metal fish sculpture

Biologists agree on many things that living organisms can do: they eat, they respire, they reproduce, they die. Many would also agree, implicitly or explicitly, that living things compute. But a trio of SFI researchers want to know: What does it mean for biological systems to carry out computations? The answer isn’t clear.

“Biologists have a vague sense of what they mean by computation,” says biologist Albert Kao, an SFI Complexity Postdoctoral Fellow.

“At a certain point, if your definition is too loose, then anything can be computation. But if you make it too tight, then maybe nothing is computational in biology.”

For three days this fall, biologists, physicists, neuroscientists, and computer scientists will come together for an SFI workshop to investigate the links between computational theory and biological systems. The workshop is the brainchild of SFI Professor David Wolpert, who leads the thermodynamics of Computation project at SFI. It was co-organized by Kao and SFI Professor Jessica Flack, who runs the Institute’s Collective Computation Group.

Flack says the workshop’s goal “is to first establish a rigorous conceptual framework for studying biological computation so that the foundations of computation in adaptive systems can be identified and compared to those in synthetic computing systems.”

Workshop participants will identify components of living systems that might look like elements of computation. They may have possible inputs and outputs, for example, or processes that look like algorithms. Researchers have also observed that the modular, hierarchical organization of biological systems resembles that of digital computers: A body is organized into separate organs, each containing cells that are organized into many separate organelles.

Finding links between computers and biological system has been a longstanding goal of researchers on the edge of both fields. “This topic has deeply puzzled scientists for decades,” says Wolpert. “However, there have recently been breakthroughs in other fields, like collective computation, coarse-graining dynamical systems, and non-equilibrium statistical physics, that we believe can help us make major progress on this topic.”

The organizers say the benefits will run both ways: Insights from computer scientists will help guide new ways of thinking about biological computation, and biologists may help computer scientists find ways to build scalable, more robust machines.

“If you remove half of a school of fish, the remaining half can still do things,” Kao says. “If you take away half a CPU, it’s dead. What new ways of computing can we think of that compute scientists don’t typically think of?

Read more about the "What is biological computation?" workshop