Modeling a Society
Main Menu Previous Topic Next Topic
The Society Model: Static Properties
Note: in order to run the simulation referred to in this slide, go here, to the Java Applet version. You will be directed to download the latest version of the Java plug-in.
Agents in our simulator are represented by the blue circles you see on the left. Each agent is given an endowment of information resources which is illustrated by the size of the circle. Larger circles represent more resources and smaller circles represent fewer resources with differences being purely relative and not absolute. In this example, all endowments are randomly generated numbers initially falling between 0 and 1.
Agents want to connect to each other in order to access the knowledge resources held by other agents. However, there are two reasons agents cannot simply connect to everybody.
The first limitation is that of channel count. This is our implementation of a rationality bound -- agents are constrained by time and energy in communicating with others. In order to capture this idea, we suppose that agents only have a certain number of channels that they can use to connect to partners in each time period. This number is known as the channel count. In the society to the left agents were given a channel count of two to connect with.
The second limitation, a main focus of this tutorial, is access. In this model, access is an integer denoting the number of potential partners that a particular agent has the ability to connect to. These partners are the ones that are geographically closest. As access increases, geography becomes more and more irrelevant, until with universal access, agents can select any other existing agent to connect to.
Because of these restrictions, agents cannot connect to everybody. Thus, with limited connections, agents prefer to connect to those with the most information, so as to gain access to the most resources possible.
Previous Slide Next Slide