A Naturalistic Approach to Adversarial Behavior: Modeling the Prisoner s Dilemma

Size: px
Start display at page:

Download "A Naturalistic Approach to Adversarial Behavior: Modeling the Prisoner s Dilemma"

Transcription

1 A Naturalistic Approach to Adversarial Behavior: Modeling the Prisoner s Dilemma Amy Santamaria Walter Warwick Alion Science and Technology MA&D Operation 4949 Pearl East Circle, Suite 300 Boulder, CO asantamaria@alionscience.com, wwarwick@alionscience.com Keywords: Naturalistic Decision Making, Human Performance Modeling, Prisoner s Dilemma ABSTRACT: In this paper we present a model of the prisoner s dilemma in which human-like, adversarial behaviors emerge from interactions between two naturalistic models. In the prisoner s dilemma, like other simple two-player games, the behaviors that arises from iterated play can be surprisingly complex. We developed a model of this task using an integrated architecture, with a task network representing the decision environment and an underlying naturalistic model of decision making determining branching in the task network. Several adaptive strategies emerged in our two-player models. In addition, we captured several qualitative effects of human performance. 1. Introduction In our previous work we have focused on demonstrating how simple, naturalistic mechanisms can produce interesting emergent behaviors that match human performance (Warwick 2006; Warwick & Santamaria, 2006; Warwick & Hutchins, 2004). Our goal in this work was to study the more complex adversarial behaviors that emerge when we play two naturalistic models against each other, with each model making decisions based on its opponent s past behavior. In this paper, we investigate how two such models interact in a common environment and generate meaningful emergent behaviors. In particular, we examined the prisoner s dilemma, a non zero-sum game studied in game theory and economics, which has been applied to many fields, including psychology, sociology, political science, and philosophy. The prisoner s dilemma is interesting from the viewpoint of military strategy because it is an example of dynamic, reciprocal, adversarial interaction based on a few simple rules. In fact, the prisoner s dilemma has been used to study arms races and other real-life military and political conflicts (see Lumsden, 1973; Smith, Sola, & Martin, 2000). It is also interesting from a naturalistic viewpoint because actual human behavior often diverges from what is predicted by a rational analysis of the prisoner s dilemma. Here we follow in the footsteps of other cognitive modelers who have demonstrated that general assumptions about the nature of cognition can lead to surprising predictions about human behavior that would be hard to come by given the normative analysis of game theory. For example, Lebiere, Wallach, and West (2000) show how memory can account for many of the qualitative aspects of adversarial behavior in 2x2 game play. That important work sets the context for our own work. 1.1 The Prisoner s Dilemma The prisoner s dilemma is a two-player game that has been used to examine adversarial behavior. In the classic form of the game, two people are arrested for a crime, but there is not enough evidence to convict without the testimony of at least one of them. Each is offered a deal: testify against the partner, and if the partner stays silent, the one who testifies goes free and the partner gets the maximum sentence. If both prisoners agree to testify, both get a reduced sentences. If both prisoners stay silent, both get a minimum sentence. The two prisoners cannot communicate about their choice. The optimal solution is for both prisoners to stay silent. However, from the perspective of a single prisoner who does not know what his partner will do, it makes more sense to agree to testify. If the prisoner agrees to testify and his partner stays silent, then he goes free. If his partner also agrees to testify, then he avoids the maximum

2 sentence. The prisoner s dilemma can also be represented abstractly as a 2x2 game (two players with two choices) with a payoff structure. Two players, who cannot communicate, can choose to either cooperate or defect, and their goal is to maximize their payoff. We used the payoff structure shown in Table 1. (The specific numeric values of the pay-offs are incidental; it only matters that the matrix preserves certain ratios between pay-offs): A Cooperate B Cooperate A Payoff=3 B Payoff =3 B Defect A Payoff =0 B Payoff=4 A Defect A Payoff=4 B Payoff =0 A Payoff =1 B Payoff=1 Table 1.1 Payoff structure for the prisoner s dilemma. presented. A similarity value (i.e., a dot product) is computed between the vector representing the new situation and each remembered situation in long term memory. The resulting similarity value is then raised to a user-defined power and used to determine the proportionate contribution that each remembered episode makes to a composite recollection. The result is then analyzed to produce an output corresponding to a specific course of action, which is implemented, evaluated, and stored as a new episode in long term memory for use in the next decision. Our task network model of the prisoner s dilemma is depicted in Figure 2.1. Only the skeleton of the task is visible because the two naturalistic decision making models, one for each of the two players, are implemented as plug-in components to the task network. (The structure of the naturalistic decisions is discussed in more detail in section 3.) While the behavior that emerges from a single iteration of the prisoner s dilemma is trivial, the adversarial behavior that emerges from iterated play can be surprisingly complex and subtle. Given two adaptive players, a type of feedback becomes possible in which the each player is simultaneously influencing and influenced by the play of the other. The behaviors that emerge from this so-called reciprocal causation (Lebiere & West, 1999) can be very difficult to unpack because of their dynamic, circular nature something like a chicken-and-egg problem with a dash of revenge. 2. Model Description We developed our models using a Micro Saint Sharp task network environment augmented with a naturalistic model of decision making (see Warwick, McIlwaine, Hutton & McDermott, 2001). In this integrated architecture, the task network is used to represent the decision environment, and the naturalistic model uses a similaritybased recognition to determine a course of action in the task network, using a store of decision making episodes which is populated by experience. Our approach to modeling naturalistic decision making was inspired by Klein s (1998) theory of the recognitionprimed decision and implemented by extending Hintzman s multiple-trace memory model (1984; 1986a; 1986b). A decision maker s long-term memory is represented as a set of episodes, each of which represents the situation that prompted a decision (encoded as a cue vector), the course of action taken, and an outcome measure (either successful or unsuccessful). There may be multiple copies of identical experiences in long-term memory. Recognition occurs when a new situation is Figure 2.1 Micro Saint task network model of the prisoner's dilemma. The heavy lines branching into Defect and Cooperate indicate where information is sent to a naturalistic decision making model. 3. Method On each trial, the modeled player must decide between defecting and cooperating. The naturalistic model learns by reinforcement; successful decisions are rewarded while unsuccessful decisions are punished. Thus, it was necessary to define criteria (transparent to the model itself) for a successful decision. Is success beating your opponent? Is it doing as well as your opponent? Is it beating some threshold? We investigated several different definitions of success: MATCH, BEAT and OPPORTUNISTIC. In MATCH, a player s decision is successful if the payoff is greater than or equal to its opponent s payoff. This rewards mutual

3 cooperation, mutual defection, or defection when the opponent cooperates. In BEAT, a player s decision is successful if the payoff is greater than its opponent. This never rewards cooperation, and it only rewards defection when the opponent cooperates. In OPPORTUNISTIC, a player s decision to defect is rewarded only if his opponent cooperates, while a decision to cooperate is rewarded only if his opponent does not defect. The first two definitions are straightforward, while the third is a bit more subtle in terms of the behavior it engenders. In addition to defining the various success criteria for the model, it was also necessary to define cue structures, the information the decision maker uses to prompt recognition. Our default model used a set of six cues, representing its own and its opponent s plays on the past three trials. 3.1 Testing for Emergent Strategies Our first goal was to observe the emergence of strategies as this default player adapted to its opponent s behavior. We tested the default player against two types of opponents: opponents whose moves were determined purely probabilistically and opponents with varying memory capacities. Each run of the model was 100 trials long, or 100 sequential prisoner s dilemma choices. First, we tested the default Player A against a player whose moves were determined probabilistically. Player B, the probabilistic player, defected with a probability of 0, 0.25, 0.5, 0.75, or 1, depending on the condition, and Player A was rewarded according to MATCH or BEAT definitions of success, for a total of 10 conditions. Each condition was run with 10 different random seeds, for a total of 100 runs (see Table 3.1). Player B Strategy BEAT MATCH P(defect) = runs 10 runs P(defect) = runs 10 runs P(defect) = runs 10 runs P(defect) = runs 10 runs P(defect) = runs 10 runs Table 3.1 Testing the model against a probabilistic opponent. Second, we tested the default Player A against players with limited memory (i.e., shorter cue vectors) under different combinations of success criteria. Fixing the memory capacity of the opponent, Player B, to only one previous trial, we tested all combinations of MATCH and BEAT definitions of success for both players, for a total of 4 conditions (MATCH-MATCH, BEAT-BEAT, MATCH-BEAT, and BEAT-MATCH). Each condition was run with 10 different random seeds, for a total of 40 runs (see Table 3.2). A: BEAT A: MATCH B: BEAT 10 runs 10 runs B: MATCH 10 runs 10 runs Table 3.2 Testing the model (Player A) against a limited memory opponent (Player B). 3.2 Evaluating the Model s Choices Our second goal was to evaluate the model s choices in order to make a comparison to human performance on the prisoner s dilemma task. To do this, we fixed the success criteria to OPPORTUNISTIC and varied the memory of the opponent, Player B, by systematically reducing the number of previous moves used as cues. Each condition was run with 20 different random seeds, for a total of 60 runs (see Table 3.3). Player B Cues OPPORTUNISTIC 0,1 20 runs 0,2 20 runs 0,3 20 runs Table 3.3 Testing the model against a limited memory opponent using OPPORTUNISTIC criteria. Executable versions of the different integrated models can be inspected and downloaded from: 4. Results and Discussion 4.1 Testing for Emergent Strategies To understand the results, it is helpful to understand the optimal strategy for both definitions of success. With MATCH as the definition of success, the optimal strategy is to always defect. Defecting means your payoff is either higher or the same as your opponent s payoff, both of which are considered successful in MATCH. With BEAT as the definition of success, the optimal strategy is to switch between defecting and cooperating. Because in BEAT your payoff has to be strictly larger than your opponent s payoff, defecting only results in success if your opponent cooperates. To encourage your opponent to cooperate, you must also cooperate some of the time. When the default Player A was tested against a probabilistic opponent with MATCH as the definition of success (see Figure 4.1), it showed that it was able to learn from its opponent s behavior and to choose the

4 appropriate strategy. When the opponent s probability of defecting was zero (it always cooperated), Player A switched between defecting and cooperating, defecting about half the time. When the opponent s probability of defecting was one (it always defected), Player A learned to defect most of the time. In the intermediate condition where the opponent defected 75 percent of the time, Player A tended to defect more than not; in the other intermediate conditions, Player A switched between cooperation and defection. Figure 4.2 The default player against a probabilistic opponent using the BEAT definition of success. Figure 4.1 The default player against a probabilistic opponent using the MATCH definition of success. When the default Player A was tested against a probabilistic opponent with BEAT as the definition of success (see Figure 4.2), it showed the reverse pattern. When the opponent s probability of defecting was one (it always defected), Player A switched between defecting and cooperating, defecting about half the time. When the opponent s probability of defecting was zero (it always cooperated), Player A learned to defect most of the time. As before, the intermediate conditions led to mixed behaviors. The default Player A was also able to converge on adaptive strategies when it was matched with a player with reduced memory. When success was defined as MATCH for both players, Player A learned to always defect for all of the ten runs, while Player B only learned this for only four of the ten runs. When success was defined as BEAT for both players, Player A learned to switch between defecting and cooperating on all of the ten runs, while Player B failed to settle on a strategy. When success was defined as BEAT for Player A and MATCH for Player B, Player A learned to defect about half the time on all of the ten runs, while Player B failed to settle on a strategy. When success was defined as MATCH for Player A and BEAT for Player B, Player A chose to defect most of the time, while Player B failed to settle on a strategy. 3.2 Evaluating the Model s Choices While the emergence of these various strategies demonstrates the naturalistic model s ability to adapt in adversarial contexts, it is still not clear whether this ability to adapt reflects human-like performance. To get at this question, we had to devise a way to score outcomes of each decision. In doing this, we hoped to reconcile our work with previous attempts to model simple games (e.g., Lebiere & West, 1999; West, Lebiere & Bothell, 2006). Toward this end, we fixed the success criteria for both Players A and B to OPPORTUNISTIC, and we defined a cumulative outcome measure such that defecting when the opponent cooperated added 1 point to a player s score, cooperating when the opponent defected subtracted 1 point from a player s score, and mutual cooperation or

5 defection resulted in no change to the score. We then ran the default Player A (with a small change to the cue structure in which only two, rather than three of the players own previous moves were remembered) against opponents of varying memory capacity. a model that implements a fixed tit-for-tat strategy. The explanation again lies in the complexity of the patternmatching. More cues mean more complex patterns to learn, which ultimately slows down the leaning process. 5. Conclusions The goal of this work has been to demonstrate how fairly simple mechanisms can engender surprisingly complex emergent behaviors that are adaptive and that reflect some of the qualitative aspects of human game play performance. Compared to previous attempts using cognitive models to account for adversarial behavior, our approach lacks some theoretical overhead (e.g., the activation calculus used in the ACT-R models of game play). This is both a blessing and a curse. We make fewer assumptions about the underlying cognitive process; however, fewer assumptions also leader to leaner foundations for explaining the effects that emerge. Figure 4.3 Cumulative score for the default player against an opponent with different levels of reduced memory, where memory is the number of previous moves (own, opponent s) used to prompt recognition. We captured two of the main qualitative effects reported by West, Lebiere and Bothell (2006). The results are shown in Figure 4.3 First, the average difference in score between opponents of equal or near-equal memory capacity is near zero, the result of averaging across many individual random walks, where a random walk is a series of successive, randomly determined steps. (We have not yet investigated the distribution of outcomes to see whether the bimodal pattern reported in Lebiere, Wallach & West (2000) is present under all the conditions we tested.) Second, having relatively more memory confers a systematic advantage. Understanding how this happens is not straightforward. Lebiere, Wallach & West (2000) identify the ability to process a greater number of previous moves and the ability to detect sequential dependencies as the driver of this behavior. The smarter model can exploit the lesser model s poor ability to predict its behavior. In our case, we believe the advantage emerges because a larger cue set allows more finely-tuned pattern recognition, while the reduced-memory model suffers because it conflates past successes and failures. We are currently working to verify this explanation. Conversely, we have observed cases where a reducedmemory model converges on a successful strategy more quickly than larger-memory models when playing against Another unexpected result of this work was the realization that emergent behavior still requires some built-in constraints. Or, to put it another way, cognitive modeling will never be free of a certain amount of knowledge engineering, even for something as seemingly straightforward as simple game play. Whether this means encoding game strategy as procedural knowledge or understanding how to reinforce particular behaviors, there will always be a fair amount of creativity required to make a model go. 6. References Hintzman, D. L. (1984). MINERVA 2: A simulation model of human memory. Behavior Research Methods, Instruments & Computers, 16, Hintzman, D. L. (1986a). Judgments of Frequency and Recognition Memory in a Multiple-Trace Memory Model. Eugene, OR: Institute of Cognitive and Decision Sciences. Hintzman, D. L. (1986b). "Schema Abstraction" in a Multiple-Trace Memory Model. Psychological Review, 93, Klein, G. (1998). Sources of Power: How People Make Decisions. Cambridge, MA: The MIT Press. Lebiere, C., Wallach, D., & West, R. L. (2000). A memory-based account of the prisoner s dilemma and other 2x2 games. In Proceedings of the 3rd International Conference on Cognitive Modeling,

6 Lebiere, C. & West, R. L. (1999). A dynamic ACT-R model of simple games. In Proceedings of the 21st Conference of the Cognitive Science Society, Lumsden, M. (1973). The Cyprus conflict as a prisoner s dilemma game. Journal of Conflict Resolution, 17, WALTER WARWICK is a Senior Systems Analyst at Alion Science and Technology. He is working on several projects having to do with the modeling and simulation of human behavior. He received his Ph.D. in History and Philosophy of Science, an Area Certificate in Pure and Applied Logic, and an M.S. in Computer Science from Indiana University. Smith, R., Sola, M. & Spagnolo, F. (2000). The prisoner s dilemma and regime-switching in the Greek-Turkish arms race. Journal of Peace Research, 37, Warwick, W. (2006). A bad Hempel day: The decoupling of explanation and prediction in computational cognitive modeling. In Proceedings of the Fall 2006 Simulation Interoperability Workshop. Warwick, W. & Hutchins, S. (2004). Initial comparisons between a "naturalistic" model of decision making and human performance data. In Proceedings of the 13th Conference on Behavior Representation in Modeling and Simulation. Warwick, W., McIlwaine, S., Hutton, R. J. B., & McDermott, P. (2001). Developing computational models of recognition-primed decision making. In Proceedings of the 10th Conference on Computer Generated Forces. Warwick, W. & Santamaria, A. (2006). Giving up vindication in favor of application: Developing cognitively-inspired widgets for human performance modeling tools. Proceedings of the 7th International Conference on Cognitive Modeling. West, R. L., Lebiere, C., & Bothell, D. J. (2006). Cognitive architectures, game playing, and human evolution. In R. Sun (ed.), Cognition and Multi- Agent Interaction: From Cognitive Modeling to Social Simulation, New York: Cambridge University Press. Author Biographies AMY SANTAMARIA is a Research Cognitive Scientist at Alion Science and Technology. Her research focuses on modeling human behavior and cognition. She received her Ph.D. in Cognitive Psychology and Neuroscience and an M.A. in Cognitive Psychology from the University of Colorado Boulder.

The Evolution of Cooperation: The Genetic Algorithm Applied to Three Normal- Form Games

The Evolution of Cooperation: The Genetic Algorithm Applied to Three Normal- Form Games The Evolution of Cooperation: The Genetic Algorithm Applied to Three Normal- Form Games Scott Cederberg P.O. Box 595 Stanford, CA 949 (65) 497-7776 (cederber@stanford.edu) Abstract The genetic algorithm

More information

Cognitive modeling versus game theory: Why cognition matters

Cognitive modeling versus game theory: Why cognition matters Cognitive modeling versus game theory: Why cognition matters Matthew F. Rutledge-Taylor (mrtaylo2@connect.carleton.ca) Institute of Cognitive Science, Carleton University, 1125 Colonel By Drive Ottawa,

More information

Modeling Social Information in Conflict Situations through Instance-Based Learning Theory

Modeling Social Information in Conflict Situations through Instance-Based Learning Theory Modeling Social Information in Conflict Situations through Instance-Based Learning Theory Varun Dutt, Jolie M. Martin, Cleotilde Gonzalez varundutt@cmu.edu, jolie@cmu.edu, coty@cmu.edu Dynamic Decision

More information

The iterated Prisoner s dilemma

The iterated Prisoner s dilemma The iterated Prisoner s dilemma U. Sperhake DAMTP, University of Cambridge PHEP Seminar, University of Cambridge 22 nd March 2013 U. Sperhake (DAMTP, University of Cambridge) The iterated Prisoner s dilemma

More information

Book review: The Calculus of Selfishness, by Karl Sigmund

Book review: The Calculus of Selfishness, by Karl Sigmund Book review: The Calculus of Selfishness, by Karl Sigmund Olle Häggström Leading game theorist Karl Sigmund calls his latest book The Calculus of Selfishness, although arguably The Calculus of Cooperation

More information

IN THE iterated prisoner s dilemma (IPD), two players

IN THE iterated prisoner s dilemma (IPD), two players IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION, VOL. 11, NO. 6, DECEMBER 2007 689 Multiple Choices and Reputation in Multiagent Interactions Siang Yew Chong, Member, IEEE, and Xin Yao, Fellow, IEEE Abstract

More information

Color Cues and Viscosity in. Iterated Prisoner s Dilemma

Color Cues and Viscosity in. Iterated Prisoner s Dilemma Color Cues and Viscosity in Iterated Prisoner s Dilemma Michael Joya University of British Columbia Abstract The presence of color tag cues and a viscous environment have each been shown to foster conditional

More information

games ISSN

games ISSN Games 2011, 2, 21-51; doi:10.3390/g2010021 Article OPEN ACCESS games ISSN 2073-4336 www.mdpi.com/journal/games Intergroup Prisoner s Dilemma with Intragroup Power Dynamics Ion Juvina 1, *, Christian Lebiere

More information

The Game Prisoners Really Play: Preference Elicitation and the Impact of Communication

The Game Prisoners Really Play: Preference Elicitation and the Impact of Communication The Game Prisoners Really Play: Preference Elicitation and the Impact of Communication Michael Kosfeld University of Zurich Ernst Fehr University of Zurich October 10, 2003 Unfinished version: Please do

More information

Effects of Sequential Context on Judgments and Decisions in the Prisoner s Dilemma Game

Effects of Sequential Context on Judgments and Decisions in the Prisoner s Dilemma Game Effects of Sequential Context on Judgments and Decisions in the Prisoner s Dilemma Game Ivaylo Vlaev (ivaylo.vlaev@psy.ox.ac.uk) Department of Experimental Psychology, University of Oxford, Oxford, OX1

More information

Irrationality in Game Theory

Irrationality in Game Theory Irrationality in Game Theory Yamin Htun Dec 9, 2005 Abstract The concepts in game theory have been evolving in such a way that existing theories are recasted to apply to problems that previously appeared

More information

Clicker quiz: Should the cocaine trade be legalized? (either answer will tell us if you are here or not) 1. yes 2. no

Clicker quiz: Should the cocaine trade be legalized? (either answer will tell us if you are here or not) 1. yes 2. no Clicker quiz: Should the cocaine trade be legalized? (either answer will tell us if you are here or not) 1. yes 2. no Economic Liberalism Summary: Assumptions: self-interest, rationality, individual freedom

More information

Introduction to Game Theory Autonomous Agents and MultiAgent Systems 2015/2016

Introduction to Game Theory Autonomous Agents and MultiAgent Systems 2015/2016 Introduction to Game Theory Autonomous Agents and MultiAgent Systems 2015/2016 Ana Paiva * These slides are based on the book by Prof. M. Woodridge An Introduction to Multiagent Systems and the online

More information

arxiv: v1 [cs.gt] 14 Mar 2014

arxiv: v1 [cs.gt] 14 Mar 2014 Generalized prisoner s dilemma Xinyang Deng a, Qi Liu b, Yong Deng a,c, arxiv:1403.3595v1 [cs.gt] 14 Mar 2014 a School of Computer and Information Science, Southwest University, Chongqing, 400715, China

More information

UC Merced Proceedings of the Annual Meeting of the Cognitive Science Society

UC Merced Proceedings of the Annual Meeting of the Cognitive Science Society UC Merced Proceedings of the Annual Meeting of the Cognitive Science Society Title Learning to cooperate in the Prisoner's Dilemma: Robustness of Predictions of an Instance- Based Learning Model Permalink

More information

Structure, Agency and Network Formation: A Multidisciplinary Perspective

Structure, Agency and Network Formation: A Multidisciplinary Perspective Structure, Agency and Network Formation: A Multidisciplinary Perspective Jason Greenberg, NYU University of Pennsylvania, Wharton School 7.25.13 1 Hockey-stick growth in social network research. 900 800

More information

The Evolution of Cooperation

The Evolution of Cooperation Cooperative Alliances Problems of Group Living The Evolution of Cooperation The problem of altruism Definition of reproductive altruism: An individual behaves in such a way as to enhance the reproduction

More information

Chapter 2 Various Types of Social Dilemma

Chapter 2 Various Types of Social Dilemma Chapter 2 Various Types of Social Dilemma In order to examine the pragmatic methods of solving social dilemmas, it is important to understand the logical structure of dilemmas that underpin actual real-life

More information

Title. Author(s)Takahashi, Nobuyuki. Issue Date Doc URL. Type. Note. File Information. Adaptive Bases of Human Rationality

Title. Author(s)Takahashi, Nobuyuki. Issue Date Doc URL. Type. Note. File Information. Adaptive Bases of Human Rationality Title Adaptive Bases of Human Rationality Author(s)Takahashi, Nobuyuki SOCREAL 2007: Proceedings of the International Works CitationJapan, 2007 / editor Tomoyuki Yamada: 1(71)-34(104) Issue Date 2007 Doc

More information

Learning to Use Episodic Memory

Learning to Use Episodic Memory Learning to Use Episodic Memory Nicholas A. Gorski (ngorski@umich.edu) John E. Laird (laird@umich.edu) Computer Science & Engineering, University of Michigan 2260 Hayward St., Ann Arbor, MI 48109 USA Abstract

More information

How to Cope with Noise In the Iterated Prisoner's Dilemma

How to Cope with Noise In the Iterated Prisoner's Dilemma How to Cope with Noise In the Iterated Prisoner's Dilemma Jianzhong Wu Institute of Automation Chinese Academy of Sciences and Robert Axelrod Institute of Public Policy Studies University of Michigan July

More information

EMOTION AS AN ENABLER OF CO-OPERATION

EMOTION AS AN ENABLER OF CO-OPERATION EMOTION AS AN ENABLER OF CO-OPERATION Martyn Lloyd-Kelly 1, Katie Atkinson 1 and Trevor Bench-Capon 1 1 Department of Computer Science, The University of Liverpool, Ashton Street, Liverpool, United Kingdom

More information

Cooperation in Prisoner s Dilemma Game: Influence of Social Relations

Cooperation in Prisoner s Dilemma Game: Influence of Social Relations Cooperation in Prisoner s Dilemma Game: Influence of Social Relations Maurice Grinberg (mgrinberg@nbu.bg) Evgenia Hristova (ehristova@cogs.nbu.bg) Milena Borisova (borisova_milena@abv.bg) Department of

More information

Disjunction Effect in Prisoners Dilemma: Evidences from an Eye-tracking Study

Disjunction Effect in Prisoners Dilemma: Evidences from an Eye-tracking Study Disjunction Effect in Prisoners Dilemma: Evidences from an Eye-tracking Study Evgenia Hristova (ehristova@cogs.nbu.bg) Maurice Grinberg (mgrinberg@nbu.bg) Central and Eastern European Center for Cognitive

More information

DIFFERENCES IN THE ECONOMIC DECISIONS OF MEN AND WOMEN: EXPERIMENTAL EVIDENCE*

DIFFERENCES IN THE ECONOMIC DECISIONS OF MEN AND WOMEN: EXPERIMENTAL EVIDENCE* DIFFERENCES IN THE ECONOMIC DECISIONS OF MEN AND WOMEN: EXPERIMENTAL EVIDENCE* Catherine C. Eckel Department of Economics Virginia Tech Blacksburg, VA 24061-0316 Philip J. Grossman Department of Economics

More information

The Role of Implicit Motives in Strategic Decision-Making: Computational Models of Motivated Learning and the Evolution of Motivated Agents

The Role of Implicit Motives in Strategic Decision-Making: Computational Models of Motivated Learning and the Evolution of Motivated Agents Games 2015, 6, 604-636; doi:10.3390/g6040604 Article OPEN ACCESS games ISSN 2073-4336 www.mdpi.com/journal/games The Role of Implicit Motives in Strategic Decision-Making: Computational Models of Motivated

More information

Reciprocity, Cooperation, and Reputation

Reciprocity, Cooperation, and Reputation Reciprocity, Cooperation, and Reputation Reciprocal altruism Prisoner s dilemma: repeated play allows cooperation Generosity, punishment, spite (Ultimatum games) Reputation Collective action and punishment

More information

A Cognitive Model of Strategic Deliberation and Decision Making

A Cognitive Model of Strategic Deliberation and Decision Making A Cognitive Model of Strategic Deliberation and Decision Making Russell Golman (rgolman@andrew.cmu.edu) Carnegie Mellon University, Pittsburgh, PA. Sudeep Bhatia (bhatiasu@sas.upenn.edu) University of

More information

Conditional behavior affects the level of evolved cooperation in public good games

Conditional behavior affects the level of evolved cooperation in public good games Center for the Study of Institutional Diversity CSID Working Paper Series #CSID-2013-007 Conditional behavior affects the level of evolved cooperation in public good games Marco A. Janssen Arizona State

More information

ETHICAL CONSIDERATION OF CONSTRUCTION UNION STRATEGIES IN JURISDICTIONAL DISPUTES BASED ON AN AGENT-BASED MODELING (ABM) AND A GAME THEORY

ETHICAL CONSIDERATION OF CONSTRUCTION UNION STRATEGIES IN JURISDICTIONAL DISPUTES BASED ON AN AGENT-BASED MODELING (ABM) AND A GAME THEORY ETHICAL CONSIDERATION OF CONSTRUCTION UNION STRATEGIES IN JURISDICTIONAL DISPUTES BASED ON AN AGENT-BASED MODELING (ABM) AND A GAME THEORY Borinara Park* Associate Professor, Construction Management Program,

More information

Models of Cooperation in Social Systems

Models of Cooperation in Social Systems Models of Cooperation in Social Systems Hofstadter s Luring Lottery From Hofstadter s Metamagical Themas column in Scientific American: "This brings me to the climax of this column: the announcement of

More information

Sympathy and Interaction Frequency in the Prisoner s Dilemma

Sympathy and Interaction Frequency in the Prisoner s Dilemma Sympathy and Interaction Frequency in the Prisoner s Dilemma Ichiro Takahashi and Isamu Okada Faculty of Economics, Soka University itak@soka.ac.jp Abstract People deal with relationships at home, workplace,

More information

Reciprocity, Cooperation, and Reputation

Reciprocity, Cooperation, and Reputation Reciprocity, Cooperation, and Reputation Reciprocal altruism Prisoner s dilemma: repeated play allows cooperation Generosity, punishment, spite (Ultimatum games) Reputation Before next lecture, see intentionality

More information

$16. Using QR in Writing Assignments AALAC Workshop on Enhacing QR Across the Curriculum Carleton College October 11, 2014

$16. Using QR in Writing Assignments AALAC Workshop on Enhacing QR Across the Curriculum Carleton College October 11, 2014 Using QR in Writing Assignments AALAC Workshop on Enhacing QR Across the Curriculum Carleton College October 11, 2014 1 2 The MegaPenny Project 1 billion pennies $16 1 trillion pennies http://www.kokogiak.com/megapenny

More information

Information Acquisition in the Iterated Prisoner s Dilemma Game: An Eye-Tracking Study

Information Acquisition in the Iterated Prisoner s Dilemma Game: An Eye-Tracking Study Information Acquisition in the Iterated Prisoner s Dilemma Game: An Eye-Tracking Study Evgenia Hristova (ehristova@cogs.nbu.bg ) Central and Eastern European Center for Cognitive Science, New Bulgarian

More information

The weak side of informal social control Paper prepared for Conference Game Theory and Society. ETH Zürich, July 27-30, 2011

The weak side of informal social control Paper prepared for Conference Game Theory and Society. ETH Zürich, July 27-30, 2011 The weak side of informal social control Paper prepared for Conference Game Theory and Society. ETH Zürich, July 27-30, 2011 Andreas Flache Department of Sociology ICS University of Groningen Collective

More information

Sequential Decision and Strategy Vector Methods in Ultimatum Bargaining: Evidence on the Strength of Other- Regarding Behavior

Sequential Decision and Strategy Vector Methods in Ultimatum Bargaining: Evidence on the Strength of Other- Regarding Behavior Department of Economics Discussion Paper 2004-04 Sequential Decision and Strategy Vector Methods in Ultimatum Bargaining: Evidence on the Strength of Other- Regarding Behavior Robert J. Oxoby University

More information

COOPERATION 1. How Economic Rewards Affect Cooperation Reconsidered. Dan R. Schley and John H. Kagel. The Ohio State University

COOPERATION 1. How Economic Rewards Affect Cooperation Reconsidered. Dan R. Schley and John H. Kagel. The Ohio State University COOPERATION 1 Running Head: COOPERATION How Economic Rewards Affect Cooperation Reconsidered Dan R. Schley and John H. Kagel The Ohio State University In Preparation Do Not Cite Address correspondence

More information

Exploration and Exploitation in Reinforcement Learning

Exploration and Exploitation in Reinforcement Learning Exploration and Exploitation in Reinforcement Learning Melanie Coggan Research supervised by Prof. Doina Precup CRA-W DMP Project at McGill University (2004) 1/18 Introduction A common problem in reinforcement

More information

Emanuela Carbonara. 31 January University of Bologna - Department of Economics

Emanuela Carbonara. 31 January University of Bologna - Department of Economics Game Theory, Behavior and The Law - I A brief introduction to game theory. Rules of the game and equilibrium concepts. Behavioral Games: Ultimatum and Dictator Games. Entitlement and Framing effects. Emanuela

More information

Reinforcement Learning : Theory and Practice - Programming Assignment 1

Reinforcement Learning : Theory and Practice - Programming Assignment 1 Reinforcement Learning : Theory and Practice - Programming Assignment 1 August 2016 Background It is well known in Game Theory that the game of Rock, Paper, Scissors has one and only one Nash Equilibrium.

More information

Evolutionary game theory and cognition

Evolutionary game theory and cognition Evolutionary game theory and cognition Artem Kaznatcheev School of Computer Science & Department of Psychology McGill University November 15, 2012 Artem Kaznatcheev (McGill University) Evolutionary game

More information

Resource Management: INSTITUTIONS AND INSTITUTIONAL DESIGN. SOS3508 Erling Berge. A grammar of institutions. NTNU, Trondheim.

Resource Management: INSTITUTIONS AND INSTITUTIONAL DESIGN. SOS3508 Erling Berge. A grammar of institutions. NTNU, Trondheim. Resource Management: INSTITUTIONS AND INSTITUTIONAL DESIGN SOS3508 Erling Berge A grammar of institutions NTNU, Trondheim Fall 2009 Erling Berge 2009 Fall 2009 1 Literature Ostrom, Elinor 2005, Understanding

More information

2012 Course: The Statistician Brain: the Bayesian Revolution in Cognitive Sciences

2012 Course: The Statistician Brain: the Bayesian Revolution in Cognitive Sciences 2012 Course: The Statistician Brain: the Bayesian Revolution in Cognitive Sciences Stanislas Dehaene Chair of Experimental Cognitive Psychology Lecture n 5 Bayesian Decision-Making Lecture material translated

More information

A Psychologically-Inspired Agent for Iterative Prisoner s Dilemma

A Psychologically-Inspired Agent for Iterative Prisoner s Dilemma Proceedings of the Twenty-Third International Florida Artificial Intelligence Research Society Conference (FLAIRS 2010) A Psychologically-Inspired Agent for Iterative Prisoner s Dilemma Rawad Al-Haddad

More information

HOT VS. COLD: SEQUENTIAL RESPONSES AND PREFERENCE STABILITY IN EXPERIMENTAL GAMES * August 1998

HOT VS. COLD: SEQUENTIAL RESPONSES AND PREFERENCE STABILITY IN EXPERIMENTAL GAMES * August 1998 HOT VS. COLD: SEQUENTIAL RESPONSES AND PREFERENCE STABILITY IN EXPERIMENTAL GAMES * August 1998 Jordi Brandts Instituto de Análisis Económico (CSIC) Barcelona and University of California Berkeley (jbrandts@econ.berkeley.edu)

More information

INTUITION: GOOD, BAD, AND UGLY INTUITION IN LEGAL DECISION MAKING

INTUITION: GOOD, BAD, AND UGLY INTUITION IN LEGAL DECISION MAKING Focused on the future of learning. INTUITION: GOOD, BAD, AND UGLY INTUITION IN LEGAL DECISION MAKING Archie Zariski Professor, Athabasca University archiez@athabascau.ca Intuition what is it? Direct perception

More information

Emergence of Cooperative Societies in Evolutionary Games Using a Social Orientation Model

Emergence of Cooperative Societies in Evolutionary Games Using a Social Orientation Model Emergence of Cooperative Societies in Evolutionary Games Using a Social Orientation Model Kan-Leung Cheng Abstract We utilize evolutionary game theory to study the evolution of cooperative societies and

More information

Intelligent Systems, Robotics and Artificial Intelligence

Intelligent Systems, Robotics and Artificial Intelligence Intelligent Systems, Robotics and Artificial Intelligence Artificial intelligence and intelligent behavior of agents Comparing artificial and natural intelligence; what should be added to agents The role

More information

Altruism. Why Are Organisms Ever Altruistic? Kin Selection. Reciprocal Altruism. Costly Signaling. Group Selection?

Altruism. Why Are Organisms Ever Altruistic? Kin Selection. Reciprocal Altruism. Costly Signaling. Group Selection? Altruism Nepotism Common-Pool Resource (CPR) Tragedy of the Commons Public Good Free-riders Competitive Altruism A 1987 Florida news story reported that 22 blood donors who had false positives on an HIV

More information

Moral Values from Simple Game Play

Moral Values from Simple Game Play Moral Values from Simple Game Play Eunkyung Kim, Ravi Iyer, Jesse Graham, Yu-Han Chang, Rajiv Maheswaran University of Southern California Los Angeles, CA 90089 {eunkyung,raviiyer,jesse.graham,yuhan.chang,maheswar}@usc.edu

More information

What Causes war? C h. 1 : T e r m i n o l o g y & Methodology. Notes by Denis Bašić

What Causes war? C h. 1 : T e r m i n o l o g y & Methodology. Notes by Denis Bašić What Causes war? C h. 1 : T e r m i n o l o g y & Methodology Notes by Denis Bašić What is called wisdom is concerned with primary causes. Aristotle The Scientific Method: Theory Theories in the social

More information

Using a Social Orientation Model for the Evolution of Cooperative Societies

Using a Social Orientation Model for the Evolution of Cooperative Societies 2010 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology Using a Social Orientation Model for the Evolution of Cooperative Societies Kan-Leung Cheng, Inon Zuckerman,

More information

In this paper we intend to explore the manner in which decision-making. Equivalence and Stooge Strategies. in Zero-Sum Games

In this paper we intend to explore the manner in which decision-making. Equivalence and Stooge Strategies. in Zero-Sum Games Equivalence and Stooge Strategies in Zero-Sum Games JOHN FOX MELVIN GUYER Mental Health Research Institute University of Michigan Classes of two-person zero-sum games termed "equivalent games" are defined.

More information

A Scoring Policy for Simulated Soccer Agents Using Reinforcement Learning

A Scoring Policy for Simulated Soccer Agents Using Reinforcement Learning A Scoring Policy for Simulated Soccer Agents Using Reinforcement Learning Azam Rabiee Computer Science and Engineering Isfahan University, Isfahan, Iran azamrabiei@yahoo.com Nasser Ghasem-Aghaee Computer

More information

Literature Henrich, Joseph, and Natalie Henrich Why Humans Cooperate A Cultural and Evolutionary Explanation. Oxford: Oxford University Press

Literature Henrich, Joseph, and Natalie Henrich Why Humans Cooperate A Cultural and Evolutionary Explanation. Oxford: Oxford University Press INSTITUTIONS AND INSTITUTIONAL DESIGN Erling Berge Henrich and Henrich 2007 Why Humans Cooperate Literature Henrich, Joseph, and Natalie Henrich. 2007. Why Humans Cooperate A Cultural and Evolutionary

More information

Epistemic Social Dilemmas

Epistemic Social Dilemmas Epistemic Social Dilemmas Kevin J.S. Zollman Carnegie Mellon University Passive individualism Not individual Not passive Inquirers must choose: What to investigate Which hypotheses to pursue How to pursue

More information

[1] provides a philosophical introduction to the subject. Simon [21] discusses numerous topics in economics; see [2] for a broad economic survey.

[1] provides a philosophical introduction to the subject. Simon [21] discusses numerous topics in economics; see [2] for a broad economic survey. Draft of an article to appear in The MIT Encyclopedia of the Cognitive Sciences (Rob Wilson and Frank Kiel, editors), Cambridge, Massachusetts: MIT Press, 1997. Copyright c 1997 Jon Doyle. All rights reserved

More information

A Note On the Design of Experiments Involving Public Goods

A Note On the Design of Experiments Involving Public Goods University of Colorado From the SelectedWorks of PHILIP E GRAVES 2009 A Note On the Design of Experiments Involving Public Goods PHILIP E GRAVES, University of Colorado at Boulder Available at: https://works.bepress.com/philip_graves/40/

More information

Evolving Strategic Behaviors through Competitive Interaction in the Large

Evolving Strategic Behaviors through Competitive Interaction in the Large Evolving Strategic Behaviors through Competitive Interaction in the Large Kimitaka Uno & Akira Namatame Dept. of Computer Science National Defense Academy Yokosuka, 239-8686, JAPAN E-mail: {kimi, nama}@cc.nda.ac.jp

More information

Exploring MultiObjective Fitness Functions and Compositions of Different Fitness Functions in rtneat

Exploring MultiObjective Fitness Functions and Compositions of Different Fitness Functions in rtneat Exploring MultiObjective Fitness Functions and Compositions of Different Fitness Functions in rtneat KJ Bredder and Cole Harbeck Abstract In machine learning, the fitness function is an incredibly important

More information

Computational Models for Belief Revision, Group Decisions and Cultural Shifts

Computational Models for Belief Revision, Group Decisions and Cultural Shifts Computational Models for Belief Revision, Group Decisions and Cultural Shifts Whitman Richards (PI), M. I. T. Computer Science and Artificial Intelligence Laboratory MURI Objective: models that forecast

More information

Generative Adversarial Networks.

Generative Adversarial Networks. Generative Adversarial Networks www.cs.wisc.edu/~page/cs760/ Goals for the lecture you should understand the following concepts Nash equilibrium Minimax game Generative adversarial network Prisoners Dilemma

More information

Trust and Situation Awareness in a 3-Player Diner's Dilemma Game

Trust and Situation Awareness in a 3-Player Diner's Dilemma Game Trust and Situation Awareness in a 3-Player Diner's Dilemma Game Yun Teng 1, Rashaad Jones 2, Laura Marusich 3, John O Donovan 1, Cleotilde Gonzalez 4, Tobias Höllerer 1 1 Department of Computer Science,

More information

Attraction and Cooperation in Space

Attraction and Cooperation in Space Attraction and Cooperation in Space Elpida S. Tzafestas, Member, IEEE Abstract This paper presents an attraction mechanism and a study of its behavior in noisy spatial IPD games. The attraction mechanism

More information

Oscillatory Neural Network for Image Segmentation with Biased Competition for Attention

Oscillatory Neural Network for Image Segmentation with Biased Competition for Attention Oscillatory Neural Network for Image Segmentation with Biased Competition for Attention Tapani Raiko and Harri Valpola School of Science and Technology Aalto University (formerly Helsinki University of

More information

Indirect Reciprocity and the Evolution of Moral Signals

Indirect Reciprocity and the Evolution of Moral Signals Indirect Reciprocity and the Evolution of Moral Signals Department of Logic and Philosophy of Science University of California, Irvine Evolution of Psychological Categories - IMBS March 16, 2008 Moral

More information

ENVIRONMENTAL REINFORCEMENT LEARNING: A Real-time Learning Architecture for Primitive Behavior Refinement

ENVIRONMENTAL REINFORCEMENT LEARNING: A Real-time Learning Architecture for Primitive Behavior Refinement ENVIRONMENTAL REINFORCEMENT LEARNING: A Real-time Learning Architecture for Primitive Behavior Refinement TaeHoon Anthony Choi, Eunbin Augustine Yim, and Keith L. Doty Machine Intelligence Laboratory Department

More information

Modeling State Space Search Technique for a Real World Adversarial Problem Solving

Modeling State Space Search Technique for a Real World Adversarial Problem Solving Modeling State Space Search Technique for a Real World Adversarial Problem Solving Kester O. OMOREGIE Computer Science Department, Auchi Polytechnic, Auchi, NIGERIA Stella C. CHIEMEKE Computer Science

More information

Altruistic Behavior: Lessons from Neuroeconomics. Kei Yoshida Postdoctoral Research Fellow University of Tokyo Center for Philosophy (UTCP)

Altruistic Behavior: Lessons from Neuroeconomics. Kei Yoshida Postdoctoral Research Fellow University of Tokyo Center for Philosophy (UTCP) Altruistic Behavior: Lessons from Neuroeconomics Kei Yoshida Postdoctoral Research Fellow University of Tokyo Center for Philosophy (UTCP) Table of Contents 1. The Emergence of Neuroeconomics, or the Decline

More information

Mini-Course in Behavioral Economics Leeat Yariv. Behavioral Economics - Course Outline

Mini-Course in Behavioral Economics Leeat Yariv. Behavioral Economics - Course Outline Mini-Course in Behavioral Economics Leeat Yariv The goal of this mini-course is to give an overview of the state of the art of behavioral economics. I hope to trigger some research in the field and will

More information

Gracious Space as a Framework For Change

Gracious Space as a Framework For Change For Change Gracious Space is a container for change. It is a crucible into which you can put the most pressing issue, conflict, culture change, strategic planning process or other dynamic change process.

More information

Cooperation in a Prisoners Dilemma with Losses. Results from an Experimental Study.

Cooperation in a Prisoners Dilemma with Losses. Results from an Experimental Study. Cooperation in a Prisoners Dilemma with Losses. Results from an Experimental Study. Roger Berger Institute for Sociology, University of Leipzig and SNF Venice International University 2006 Rational Choice

More information

Identity, Homophily and In-Group Bias: Experimental Evidence

Identity, Homophily and In-Group Bias: Experimental Evidence Identity, Homophily and In-Group Bias: Experimental Evidence Friederike Mengel & Sergio Currarini Univeersity of Nottingham Universita Ca Foscari and FEEM, Venice FEEM, March 2012 In-Group Bias Experimental

More information

How rational are your decisions? Neuroeconomics

How rational are your decisions? Neuroeconomics How rational are your decisions? Neuroeconomics Hecke CNS Seminar WS 2006/07 Motivation Motivation Ferdinand Porsche "Wir wollen Autos bauen, die keiner braucht aber jeder haben will." Outline 1 Introduction

More information

Institutions and Cooperative Behavior

Institutions and Cooperative Behavior Institutions and Cooperative Behavior by Arild Vatn Department of International Environmental and Development Studies Norwegian University of Life Sciences Complexity Economics for Sustainability Seminar

More information

Appendix: Instructions for Treatment Index B (Human Opponents, With Recommendations)

Appendix: Instructions for Treatment Index B (Human Opponents, With Recommendations) Appendix: Instructions for Treatment Index B (Human Opponents, With Recommendations) This is an experiment in the economics of strategic decision making. Various agencies have provided funds for this research.

More information

Emotion, Trustworthiness and Altruistic Punishment in a Tragedy of the Commons Social Dilemma

Emotion, Trustworthiness and Altruistic Punishment in a Tragedy of the Commons Social Dilemma Emotion, Trustworthiness and Altruistic Punishment in a Tragedy of the Commons Social Dilemma Garrison Greenwood 1, Hussein A. Abbass 2(B), and Eleni Petraki 3 1 Electrical and Computer Engineering Department,

More information

Hebbian Plasticity for Improving Perceptual Decisions

Hebbian Plasticity for Improving Perceptual Decisions Hebbian Plasticity for Improving Perceptual Decisions Tsung-Ren Huang Department of Psychology, National Taiwan University trhuang@ntu.edu.tw Abstract Shibata et al. reported that humans could learn to

More information

WORKING PAPER GAME DYNAMICAL ASPECTS OF THE PRISONER'S DILEMMA WP December Martin Novak Karl Sigmund

WORKING PAPER GAME DYNAMICAL ASPECTS OF THE PRISONER'S DILEMMA WP December Martin Novak Karl Sigmund WORKING PAPER GAME DYNAMICAL ASPECTS OF THE PRISONER'S DILEMMA Martin Novak Karl Sigmund December 1988 WP-88-125 L International Institute for Applied Systems Analysis GAME DYNAMICAL ASPECTS OF THE PRISONER'S

More information

Class 6 Overview. Two Way Tables

Class 6 Overview. Two Way Tables Class 6 Overview Two Way Tables Example: Using data from 2004-2008, the Steelers won 56 / 80 or 70% of their games. Of the games won, 31 / 56 or 55% were played at home. Of the games lost, 9 / 24 or 38%

More information

Chapter 1 Social Science and Its Methods

Chapter 1 Social Science and Its Methods Chapter 1 Social Science and Its Methods MULTIPLE CHOICE 1) Scientific knowledge is knowledge that has been: A) systematically gathered, classified, related, and interpreted. B) rediscovered and cherished

More information

Reverse Appraisal: Inferring from Emotion Displays who is the Cooperator and the Competitor in a Social Dilemma

Reverse Appraisal: Inferring from Emotion Displays who is the Cooperator and the Competitor in a Social Dilemma Reverse Appraisal: Inferring from Emotion Displays who is the Cooperator and the Competitor in a Social Dilemma Celso M. de Melo (demelo@ict.usc.edu) Jonathan Gratch (gratch@ict.usc.edu) Institute for

More information

SUPPLEMENTAL MATERIAL

SUPPLEMENTAL MATERIAL 1 SUPPLEMENTAL MATERIAL Response time and signal detection time distributions SM Fig. 1. Correct response time (thick solid green curve) and error response time densities (dashed red curve), averaged across

More information

Strong Reciprocity and Human Sociality

Strong Reciprocity and Human Sociality Notes on Behavioral Economics 1 Strong Reciprocity and Human Sociality Human groups are highly social despite a low level of relatedness. There is an empirically identifiable form of prosocial behavior

More information

Supporting Information

Supporting Information Supporting Information Burton-Chellew and West 10.1073/pnas.1210960110 SI Results Fig. S4 A and B shows the percentage of free riders and cooperators over time for each treatment. Although Fig. S4A shows

More information

ABSTRACT. sometimes an agent s preferences may involve not only its own payoff (as specified

ABSTRACT. sometimes an agent s preferences may involve not only its own payoff (as specified ABSTRACT Title of dissertation: AGENT MODELING IN STOCHASTIC REPEATED GAMES Kan Leung Cheng, Doctor of Philosophy, 2014 Dissertation directed by: Professor Dana Nau Department of Computer Science There

More information

Exploring the Influence of Particle Filter Parameters on Order Effects in Causal Learning

Exploring the Influence of Particle Filter Parameters on Order Effects in Causal Learning Exploring the Influence of Particle Filter Parameters on Order Effects in Causal Learning Joshua T. Abbott (joshua.abbott@berkeley.edu) Thomas L. Griffiths (tom griffiths@berkeley.edu) Department of Psychology,

More information

Formulating Emotion Perception as a Probabilistic Model with Application to Categorical Emotion Classification

Formulating Emotion Perception as a Probabilistic Model with Application to Categorical Emotion Classification Formulating Emotion Perception as a Probabilistic Model with Application to Categorical Emotion Classification Reza Lotfian and Carlos Busso Multimodal Signal Processing (MSP) lab The University of Texas

More information

Work, Employment, and Industrial Relations Theory Spring 2008

Work, Employment, and Industrial Relations Theory Spring 2008 MIT OpenCourseWare http://ocw.mit.edu 15.676 Work, Employment, and Industrial Relations Theory Spring 2008 For information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms.

More information

Racing for the City: The Recognition Heuristic and Compensatory Alternatives

Racing for the City: The Recognition Heuristic and Compensatory Alternatives Racing for the City: The Recognition Heuristic and Compensatory Alternatives Katja Mehlhorn (s.k.mehlhorn@rug.nl) Dept. of Artificial Intelligence and Dept. of Psychology, University of Groningen, the

More information

Personality and social behavior. Aldo Rustichini Conference on Measuring and Assessing Skills Chicago, October 1-2

Personality and social behavior. Aldo Rustichini Conference on Measuring and Assessing Skills Chicago, October 1-2 Personality and social behavior Aldo Rustichini Conference on Measuring and Assessing Skills Chicago, October 1-2 Outline: Three main themes Conscientiousness and strategic behavior Intelligence and strategic

More information

Department of Psychological and Behavioural Science public lecture Human Cooperation

Department of Psychological and Behavioural Science public lecture Human Cooperation Department of Psychological and Behavioural Science public lecture Human Cooperation Dr David Rand Associate Professor of Psychology, Economics, and Management, Yale University Director of Human Cooperation

More information

Modeling Social Systems, Self-Organized Coordination, and the Emergence of Cooperation. Dirk Helbing, with Anders Johansson and Sergi Lozano

Modeling Social Systems, Self-Organized Coordination, and the Emergence of Cooperation. Dirk Helbing, with Anders Johansson and Sergi Lozano Modeling Social Systems, Self-Organized Coordination, and the Emergence of Cooperation Dirk Helbing, with Anders Johansson and Sergi Lozano Sociology - The Queen of Sciences? Auguste Comte (1798-1857)

More information

Reinforcement Learning

Reinforcement Learning Reinforcement Learning Michèle Sebag ; TP : Herilalaina Rakotoarison TAO, CNRS INRIA Université Paris-Sud Nov. 9h, 28 Credit for slides: Richard Sutton, Freek Stulp, Olivier Pietquin / 44 Introduction

More information

Partners versus Strangers: Random Rematching in Public Goods Experiments

Partners versus Strangers: Random Rematching in Public Goods Experiments Partners versus Strangers: Random Rematching in Public Goods Experiments by James Andreoni Department of Economics University of Wisconsin-Madison Madison, Wisconsin 53711 and Rachel Croson Wharton School

More information

THE SOCIAL EXCHANGE HEURISTIC

THE SOCIAL EXCHANGE HEURISTIC THE SOCIAL EXCHANGE HEURISTIC MANAGING ERRORS IN SOCIAL EXCHANGE Toshio Yamagishi, Shigeru Terai, Toko Kiyonari, Nobuhiro Mifune, and Satoshi Kanazawa ABSTRACT We extend the logic of Haselton and Buss

More information