Decision Making In Prisoner’s Dilemma

Masaryk University

Faculty of Arts

Department of Psychology
Psychology of decision making: Effect of learning, end-effect, cheap talk, and other variables influencing decision making in the Prisoner’s Dilemma game

Diploma thesis

Author: Štěpán Veselý

Supervisor: doc. PhDr. Mojmír Tyrlík, Ph.D.

Brno, 2011

Declaration

Hereby I declare, that this thesis is my original authorial work, which I have worked

out by my own. All sources, references, and literature used or excerpted during elaboration of this work are properly cited and listed in complete reference to the due source.

Štěpán Veselý

Brno, April 2011

Acknowledgement
I would like to thank my supervisor doc. PhDr. Mojmír Tyrlík, Ph.D. for inspiration and for his understanding approach. I wish to thank Mgr. Helena Klimusová, Ph.D. and PhDr. Martin Jelínek, Ph.D. for their kind help with certain statistical analyses. I thank my family for help and support. I am also grateful to all people, who participated in my experiments.

Contents:

1. Introduction 6

1.1 Theoretical part 6

1.2 Practical part 7

I. Theoretical part 8

2. Classical utility theory and classical decision making 8

2.1 Mathematically calculated utility 8

2.2 Expected utility with diminishing returns 9

3. Bounded rationality theory 11

3.1 Bounded rationality vs. classical utility theory 11

3.2 Basic features 12

3.3 Bounded rationality: developments 13

4. Prospect theory 16

4.1 Applications: prospect theory (and heuristics) in behavioral finance 22

5. Heuristics and biases 26

5.1 Representativeness 26

5.11 Base rates 27

5.12 Samples of a population 29

5.2 Availability 31

5.21 Recall: differential availability 31

5.22 Construction: prediction and attribution based on scenarios 33

5.3 Anchoring 34

5.4 Heuristics – natural selection and modularity 35

6. Theory of games 36

6.1 How to play the game 36

6.11 Beginnings and applications 36

6.12 A 2 x 2 game 37

6.13 Question of trust 37

6.14 Strictly dominant strategy 38

6.15 Superrationality and conscience 39

6.16 Matching heuristic and control heuristic 40

6.17 Iterated Prisoner’s Dilemma 40

6.18 Teach and be taught 43

6.19 Shadow of the future 44

6.2 How to study the game (the most important independent variables) 46

6.21 Effects of learning 47

6.22 Characteristics of subjects 47

6.23 The strategy against which the player plays 50

6.24 Payoff matrix and incentives 52

6.25 Experimental setting 54

6.3 Other independent variables and modifications of the game 55

6.31 Continuous development of Prisoner’s Dilemma research 59

II. Practical part 61

7. Overview 61

8. The amount of cooperation in the single move Prisoner’s Dilemma game and in the iterated Prisoner’s Dilemma game 62

8.1 Task 62

8.2 Single-move Prisoner’s Dilemma 62

8.3 Iterated Prisoner’s Dilemma 63

8.4 Procedure 63

8.41 Subjects sample 64

8.42 Prisoner’s Dilemma game introduced to subjects 64

8.43 Single move Prisoner’s Dilemma 65

8.44 Iterated Prisoner’s Dilemma 65

8.5 Variables 69

8.6 Hypotheses 70

8.7 Results 71

8.8 Discussion 79

9. End-game effect 82

9.1 Task 82

9.2 Procedure 82

9.3 Variables 82

9.4 Hypotheses 84

9.5 Results 86

9.6 Discussion 97

10. Effect of learning 100

10.1 Task 100

10.2 Procedure 101

10.3 Variables 101

10.4 Hypothesis 103

10.5 Results 104

10.6 Discussion 112

11. Cheap-talk 114

11.1 Task 114

11.2 Procedure 114

11.3 Variables 115

11.4 Hypotheses 117

11.5 Results 118

11.6 Discussion 123

12. Personality variables 125

12.1 Task 125

12.2 Procedure 126

12.3 Variables 127

12.4 Results 129

12.5 Discussion 133

13. Conclusions 137

14. References 140

Appendix 159

1. Introduction

1.1 Theoretical part

The fundamental question is simple: How do we make decisions? Well, the obvious answer is that we weigh the respective utility values of the alternatives together with the probability of their occurrence, then we do the same for the respective costs, and then subtract the net costs from the net values. The alternative with the greatest net utility value after the subtraction is what we decide to do. In other words: we compare the outcomes. Decision makers as described in expected-utility theory (see section 2.), prospect theory (section 4.), heuristics and biases approach (section 5.), and theory of games (section 6.) compare outcomes, or they try their best to compare them. Each of these theories, however, came with its own special answers to the “How do we make decisions” question.

The task of the theoretical part of my thesis is to examine some major contributions to the decision making theory. The following approaches will be discussed: In section 2. the economics-based classical utility theory and classical decision making. In section 3. Simon’s bounded rationality theory. In section 4. prospect theory developed by Kahneman and Tversky. In section 5. heuristics and biases approach, and finally in section 6. game theory and human strategic decision making in games with conflict of interest (e. g. Prisoner’s Dilemma).
In the theoretical part we will also deal shortly with considerations coming from the evolutionary theory and psychology, that offer further explanations of several issues at hand. We will, unfortunately, have to omit a truly promising neuroscientific approach, which proposes valuable explanations of human decision making behavior, such as the somatic markers hypothesis (Damasio, 1996; Bechara et al., 1997; Bechara et al., 2000a; Bechara et al., 2000b). For instance, an empirical study of decision making, that takes into account physiological bodily and brain changes, and both conscious and subconscious emotional states of actors (registered for example with help of facial expressions and microexpressions reading and introspective protocols) as a mediator between outcomes-information (“inputs”) and choice outputs, might bring some notable improvements and refinements of our view of human decision making.