When it comes to teamwork, there's a lot we can learn from ants. In Panama, a particular species of army ant displays inspiring acts of solidarity with a distinctly architectural flare: when the terrain gets dicey, they band together to form scaffolds with their bodies. These army ants usually march along at a quick pace, so it's remarkable that they stop completely to create a safer route for the colony as a whole. Researchers at the Max Planck Institute of Animal Behavior, Germany, led by Ian Couzin, decided to study these feats of camaraderie as a model for how the behavior of individuals can give rise to order in a complex system without explicit directions.
To characterize whether aspects of the environment influence the scaffolds that the ants form, the research team turned the ants’ usual route into a perilous highway. They elevated a stretch of road a few centimeters above the ground and used sticks and leaves to create ramps on either end. In the middle of the elevated route, they positioned a platform which could be positioned at different angles to create a stretch of sloped terrain. The platform was covered in sandpaper to allow the ants to get a stable footing, so that any slipperiness would be due to the angle of the terrain. By watching this hazardous stretch of road, the researchers found that ants were more likely to form scaffolds on steeper terrain and that more ants joined scaffolds at steeper angles. They also observed that fewer ants tumbled off the platform once the scaffolds formed, which shows that scaffolds are an effective strategy for preserving the colony.
Next, the researchers wanted to test whether the ants start forming a collective structure as a result of their own experience of the terrain, possibly prompted by a loss of footing. They discovered that as the terrain becomes more inclined, an ant is more likely to slip, which would then trigger an instinct to stop marching. They built a computational model to test whether they could predict how many ants would join the scaffold and how quickly it would form based on the steepness of the incline and the density of traffic. If their model fell short of what they observed in the real ant colony, it would suggest that there are other factors governing the ants’ collective behavior. Instead, they found that their models accurately predicted what they had observed in the real ant colony. The model predicted that steeper inclines would lead to scaffolds made up of more ants at a similar rate to what they had observed. The computational model also predicted the success of the structures: fewer ants would fall as time elapsed and the scaffold grew.
The algorithms didn't include an expressed objective to build a structure, so the accurate predictions validated their theory that the ants start to form scaffolds based on their own individual experiences of their environments, not as a result of a widely broadcasted signal to do so. For an army ant, taking a moment to stop in your busy trek to ensure your comrades’ safety is a no-brainer.
