A neural model of bilateral negotiation consisting of one and two issues

Abstract

This thesis demonstrates that neural technology may be successfully employed to mimic some of the thought process of a negotiator during a bilateral negotiation. Using the constraint satisfaction paradigm, originally developed to explore parallel distributed processing, a neural network is proposed to emulate the thought process of a buyer who negotiates the purchase of a good based on price and quality. The findings of this thesis suggest that continued research in neural networks to replicate the mental model of the negotiator holds great promise. The ability to model true beliefs and evaluation methods has an advantage over more traditionally prescriptive models. The neural network model allows incorporation of human irrationality and provides an ability to assess how that irrationality affects the negotiation outcome.