Contrary to expectations and Star Wars propaganda, an army of clones might not be a particularly effective fighting force. To create something that works together most effectively, whether for war or peace, it is better to build it from elements that differ from each other than have them all be exactly the same. The proof is a physical one, but it could have applications from renewable energy to biology.
We naturally assume the more similar items are, the more likely they are to act in the same way. However, science is all about overthrowing assumptions. Four years ago, Dr Takashi Nishikawa and Professor Adilson Motter published a paper arguing that not only are differences between components of a system no barrier to producing matching behavior, they are sometimes essential.
This sounds like the sort of abstract mathematics that barely connects with reality, but in Nature Physics, Nishikawa, Motter, and Dr Ferenc Molnar have demonstrated what they call “scenarios in which interacting entities are required to be non-identical in order to exhibit identical behavior” in the real world.
We know from examples such as heart cells and pedestrians falling into step that, in the right circumstances, things with naturally different rhythms can synchronize. Even physicists expected similarities in the components would assist this process (for example, if the walkers naturally walked at the same pace).
Molnar built a set of electrically powered oscillators that could be coupled together and applied friction as a dampening force. In some experiments, all three oscillators were matched. In others, the friction applied to each varied. As Motter noted in a statement, identical objects will behave identically when separate, but this can go out the window once they start to interact, with feedback between them inducing changes. Hard as it is to visualize, the three found the output frequencies matched best when the oscillators experienced different, but not too different, dampening.
“This is remarkable mathematically, let alone physically. So, many colleagues thought that experimentally demonstrating this effect was impossible,” Nishikawa said.
As unrelated to the practical world as this may seem, the authors argue: “The results presented here will naturally extend to real systems with tunable node parameters, such as networks of logic gates, neuronal systems, coupled lasers, and networks of mechanical, electrical and chemical oscillators.”
The authors are already trying to use their work to improve stability in electricity grids dependent on wind and solar power. However, the implications extend far further. Fish and birds act in unison to evade predators, and male Asian fireflies are most attractive to mates when their flashing keeps time with each other. This finding raises the possibility that animals' synchronization may work best when diversity exists within those that need to align, a potentially important lesson if we seek to replicate or harness these systems.