Cultural Dissemination/Assimilation

A culture can be viewed as a collection of ethnic traits: norms, customs, technologies, laws, politics, language, art. Cultural dissemination refers to the tendency for one culture to adopt some of the traits of a neighboring culture.

But why are there different cultures today? Why haven't all cultures blended into one universal culture by this time? Why do cultural islands form? Is there some natural dynamic that explains this?

Assimilation Simulator (AS)

AS is a customization of the CA framework that allows users to simulate the spread of cultural traits. In AS the only agents are NetLogo patches. Each patch represents an ethnic region. The cultural traits of an agent are represented by a list of integers.

For example, suppose we are only interested in four cultural traits: economy, language, technology, and political organization. For simplicity, let's assume that each of these traits has only four possible values. For example:

In this case every patch owns a list containing four integers in the range zero to four inclusive. For example, here is a snapshot of nine patches at a particular moment in time:

Inhabitants of the patch in the middle are farmers (economy = 2). They speak a Nahwatal language (language = 2). They use bronze tools and weapons (technology = 1). Politically, they are a democracy (organization = 3).

Measuring Cultural Similarity

We can define the cultural similarity between two cultures as the percentage of traits they have in common. For example, the following table shows the percentage of traits the middle patch has in common with each of its eight neighbors:

Which two cultures shown are the most similar?

Cultural Adaptation

During each cycle (cycle = year, decade, century, etc.) each patch, P, is given a chance to borrow a cultural trait of one of its neighbors. P randomly picks one of its neighbors, N. If the cultural similarity between P and N is greater than some randomly generated percentage, then P randomly picks one of N's traits, t, and replaces its corresponding trait by t.

Domain Model

Here's a class diagram that models some of the concepts we have discussed:

Using AS

The interface for AS is simple. It consists of an initialization button, a update button, a cycle counter, and a view. (Students are welcome to add other controls and plots as they see fit.)

Initializing a patch involves giving it a list of five random integers between 0 and 4, then calling color-patch. The color-patch procedure is called after each update. This procedure is discussed in the Information tab of ca.nlogo. The problem is that there are 4⁴ = 256 possible cultures, and only 140 NetLogo colors. This means two cultures will have to share a color. Also, we would like similar cultures to have similar colors.

Cycle = 0

Here's a screen shot just after initialization:

Cycle = 8502, Green Hegemony

Clicking the run button initially starts the simulation. The cycle counter starts updating and the color patterns begin shifting. Subsequently clicking on the run button pauses and resumes the simulation.

Here's what the world looks like after 8502 cycles:

Notice the hegemony of the greenish cultures.

Cycle = 12352, The Yellow (and Orange) Peril

Times change, the introduction of TV, Internet, cars, and jumbo jets allows culture to disseminate in ways that are difficult to predict. Here's what our world looks like after 12353 cycles:

Note that the greenish cultures are in abeyance while the yellowish and orange-ish cultures are beginning to dominate.

Cycle = 2184, The Greens are Back!

Cycle = 26907, The World is Green!

Imagine a day when all people wear Michael Jordan shoes, even Earth's president. Everyone is a born-again Christian, and there is a MacDonald's on the summit of Everest.

Of course my getColor function is defined in such a way that it is possible that there could still be spontaneous eruptions of non-green cultural traits.

Implementation of AS

Extensions

A. . Add text controls that allow the user to adjust the values of numTraits and maxTraitValue. This means generalizing the color-patch procedure.

B. For each trait, add a pen to a plotter that shows how many patches have that trait.

References

The idea for this project came from Robert Axelrod's essay Disseminating Culture, which can be found in his book: The Complexity of Cooperation: Agent-based Models of Competition and Collaboration.