Half an inch long and blind, a single worker army ant may seem harmless. But when millions of army ants gather in colonies, they can hunt, kill, and eat animals far larger than themselves, such as worms, spiders, and even lizards. One secret to their success: collectively, they know how to find the fastest path between them and their meal. The ants in front first scramble forward randomly, depositing pheromone marks on their chosen path. Ants behind them follow the chemical trail. If they find food, they double back, reinforcing the path by leaving more pheromones. The pheromones evaporate over time—but the markings on the fastest path are strongest, and gradually all the ants follow that one. Together, the ants essentially solve a math problem.
In some situations, an ant colony “thinks” as a unit, in ways similar to a human brain, says SFI External Professor Ricard Solé (Universitat Pompeu Fabra). Like a brain, the colony makes decisions by receiving and processing information. Each ant is like a neuron, sort of. “Individually, the ants don’t know much,” Solé says. “But as a whole, the colonies are problem solvers.” Many other brain-like systems exist throughout nature, such as the human immune system, whose cells learn and remember to identify unfriendly microbes through exposure.
Neuroscience researchers have already developed many mathematical models and tools to describe literal brains. This December 4-5, Solé, along with External Professors Stephanie Forrest (Arizona State University) and Melanie Moses (University of New Mexico) are convening a workshop to discuss how to adapt those tools to study figurative brains such as ant colonies, microbe ecosystems, and the immune system. To do this, they must analyze in detail how the structure and connectivity of a literal human brain differs from these other brain-like systems.
“This working group will explore ways to think differently about the structures that enable ‘thinking’,” says Moses.