Collective Chaos Spawns a New Science

How can insect swarms help researchers tease out patterns in online chatter? It turns out ant colonies and Facebook commenters are similar; both are more than the sum of their parts.

When collective behavior emerges from individuals—be they insect swarms, traffic jams or Internet users—scientists call the resulting system a complex network. With some digging, researchers at UC Davis are discovering basic mathematical principles can help predict behavior in complex networks. The findings could lead to improvements in real-world human networks, including air traffic, the stock market and the U.S. power grid.

Scientists aren’t the only ones who seek patterns in the natural world. Artists also extract meaning from chaos. At UC Davis, an alliance between art and physics served to illuminate nature’s beautiful complexity.

The collaboration includes Dawn Sumner, chair of the Department of Earth and Planetary Sciences, James Crutchfield, professor of physics, and artist/writer Meredith Tromble of the San Francisco Art Institute. The team also included researchers at the college’s Complexity Sciences Center (CSC) and KeckCAVES, a 3-D data visualization facility.

Tromble created a virtual installation built from the inner lives of physicists, Take Me To Your Dream (Dream Vortex). Made with drawings projected in the KeckCAVES 3-D space, a vortex of dream images appears in the air before the viewer; the images can be “touched” and handled, as if they were physical objects. Seed dreams for the prototype were collected from scientists, such as a physicist’s dream of lecturing to daisies.

Melding art and science offers an interesting perspective on organizing information, said Crutchfield, director of the CSC.

“The collisions that happen when you talk across disciplines are very helpful in coming up with new insights into complex net-works,” Crutchfield said. “And here at Davis, as an ag school, we have a history of people from different backgrounds solving problems.”

Complexity science

At UC Davis, the CSC provides an organizing home for faculty with interests in complex systems. Complexity science is a particularly fertile field for interdisciplinary work, Crutchfield said.

Everything on Earth is part of a complex network. These networks can be physical, such as neurons, or abstract, as with online acquaintances. 

The CSC is home to more than 30 researchers who work in both theory and applications of complex networks. One project, led by Raissa D’Souza, professor of mechanical and aerospace engineering and computer science, explores how to intervene in networks and make them more robust. The research could help alleviate widespread power outages by preventing one power grid failure from taking down the national network.

Crutchfield has been interested in complex systems and chaotic behavior since the 1970s, when he was a UC Santa Cruz graduate student. At UCSC, Crutchfield was part of a group called the Chaos Cabal, which did much of the work in developing chaos theory. Today, Crutchfield explores how humans recognize new patterns in nature—patterns we have never seen before. “There is a very real possibility we can use machines to find emergent organization in our vast data sets,” Crutchfield said. 

Big data

While the CSC focuses on the physics of information, campus researchers also explore the intersection of big data and science.

To gather and study the vast amounts of data available from networks requires new tools and skills from scientists. In 2014, the provost and chancellor funded a new initiative to tackle these big data challenges.

The Data Science Initiative (DSI) has several goals, including training researchers at all levels, from undergraduates to faculty and sparking new collaborations between faculty. “We want to give people the skills to ask good questions of their data and interpret the results,” said professor of statistics Duncan Temple Lang, director of the DSI and researcher in the CSC.

Data science collaborations already underway on campus include analyzing rivers and watersheds, parsing health records and scanning historical texts. “The thing that brings everybody together is data,” said Temple Lang, who focuses on integrating computer science research with scientific and statistical research. 

— Becky Oskin, content strategist in the UC Davis College of Letters and Science