1、目 录 外文文献 . 1 1. Introduction . 1 2. Games and game theory . 2 3. Theories of social preferences . 3 4. Why do game experiments? And which games?. 3 5. Conclusions . 4 中文翻译 . 5 1.摘要 . 5 2.博弈和博弈论 . 5 3.社会偏好理论 . 6 4.为什么用博弈做实验？用什么博弈？ . 6 5. 结论 . 7 外文文献 Measuring Social Norms and Preferences using Experi
2、mental Games: A Guide for Social Scientists Colin F. Camerer and Ernst Fehr 1. Introduction The purpose of this chapter is to describe a menu of experimental games that are useful for measuring aspects of social norms and social preferences. Economists use the term “preferences” to refer to the choi
3、ces people make, and particularly to tradeoffs between different collections (“bundles”) of things they valuefood, money, time, prestige, and so forth. “Social preferences” refer to how people rank different allocations of material payoffs to themselves and others. Self-interested individuals care o
4、nly about their own material payoffs. The past two decades of experimental research have shown, however, that a substantial fraction of people in developed countries (typically college students) also care about the payoffs of others. In some situations, many people are willing to spend resources to
5、reduce the payoff of others. In other situations, the same people spend resources to increase the payoff of others. As we will see, the willingness to reduce or increase the payoff of relevant reference actors exists even though people reap neither present nor future material rewards from reducing o
6、r increasing payoffs of others. This indicates that, in addition to self-interested behavior, people sometimes behave as if they have altruistic preferences, and preferences for equality and reciprocity.1 Reciprocity, as we define it here, is different from the notion of reciprocal altruism in evolu
7、tionary biology. Reciprocity means that people are willing to reward friendly actions and to punish hostile actionsalthough the reward or punishment causes a net reduction in the material payoff of those who reward or punish. Similarly, people who dislike inequality are willing to take costly action
8、s to reduce inequality although this may result in a net reduction of their material payoff. Reciprocal altruism typically assumes that reciprocation yields a net increase in the material payoff (for example, because one players action earns them a reputation which benefits them in the future). Altr
9、uism, as we define it here, means that an actor takes costly actions to increase the payoff of another actor, irrespective of the other actors previous actions. Altruism thus represents unconditional kindness while reciprocity means non-selfish behavior that is conditioned on the previous actions of
10、 the other actor. Reciprocity, inequality aversion and altruism can have large effects on the regularities of social life and, in particular, on the enforcement of social norms. This is why the examination of the nature of social preferences is so important for anthropology and for social sciences i
11、n general. There is, for example, an ongoing debate in anthropology about the reasons for food-sharing in small-scale societies. The nature of social preferences will probably have a large effect on the social mechanism that sustains food-sharing. For example, if many people in a society exhibit ine
12、quality aversion or reciprocity, they will be willing to punish those who do not share food, so no formal mechanism is needed to govern food-sharing. Without such preferences, formal mechanisms are needed to sustain food-sharing (or sharing does not occur at all). As we will see there are simple gam
13、es that allow researchers to find out whether there are norms of food-sharing, and punishment of those who do not share. In the following we first sketch game theory in broad terms. Then we describe some basic features of experimental design in economics. Then we introduce a menu of seven games that
14、 have proved useful in examining social preferences. We define the games formally, show what aspects of social life they express, and describe behavioral regularities from experimental studies. The behavioral regularities are then interpreted in terms of preferences for reciprocity, inequity aversio
15、n or altruism. The final sections describe some other games anthropologists might find useful, and draw conclusions. 2. Games and game theory Game theory is a mathematical language for describing strategic interactions and their likely outcomes. A game is a set of strategies for each of several play
16、ers, with precise rules for the order in which players choose strategies, the information they have when they choose, and how they rate the desirability (utility) of resulting outcomes. Game theory is designed to be flexible enough to be used at many levels of detail in a broad range of sciences. Pl
17、ayers may be genes, people, groups, firms or nation-states. Strategies may be genetically-coded instincts, heuristics for bidding on the e-Bay website, corporate routines for developing and introducing new products, a legal strategy in complex mass tort cases, or wartime battle plans. Outcomes can b
18、e anything players value- prestige, food, control of Congress, sexual opportunity, returning a tennis serve, corporate profits, the gap between what you would maximally pay for something and what you actually pay (“consumer surplus”), a sense of justice, or captured territory. Game theory consists o
19、f two different enterprises: (1) Using games as a language or taxonomy to parse the social world； and (2) deriving precise predictions about how players will play in a game by assuming that players maximize expected “utility” (personal valuation) of consequences, plan ahead, and form beliefs about o
20、ther players likely actions. The second enterprise dominates game theory textbooks and journals. Analytical theory of this sort is extremely mathematical, and inaccessible to many social scientists outside of economics and theoretical biology. Fortunately, games can be used as a taxonomy with minima
21、l mathematics because understanding prototypical games like those discussed in this chapter requires nothing beyond simple logic. The most central concept in game theory is Nash equilibrium. A set of strategies (one for each player) form an equilibrium if each player is choosing the strategy which i
22、s a best response (i.e., gives the highest expected utility) to the other players strategies. Attention is focussed on equilibrium because players who are constantly switching to better strategies, given what others have done, will generally end up at an equilibrium. Increasingly, game theorists are
23、 interested in the dynamics of equilibration as well, in the form of evolution of populations of player strategies (Weibull, 1995)； or learning by individuals from experience (e.g., Fudenberg and Levine, 1998； Camerer and Ho, 1999). 3. Theories of social preferences Within economics, the leading exp
24、lanation for the patterns of results described above is that agents have social preferences (or “social utility”) which take into account the payoffs and perhaps intentions of others. Roughly speaking, social preference theories assume that people have stable preferences for how money is allocated (
25、which may depend on who the other player is, or how the allocation came about), much as they are assumed in economics to have preferences for food, the present versus the future, how close their house is to work, and so forth.10 Cultural anthropologists and evolutionary psychologists have sought to
26、explain the origin of these preferences. One idea is that in the environment of evolutionary adaptation (EEA) or ancestral past, people mostly engaged in repeated games with people they knew. Evolution created specialized cognitive heuristics for playing repeated games efficiently. It is well-known
27、in game theory that behavior which is optimal for a selfinterested actor in a one-period game with a stranger - such as defecting or free riding, accepting all ultimatum offers - is not always optimal in repeated games with partners. In a repeated ultimatum game, for example, it pays to reject offer
28、s to build up a reputation for being hard to push around, which leads to more generous offers in the future. In the unnatural habitat view, subjects cannot “turn off” the habitual behavior shaped by repeated-game life in the EEA when they play single games with strangers in the lab. An important mod
29、ification of this view is that evolution did not equip all people with identical hard-wired instincts for playing games, but instead created the capacity for learning social norms. The latter view can explain why different cultures would have different norms. 4. Why do game experiments? And which games?
