Persuasive Argumentation in Negotiation

Abstract

This paper presents Persuasive Argumentation as a means of guiding the negotiation process to a settlement. Decision theoretic approaches construct prescriptive models of the negotiation process that make various assumptions about the behavior of the negotiation participants but do not model changes in behavior. On the other hand, models for decision support leave the actual decisions to human negotiators, again not modeling or automating the negotiating process. In contrast to both approaches, our work deals with automating the negotiation process. This paper focuses on modeling the process by which the beliefs and behavior of negotiators are changed via persuasive argumentation. We claim that persuasive argumentation lies at the heart of negotiation and embodies the dynamics of negotiation. We present a model of persuasive argumentation that integrates Artificial Intelligence and decision theoretic methods. The model has been implemented as part of the PERSUADER, a multi-agent computer program that operates in the domain of labor negotiations.