, with increasing financial temptations, self-interested action becomes more likely and often occurs almost automatically. Reanalysis of the data revealed many instances of bounded personal ethics. The paper concludes with a discussion of the dynamics of agency relationships and the implications of bounded personal ethics.
-
Teaching Materials
426-1 Economics, Social Psychology, and their Experiments Fall 2009
This course focuses on individual behavior in organizational settings. Topics for analysis include social cognition, decision making, negotiation groups, social influence, norms, fairness and equity theory. Recent empirical research will be evaluated in each of these areas, and implications will be studied in terms of theoretical advancement, empirical study and practical applications.
Special Topics in Management and Organizations: Economics, Social, Psychology and their Experiments (MORS-530-0)
Doctoral-level course offered on a one-time basis dealing with a special topic in the Management and Organizations field. Addresses a specific need within the program’s curriculum and/or a trend in the field.
This course counts toward the following majors: Management & Organizations.
This course combines the approaches of game theory, organization behavior and negotiations to improve students' strategy formation and implementation in a variety of bargaining games. Using hands-on exercises, students are taught to construct and enact immediate strategies, learn how well they have bargained and analyze their actions in depth. The course provides experiences in competitive and cooperative negotiations and culminates in an extended, complex bargaining simulation
This course focuses on individual behavior in organizational settings. Topics for analysis include social cognition, decision making, negotiation groups, social influence, norms, fairness and equity theory.
Ethics and Leadership (MORSX-472-0)
Ethics and Leadership examines the anatomy of leadership in modern organizations, highlighting the ethical challenges facing corporate leaders in the rapidly changing business environment.
PHONE: 847-467-3566
FAX: 847-491-8896
Jacobs Center Room 360