Premium
An experimental investigation into the effects of uncertainty on rational behaviour in two person symmetric games
Author(s) -
Butler David J.
Publication year - 1992
Publication title -
journal of behavioral decision making
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.136
H-Index - 76
eISSN - 1099-0771
pISSN - 0894-3257
DOI - 10.1002/bdm.3960050405
Subject(s) - allowance (engineering) , stochastic game , set (abstract data type) , test (biology) , variety (cybernetics) , independence (probability theory) , mathematical economics , social psychology , psychology , econometrics , mathematics , economics , computer science , statistics , paleontology , operations management , biology , programming language
Abstract This paper reports the results of two experiments to investigate the effects of payoff alterations in two‐person symmetric games. The initial experiment involved 60 subjects, each of whom could earn from £0 to £15, depending on the interaction of their decisions and the unknown decisions of one other person. All decisions were made without feedback to isolate the impact on subjects' behaviour of the game's parameters. A primary aim was to see if the games had strategically distinct structures, or whether uncertainty over the choice rule of others would influence choice behaviour and lessen this independence. An alternative model of rational choice, making allowance for uncertainty in the decision environment, was proposed to capture subjects' behaviour. The results indicated that the frontiers of most of the games had no impact on choice behaviour other than through the magnitude of the change in payoffs. The model received strong support across an identifiable set of ‘frontierless’ games. However, the variety of games was not sufficient to provide a comprehensive test. A second experiment was recently held to close these gaps. Fifty subjects took part, each of whom earned from A$0 to A$24, averaging A$17. The findings provided broad additional support for the theory.