This work presents a goal-based model of decision making in which the relative priorities of goals drive the decision process -- a psychological alternative to traditional decision analysis. Building on the work of Schank and Abelson, the author uses goals as the basis for a model of interpersonal relations which permits decisions to incorporate personal and adopted goals in a uniform manner. The theory is modelled on the VOTE computer program which simulates Congressional roll-call voting decisions.
The VOTE program expands traditional decision making and simulation models by providing not only a choice, but also a natural language explanation, in either English or French. It simulates real members of Congress voting on real bills, and producing reasonable explanations. The program is consistent with much of the descriptive political science literature on Congressional decision making and provides an explicit model of political issues, relationships, and strategies that converge in voting behavior.
In developing the VOTE program, the author draws on his own practical experience in politics from four presidential campaigns and the White House. Given the underlying psychological basis of the program, VOTE can be extended to other decision making domains different from politics. Another use for the program is to simulate business decisions such as securities analysis, as well as mundane decision making such as choosing a college or deciding whether to get a Mohawk haircut.
Table of Contents
Contents: Goals and Decisions. Overview of the VOTE Program. Goals. Resources. Interpersonal Relations. The VOTE Program. Decision Strategies. Related Work. Future Work. Appendices: Natural Language Generation. Inside VOTE. Issues. Bills. Groups. Members.
"Two central issues in psychology and artificial intelligence are goals and interpersonal relationships. Reasoning about goals is a fundamental aspect of human behavior. Intelligent agents are not solitary actors, but must make decisions with respect to the goals and plans of others. Slade has built a model that uses the framework of Schank and Abelson (1977) to predict the voting behavior of members of Congress. To do this he had to deal with much that was left out of that initial work and has thus built a stronger model of planful behavior. Slade's work constitutes a fascinating new model of decision making done within a strong theoretical framework based on an analysis of human memory."
"...well-researched and thorough, as well as broadly appealing. I would expect it to be of interest to cognitive scientists working on models of decision making and problem solving, and to artificial intelligence researchers."
ILS, Northwestern University