Efficiency, Reciprocity and Expectations in an Experimental Game

M. Dufwenberg, U. Gneezy

Research output: Working paperDiscussion paperOther research output

306 Downloads (Pure)

Abstract

We experimentally investigate the nature of strategic interaction in a 2-player game.Player 1may take x Dutch guilders (f x) and end the game (player 2 then gets f 0), or let player 2 split f 20 between the players. x is a treatment variable taking values of f 4, 7, 10, 13, and 16.We find that most players 2 "give away" positive amounts (f6 on average), but their choices are independent of x.We explicitly measure the players' beliefs and find that many players 1 expect to get back no more than f x but nevertheless let player 2 split the f 20, and that the behavior by the players 2 is consistent with a theory of a guilt based on psychological game theory.
Original languageEnglish
Place of PublicationTilburg
PublisherMicroeconomics
Number of pages20
Volume1996-79
Publication statusPublished - 1996

Publication series

NameCentER Discussion Paper
Volume1996-79

Fingerprint

Experimental games
Guilt
Psychological
Game theory
Strategic interaction

Keywords

  • game theory

Cite this

Dufwenberg, M., & Gneezy, U. (1996). Efficiency, Reciprocity and Expectations in an Experimental Game. (CentER Discussion Paper; Vol. 1996-79). Tilburg: Microeconomics.
Dufwenberg, M. ; Gneezy, U. / Efficiency, Reciprocity and Expectations in an Experimental Game. Tilburg : Microeconomics, 1996. (CentER Discussion Paper).
@techreport{96e64bdc6dc34d499d1cb20c370dddf5,
title = "Efficiency, Reciprocity and Expectations in an Experimental Game",
abstract = "We experimentally investigate the nature of strategic interaction in a 2-player game.Player 1may take x Dutch guilders (f x) and end the game (player 2 then gets f 0), or let player 2 split f 20 between the players. x is a treatment variable taking values of f 4, 7, 10, 13, and 16.We find that most players 2 {"}give away{"} positive amounts (f6 on average), but their choices are independent of x.We explicitly measure the players' beliefs and find that many players 1 expect to get back no more than f x but nevertheless let player 2 split the f 20, and that the behavior by the players 2 is consistent with a theory of a guilt based on psychological game theory.",
keywords = "game theory",
author = "M. Dufwenberg and U. Gneezy",
note = "Pagination: 20",
year = "1996",
language = "English",
volume = "1996-79",
series = "CentER Discussion Paper",
publisher = "Microeconomics",
type = "WorkingPaper",
institution = "Microeconomics",

}

Dufwenberg, M & Gneezy, U 1996 'Efficiency, Reciprocity and Expectations in an Experimental Game' CentER Discussion Paper, vol. 1996-79, Microeconomics, Tilburg.

Efficiency, Reciprocity and Expectations in an Experimental Game. / Dufwenberg, M.; Gneezy, U.

Tilburg : Microeconomics, 1996. (CentER Discussion Paper; Vol. 1996-79).

Research output: Working paperDiscussion paperOther research output

TY - UNPB

T1 - Efficiency, Reciprocity and Expectations in an Experimental Game

AU - Dufwenberg, M.

AU - Gneezy, U.

N1 - Pagination: 20

PY - 1996

Y1 - 1996

N2 - We experimentally investigate the nature of strategic interaction in a 2-player game.Player 1may take x Dutch guilders (f x) and end the game (player 2 then gets f 0), or let player 2 split f 20 between the players. x is a treatment variable taking values of f 4, 7, 10, 13, and 16.We find that most players 2 "give away" positive amounts (f6 on average), but their choices are independent of x.We explicitly measure the players' beliefs and find that many players 1 expect to get back no more than f x but nevertheless let player 2 split the f 20, and that the behavior by the players 2 is consistent with a theory of a guilt based on psychological game theory.

AB - We experimentally investigate the nature of strategic interaction in a 2-player game.Player 1may take x Dutch guilders (f x) and end the game (player 2 then gets f 0), or let player 2 split f 20 between the players. x is a treatment variable taking values of f 4, 7, 10, 13, and 16.We find that most players 2 "give away" positive amounts (f6 on average), but their choices are independent of x.We explicitly measure the players' beliefs and find that many players 1 expect to get back no more than f x but nevertheless let player 2 split the f 20, and that the behavior by the players 2 is consistent with a theory of a guilt based on psychological game theory.

KW - game theory

M3 - Discussion paper

VL - 1996-79

T3 - CentER Discussion Paper

BT - Efficiency, Reciprocity and Expectations in an Experimental Game

PB - Microeconomics

CY - Tilburg

ER -

Dufwenberg M, Gneezy U. Efficiency, Reciprocity and Expectations in an Experimental Game. Tilburg: Microeconomics. 1996. (CentER Discussion Paper).