Computer Science 72: Artifical Life, Culture, and Evolution

Nicholas Gessler

Nicholas Gessler. Les Todd

To most people, the tiny squares blinking red and green all over Nicholas Gessler's computer screen are nothing more than a blinding mess of color. For Gessler, a research scholar in the Information Science and Information Studies program, they help explain the universe.

The conglomeration of squares is part of Conway's Game of Life, a computer program Gessler and students in his course on "Artificial Life, Culture, and Evolution" use to simulate complex social systems.

Early in his career, Gessler worked as an archaeologist. Over time, he became dissatisfied with written language's ability to explain the myriad simultaneous forces behind great social changes. Written records seemed too linear, too oversimplified. "You can explain culture change by telling stories and talking and writing, but those methods don't adequately capture the dynamics of real events," he says.

He soon came upon a promising alternative: computer modeling. Unlike a textbook, computer simulations show multiple causes of culture change acting concurrently and can take into account the parallel, or competing, forces of many agents.

Gessler's class lets students use computers to model behavior patterns of complex social systems ranging in scale from microbes to galaxies. In one program, students explore the effects of racism on society. Squares of many colors, similar to the red and green ones present in Conway's Game of Life, represent people of varying races. Students alter the degree to which squares of the same color favor each other and run their program to see how a city might become segregated. "The course introduces the notion that complex global patterns or behaviors arise from relatively simple local rules, " Gessler says.

By introducing two or three rules that will be followed by individual agents in a programmed system, Gessler's students observe everything from patterns of bird flocking to urbanization. Their models illustrate that most of the successes and failures of culture result not from rational plans, but from many individuals pursuing their own independent goals.

Gessler says cognitive limitations prevent us, as rational human beings, from identifying and manipulating global patterns that can ultimately be traced to our daily behavior. However, by running computer simulations of a cultural system, students can see probable results of social behavior trends before they occur. They learn, Gessler says, how massive change can result from seemingly insignificant actions of individuals.

Through various projects, students attempt to replicate social phenomena from the past, plug in qualities of current society to predict the direction it's headed, and tinker with those qualities to explore what Gessler calls "what-if" scenarios.

However, "Artificial Life, Culture, and Evolution" is not just about examining causes of change in complex systems. The class teaches students to take control of a computer and make it do what they want.

"Computers are not alien devices," says Gessler. "They have really been designed to deal with human problems."

Professor
Nicholas Gessler, whose primary research interests now include artificial culture and experimentation in synthetic anthropology, earned his bachelor's and master's degrees from the University of Alberta. He served as director of what was then the Queen Charlotte Islands Museum in British Columbia from 1973 to 1988 before working in several research posts at the University of California at Los Angeles, where he earned his Ph.D. in anthropology in 2003. He joined the Duke faculty last year.

Prerequisites
None. Open to upperclassmen and graduate students

Readings
Articles relating to anthropology and computing

Assignments
Three hands-on computer programming exams
Five simulation challenges and critiques
Five written responses to readings
One final project: a simulation, analysis, or critique

Share your comments

Have an account?

Sign in to comment

No Account?

Email the editor