Laboratory Strife: Higgling as Experimental Science in Economics and Social Psychology

Abstract

Ce texte a ete prepare pour le colloque HIGGLING: Markets and their Transactors in the History of Economic Thought, a Duke University, mars 1993. Il s'agit de l'experimentation en laboratoire des jeux denegotiation a deux personnes, dans la theorie economique et la theorie psycho-sociale, depuis la deuxieme guerre mondiale. Pour ce qui a trait a ce premier, nous examinons les premieres experiences de Siegel et Fouraker, puis celles basees sur le modele axiomatique de Nash, et enfin les approches strategiques qui font suite au travail de Rubinstein. En ce qui a trait a la psychologie, nous examinons plusieurs experiences qui se revelent tres heterogenes etsans base theorique aussi rigoureuse que celle de l'economie. Neanmoins, les facteurs sociaux et culturels, qui sont mis en avant dans les experiences psychologiques, se manifestent de plus en plus dans celles de l'economie. Mais, la theorie d'utilite se revele insuffisamment souple pour l'integration theorique de ces facteurs. This paper was prepared for a conference on Higgling: Markets and their Transactors in the History of Economic Thought, held at Duke University, March 1993. The subject is laboratory experimentation on two-person bargaining in both the economics and social psychology literatures since World War II. In the former, we look at the early experiments of Siegel and Fouraker, and those based on the axiomatic models following Nash, and the strategic approaches following Rubinstein. In the latter, we survey a range of approaches, whose very heterogeneity reflects the absence of an even relatively narrow theoretical basis. What emerges is that while the economics experiments perpetually return to the importance of those social and cultural factors that social psychology literature emphasizes, the constraints of utility-based models sharply limit the extent to which these variables can be incorporated into the theory.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 15 citations exist upstream.