Mind Games: The Mental Representation of Conflict Nir Halevy Stanford University Eileen Y. Chou and J. Keith Murnighan Northwestern University Perception and misperception play a pivotal role in conflict and negotiation. We introduce a framework that explains how people think about their outcome interdependence in conflict and negotiation and how their views shape their behavior. Seven studies show that people’s mental representations of conflict are predictably constrained to a small set of possibilities with important behavioral and social consequences. Studies 1 and 2 found that, when prompted to represent a conflict in matrix form, more than 70% of the people created 1 of 4 archetypal mixed-motive games (out of 576 possibilities): Maximizing Difference, Assurance, Chicken, and Prisoner’s Dilemma. Study 3 demonstrated that these mental representations relate in predictable ways to negotiators’ fixed-pie perceptions. Studies 4 – 6 showed that these mental representations shape individuals’ behavior and interactions with others, including cooperation, perspec- tive taking, and use of deception in negotiation, and through them, conflict’s outcomes. Study 7 found that the games that people think they are playing influence how their counterparts see them, as well as their counterparts’ negotiation expectations. Overall, the findings document noteworthy regularities in people’s mental representations of outcome interdependence in conflict and illustrate that 4 archetypal games can encapsulate fundamental psychological processes that emerge repeatedly in conflict and negotiation. Keywords: conflict, negotiation, mental representations, outcome interdependence, mixed-motive games Outcome interdependence—the compatibility or incompatibility of people’s interests and goals—is a defining element of conflict situations (Bornstein, 2003; De Dreu, 2010). Indeed, influential theoretical approaches to conflict, including interdependence the- ory (Kelley & Thibaut, 1978) and game theory (Luce & Raiffa, 1957), rely on systematic and nuanced variations of the parties’ outcome interdependence to analyze and study social interactions. Although there is considerable consensus that outcome interdepen- dence is a critical feature of social interactions and that subjective perceptions play a pivotal role in shaping the course and outcomes of conflicts (Deutsch, 1973; Jervis, 1976; Ross & Ward, 1996), research at the intersection of these two key observations is sparse: Little is known about individuals’ subjective perceptions of their outcome interdependence. The current research investigates individuals’ subjective percep- tions of their outcome interdependence, as well as the behavioral and social consequences of these perceptions, with a particular focus on dyadic negotiation. We first identify a small set of interdependence structures that seem to have psychological prom- inence in people’s minds when they think about conflict and negotiation situations. Then we investigate the consequences of these conflict perceptions by studying how particular outcome interdependence perceptions affect negotiator behavior and, ulti- mately, a conflict’s outcomes. Thus, this article addresses two important questions: (a) What games do people think they are playing? 1 and (b) What are the behavioral consequences of view- ing conflict in terms of a particular game? From Matrix Transformations to Functional Construal Researchers often use matrix games—abstract representations of interactions that include a set of players, the strategies available to each player, and the payoffs that are associated with their actions—to study conflict (Camerer, 2003; Plott & Smith, 2008). Although researchers commonly recognize that “how competitors define the game may be more important than the moves they make within the game” (Bazerman, Curhan, Moore, & Valley, 2000, p. 286), research in behavioral game theory typically has focused on the decisions that individuals and groups make within a given game rather than on how players define the game, plausibly because economics has traditionally been more interested in action than in cognition (cf. Chou, McConnell, Nagel, & Plott, 2009; Devetag & Warglien, 2008). Previous research emanating from interdependence theory, how- ever, has investigated individuals’ tendencies to mentally trans- form exogenously determined payoff matrices (i.e., the objective, “given” matrices) and make their decisions upon the new, “effec- tive” matrices (Kelley et al., 2003; Kelley & Thibaut, 1978). For 1 This question is considered one of the top 10 open research questions in behavioral game theory (Camerer, 2003, p. 474). This article was published Online First September 12, 2011. Nir Halevy, Graduate School of Business, Stanford University; Eileen Y. Chou and J. Keith Murnighan, Kellogg School of Management, North- western University. This research was supported by the Dispute Resolution Research Center at the Kellogg School of Management and benefitted from helpful sugges- tions by Evan Apfelbaum, Guy Arie, Jonathan Bendor, Colin Camerer, Taya Cohen, Adam Galinsky, Dale Miller, Benoit Monin, Christian Wheeler, and Frank Yates. Correspondence concerning this article should be addressed to Nir Halevy, Graduate School of Business, Stanford University, 518 Memorial Way, Stanford, CA 94305-5015. E-mail: nhalevy@standford.edu Journal of Personality and Social Psychology © 2011 American Psychological Association 2012, Vol. 102, No. 1, 132–148 0022-3514/11/$12.00 DOI: 10.1037/a0025389 132