Mapping Tokyo's Train System in Slime Mold
In the future, urban planners could lose their jobs to ultra-efficient slime molds.
That's if the research of an international consortium of protist-obsessed scientists translates to real-life railroad building. For more than a decade, Atsushi Tero and his posse have been investigating the path-finding ways of Physarum polycephalum, the “many-headed slime,” which survives by sending out tendrils in all directions in the quest for food. The unsuccessful tentacles wither and die, while the ones that locate a food source in the most efficient fashion grow fat and juicy. This setup produces a network of trails that just so happen to look a lot like your city's commuter-rail map.
Back in 2010, Tero, whom you may recall from the paper "Simulation of a soft-bodied fluid-driven amoeboid robot that exploits thixotropic flow," put his army of molds to the ultimate public-transit test: recreating the byzantine rail system coursing around Tokyo. Once the pieces were in place, this turned out to be extraordinarily easy.
The researchers arranged little oat flakes on a gel in the same pattern of real cities near Tokyo in the Kanto region. Then they stimulated a slime mold to grow from the center outward, connecting with all the delicious Oatvilles in a highly precise and unappetizing game of connect-the-dots. The whole process took just over a day, and at the end there was the environs of Tokyo, outlined in ooze. (Abstract and study.)
Simulation results, taken from the paper "Rules for Biologically Inspired Adaptive Network Design."
Basking in their success, the researchers noted that even when the mold didn't zig or zag exactly like its concrete-and-steel analogue, it found an equally efficient path. Such flexibility makes them think that molds will one day serve in place of human brains and computers when it comes to designing networks, whether for public transit or power lines or something else. From an AFP article on the ongoing revolution in slime:
Atsushi Tero at Kyushu University in western Japan, said slime mold studies are not a "funny but quite orthodox approach" to figuring out the mechanism of human intelligence.
He says slime molds can create much more effective networks than even the most advanced technology that currently exists.
"Computers are not so good at analysing the best routes that connect many base points because the volume of calculations becomes too large for them," Tero explained.
"But slime molds, without calculating all the possible options, can flow over areas in an impromptu manner and gradually find the best routes.
Losing your job to something that reproduces via spores – think it couldn't happen? Slime molds are already solving mazes, so if you're not doubling down on your studying you might have something to worry about.