The trails of harvester ant foragers are captured by image analysis.

The trails of harvester ant foragers are captured by image analysis. Photo: AnTracks

"Ants, it turns out, have much to teach us about the decentralized networks that operate in many biological systems, in which local interactions produce global behavior, without the guidance of any central intelligence or authority." – Deborah Gordon, Boston Review, 2010

Ant colonies are fascinating because no ant is in charge or tells any other ant what to do. An ant uses only what it perceives nearby, mostly odor and vibration, to figure out what to do next. I study how the interactions of ants produce the organization of the colony. To do this, I watch ants and try to figure out what they are doing. At first I focus on function. Is the ant collecting food or building a nest? Then I ask how the ants work together.

Ants interact by touching antennae. Ants smell with their antennae, and when one ant touches another, it can assess whether the other ant belongs to the same colony. There are more than 11,000 species of ants. For at least some species, brief antennal contact is a cue to the task the other ant is performing. An ant uses these recent interactions to decide what to do. To understand how colonies work together is to understand what ants react to and how simple interactions produce the colonies’ responses.

To learn about ants, I often count something, such as how often an ant met other ants. Sometimes I intervene to see how the ants react. Watching, thinking about how to interpret what I see, figuring out how to set up a way to measure what the ants are doing so I can find out if I’m wrong, finding out how to change their world enough to make a difference but not so drastically as to push them outside the range of ordinary behavior—this is the practice of science. It is both systematic and creative.

I think that science is much like art, except that science has formal rules that, we hope, allow us to work in a way that would lead anyone else who did the same thing to the same conclusion. The scientist seeks generality in results that can be replicated. Good science is produced by practices that set an agreed standard for what is true.

Deborah Gordon and Felipe Campos Cerda observing ants in the tropical forest at Chamela, Mexico.

Deborah Gordon and Felipe Campos Cerda observing ants in the tropical forest at Chamela, Mexico. Photo: Courtesy of Deborah Gordon

I find ways to visually explore patterns that ants generate with their behavior. This process shapes my decisions about what questions to ask next. We use image analysis to trace the paths of ants. We can extract data from this visual information; for example, we can measure the flow of foragers by asking how quickly their paths pass an imaginary line across the foraging trail. The images also give a sense of the pattern created by the ants’ paths, and thinking about these patterns raises new questions, which in turn require further data to answer. Representations of nature in many forms – visual, mathematical and metaphorical – are all a part of the process of scientific research. Practicing science often involves making art, as making art involves experimentation.