Studying the Evolution of Higher Intelligence on MSU's Supercomputer

Dr. Chris Adami is a Professor of Microbiology and Molecular Genetics and of Physics and Astronomy at Michigan State University. He utilizes tools from evolutionary biology, systems biology, computer science and cognitive sciences to explore subjects such as the evolution of complex behaviors and intelligence and the evolution of cooperation.

“My lab works on the interface between life sciences and physics," Dr. Adami says. “We use theoretical and computational tools to understand how simple rules give rise to complex systems and behaviors.”

For one of his intelligence studies, Dr. Adami’s lab created “artificial brains” on MSU’s high-performance computer. These programs were allowed to evolve and subsequently analyzed using information theory and automata theory. In the cooperation study, they evolved game-playing strategies within an agent-based scenario, characterizing the emerging strategies from the point of view of information transmission and communication.

Other research topics include the evolution of drug resistance in HIV-1, the evolution of behavior in animals, and the evolution of networks. This research utilizes large-scale digital experiments to study novel effects in the evolution of populations and to test recent mathematical formulations of evolution. In this research, Dr. Adami and his team worked closely with MSU’s Institute for Cyber-Enabled Research (ICER) and the High Performance Computing Center (HPCC).

“The computational work in my group has the crucial support of the HPCC,” says Dr. Adami. “Because evolution is intrinsically unpredictable, making statements about the significance of an evolutionary outcome requires tremendous replication, which we can do by running on hundreds of nodes in parallel. This research would not be thinkable without high-performance computing support."

Dr. Adami says high-performance computing sits at the intersection of work being done by biologists, physicists, engineers and computer scientists. He says his field exists in this interface between the physical sciences and the biological sciences.

“ICER here at Michigan State University is ideal because we have the type of computer scientists who understand biological problems and we have biologists that also understand the computer aspects of the problem,” Dr. Adami says. “There are not that many institutes in the world that are so interdisciplinary where we can ask these questions and get the kind of support from both sides of the field.”