Yidan Chen Period 2
Project Proposal 

Title
Multiagent Modeling of Societal Development and Cultural Evolution

Purpose
The purpose is to simulate human social evolution from the application of 
agent-based modeling techniques. The project will allow for an in-depth 
understanding of computer-based simulations, as well as the conduct of a study 
into certain behavioral phenomenons that occur in human gatherings, an 
important aspect of social sciences and economics.

Scope
Although the dynamics of a behavioral model can become ever more complex in 
administrating new social rules, for the purpose of this class, the artificial 
society will initially limit itself to the basic rules of human survival (ie: 
food and reproduction). Then, the society will implement more advanced rules of
civilization (ie: disease, migration, trade). If time permits, the society 
will further its activities to intergroup relations (ie: formation of tribes, 
cultural exchanges and tribal conflicts). Basic parameters such as natural 
resources and initial population are subject to change.

Background
This project is based on the original Sugarscape, outlined in Growing Artificial
Societies (Epstein and Axtell, 1996). The code of the original Sugarscape is 
written using Ascape. Other developers, such as Tony Bigbee, have implemented 
the project in other configurations (ie: MASON, NetLogo) as well. Other forms 
of agent-based modeling systems (ie: RePast, Swarm) also exist, however the 
system currently in development is MASON from GMU.

Methodology
The project will build on the algorithm of another agent-based modeling program 
(ie. Sugarscape in NetLogo or RePast) and begin with a prototype in MASON to 
duplicate the fundamental rules of Sugarscape, G for reproduction of essential 
resources and M for agent movement. After establishing the initial parameters 
of the agent world, the parameters of the agents will encompass gender, age, 
multiple resources, culture, etc to establish reproduction, trade, cultural 
exchange, etc. Data recording will be implemented along with the development of
the agent-world rules, to display statistical information of the artificial 
society.

Results
The basic statistics of agent movement, reproduction, wealth distribution based
on the accumulation of resource will initially be recorded. More complex 
economic data will later be compared to the formation of primitive tribes near 
resources and the exchange of culture, disease, and/or war. The array of events 
and changes in the agent world will be displayed through a Java applet, updating
itself with each iteration. Statistics graphs will also display the data 
collected in real time, with the development of the artificial society.