Emotion

Document Sample
Emotion Powered By Docstoc
					How the Emotions Enforce the
Cooperative Behavior in Social
Dilemma Games

              Speaker:林 家 仰
Before start

   The main contents in this speech are discussing
    Prof. Ernesto Reuben’s paper and using the
    materials in his courses
   http://ereuben.googlepages.com/home
Outline

   What’s experimental economics?
   A very brief history
   Neuroeconomics
   Public goods
   Punishment
   Emotions
What’s experimental economics?


   A discipline in
    which data are
    collected in a
    controlled
    environment.
A Brief History
Market experiments
 Decentralized markets
    Chamberlin (1948) induced demand and cost structure

 Double auction
    Vernon Smith (1962, 1964)

Game experiments
 Prisoners‘dilemma 1950’s
    Originally by psychologists and sociologists

 Oligopoly games
    Reinhard Selten(1959)

Individual choice experiments
 Choice under uncertainty
    Savage (1954)

 Allais paradox (1953)
Why experimental method
   Advantages of the experiments
       Control
           Institutions (e.g. voting rules, communication, etc.)
               Not always complete control (e.g. social norms)
           Incentives (payoffs)
               Not always complete control (e.g. altruism)
           Measure confounding variables (e.g. beliefs)
           Randomization (avoids some self-selection problems)
       Replication
           Check for robustness, experimenter effects, etc.
           Gives an incentive to do it right
           Make available: data, instructions, program, and procedures
Importance
   Economics is becoming an experimental science like physics and
    biology.����
   Like theory, running experiments is an established method to explain
    and/or describe economic activity.
Neuroeconomics

   See lecture 1 on Prof. Ernesto Reuben’s web site
    http://ereuben.googlepages.com/expteach
Public Goods
Real-world problems of cooperation
 Cooperative hunting and warfare

 Teamwork in firms

 Charities and gift-giving

 Environmental protection

 Economic public goods

       Paying taxes
       Fishing
       Security
   Political collective action
       Voting
       Lobbying
       Revolutions
The voluntary contributions mechanism
(VCM)
   The profit of each subject


   endowment: ei
   Contribute to public account: ci
   Contributions to the public good benefit each
    member by: αi (marginal per capita return, MPCR)
   Group of n members
Reasons for cooperation

   By mistake!
       Do not understand that ci= 0 is dominant
       Do understand dominance but make systematic errors
   Social preferences
       Altruism, warm glow, efficiency-seeking motives
       Conditional cooperation, reciprocity
   Strategic cooperation
       Strategies such as Tit-for-Tat can support cooperation among
        selfish players
           mostly infinitely repeated games but see also Kreps et al. (1982)
By mistake

   Cooperating by Mistake Brandts et al. 2004
   Design
       VCM: n= 4, e= 9, repeated for 10 periods
       72 subjects, within subjects
       On every period the MPCR is randomly drawn from 10
        values
           MPCR ≤0.1875: efficient ���� ci= 0, dominant strategy ����
            ci= 0
           0.3125 ≤MPCR ≤0.9375: efficient ���� ci= 9, dominant
            strategy ���� ci= 0
           MPCR ≥1.0625:efficient ���� ci= 9, dominant strategy ����
            ci= 9
By mistake

   Result
Conditional Cooperation
   Conditional or unconditional cooperation? Fischbacher & Gächeter
    2006
   Design
       140 subjects, within subjects
       VCM: n= 4, e= 20, MPCR = 0.4
       2 stages: strategy method and normal
       Decisions 1: unconditional contribution decision
       Decision 2: use strategy method to elicit contribution schedule with respect to
        average contribution of decision 1
       Pick randomly three decision 1’s and one decision 2
   Predictions
       Altruism or ‘warm glow’: contribution is positive but independent of others
        contribution
       Reciprocity: contribution increases with the average contribution of others
       Selfish: always contribute zero
Conditional Cooperation

Results����
   Little unconditional
    cooperation����
   Heterogeneity in
    types:
       55% conditional
        cooperators
       23% selfish
       12% ‘hump-
        shape’contributors
       10% other
Strategic cooperation

   Is there strategic cooperation? Keser& van Winden2000
   Compare partners vs. strangers
       If partners cooperate more → supports strategic cooperation?
Design����
 VCM:

   n= 4, e= 10, 25 periods, MPCR

     = 0.5
   160 subjects between subjects

Results
 Strong effect of the partner’s
  treatment
Explaining the decline in cooperation Croson 1996
 Why does cooperation decline with time
       Strategic cooperation
       Learning to play the
        dominant strategy
   Design: surprise restart
       VCM:
       n= 4, e= 25, MPCR = 0.5
       Repeated 10 + 10 periods
       24 subjects
   Results
       Clear evidence of restart
        specially for partners
Punishment

   The effectiveness of punishment Nikiforakis and Normann 2006
   Results����
       Punishment sustains
        cooperation with a
        damage/cost ratio
        greater than 2/1����
       Punishment
        increases welfare
        with a damage/cost
        ratio greater than
        3/1
To punish or not to punish Sanfreyet al. 2003
 Results
     Higher activation in anterior insulafor unfair
      human offers
     Activation is higher with degree of unfairness
Results
 Higher activation in anterior insulafor
  unfair human offers
       Activation is higher with degree of
        unfairness
       Activation is highest with rejection
   Higher activation in right dorsolateral
    prefrontal cortex
       Not sensitive to rejection
Emotions
Prosocial emotions Hopfensitz and Reuben
    2006
   For punishment to be effective:
       Punished subjects should switch to
        cooperation
       Punished subjects should not punish
        back

   Design
       Trust game with ‘infinite’ rounds of
        punishment (costs 1 to reduce 4)
       2 periods, perfect strangers
       emotions are measured before making
        decisions
Measurement of Emotions

   Is self reports a good idea ?
   experimental economists
       self reports is sometimes regarded with suspicion
        (Smith and Walker, 1993)
       self reports will often overstate the true amount that
        individuals are will-ing to pay (Murphy et al., 2005)
Measurement of Emotions

   Social psychology
       emotions are internal, difficult to observe states, self
        reports of emotions are an often used technique
        (Robinson and Clore, 2002)
       self reports of anger have been related to skin
       conductance levels for emotional reactions in the power
        to take game (Ben-Shakhar et al., 2007
       self reported are indeed reliable can be deduced from a
        recent neuroimaging study (Takahashi et al., 2004)
Measurement of Emotions

   measured emotions included: anger, gratitude,
    guilt, happiness, irritation, shame, and surprise.
   Results
       first movers cooperate more often and second movers
        return more in the presence of punishment
   Results
       2nd movers cooperate after being punished only if they
        feel guilt
   Results
       2nd movers cooperate after being punished only if they
        feel guilt
Results
 Considerable retaliation after receiving punishment
       40% of second movers punish back if punished
   2nd movers retaliate because:
       They are angry and feel no guilt

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:7
posted:5/16/2012
language:English
pages:29
fanzhongqing fanzhongqing http://
About