Main Menu           Previous Topic                                                           Next Topic

Assumptions About Information Interactions

In order to construct a model that simulates a real population, it is necessary to single out the characteristics of real interaction that influence behavior. The Information Elites model is based on four main assumptions about information interactions:

1. People cannot talk to or listen to everyone else at once. As a proxy for bounded rationality, this means that they will need to select some subset of the population with which to communicate. It's been suggested that we might be able to communicate with as many as 3000 people, but out of 270 million in the US alone, this is a pretty small fraction (see the paper for this and all subsequent references).

2. We assume that information isn't lost, consumed, or destroyed when it is shared with others. In economic terms, information goods are "nonrival." If one person gives another a favorite recipe, for example, that information doesn't necessarily become unavailable to the giver -- information isn't destroyed and multiple agents can still use it. This also allows us to model the possibility, however, that people may lose strategic advantage if resources are transfered.

3. Private resources differ in quality, amount, value etc. That is, people know different things and private resources are heterogeneous. It has also been suggested that information which might once have been public is becoming increasingly privatized as people recognize its value. For example, trade secrets, patents and copyrights increasingly privatize software, genetic codes, and protein manufacturing.

4. People can improve their resources based on the resources to which they can gain access. If one person can gain access to a Nobel prize winner for example, but another cannot, then the first is likely to have access to a richer source of information. This implies that social networks or networks of connections can become very important in improving resources over time.

Only the first two assumptions are important for our observation that increasing access and decreasing search costs brought on by information technology are likely to increase "cyber-stratification." The other assumptions help us make a few observations about the improvements to resources over time.

                   Previous Slide                                                           Next Slide