While much of science seeks to understand complex systems by reducing them to their smallest elements, a team of University of Wisconsin School of Medicine and Public Health neuroscientists argues that studying the big picture can be superior.

William Marshall, PhD, and Larissa Albantakis, PhD, of the Wisconsin Center for Sleep and Consciousness, say that borrowing the notion of “black boxes” from the field of engineering can also help to understand cause and effect in complex biological systems.

At their simplest, black boxes are parts of a larger system that receive inputs and give outputs, and can be measured without understanding exactly what goes on in each box. In everyday life, you can link up and use a new computer, monitor and printer without understanding how each component is put together.

To understand biological systems, sometimes a coarse grain is superior to the fine grain.

While their laboratory studies the brain and consciousness, Marshall and colleagues write that the tool works equally well on biological systems, applying it to understand a fast-growing type of yeast used in laboratory experiments.

“We describe an algorithm that explains how to understand all the causal relationships in the system,” says Albantakis. “And sometimes the parts together have effects that can’t be found by looking at the individual level.”

Marshall says the paper offers an argument that to understand biological systems, sometimes a coarse grain is superior to the fine grain.

“There is a commonly held view that systems should be studied at the finest possible scale, and that higher levels of description are merely approximations that are necessary in practice,” Marshall says. “The result of our work is that – contrary to this view – causal power can increase at higher levels of description.”

Marshall says the hope is to create a practical tool that offers scientists another way to analyze and understand complicated biological systems by studying their larger-scale components. He says the mathematically rigorous algorithm is fully general, and thus applies across many disciplines.

“One benefit of our analysis is that it is observer independent – it provides an objective measure of causal power that can be used to 'carve nature at its joints' and identify the scales at which complex interactions in physical systems come into focus,” Marshall says.

The study was published in the journal PLOS-Computational Biology. The senior author is Giulio Tononi, MD, PhD, director of the Wisconsin Center for Sleep and Consciousness. The work was supported by the Templeton World Charity Foundation, Inc.