In 2009, scientists unleashed an amoeba-like blob on to Tokyo, and watched as it consumed everything in sight. In less than a day, the blob had spread throughout the entire city, concentrating itself along major transport routes.

Fortunately for the citizens of the great Japanese metropolis, the blob did its work on a model. Flakes of oats stood in for the major urban zones and the scientists involved were no B-movie villains. Rather, they were biologists studying the sophisticated behaviour of a slime mould, an oozing blob of goo that performs feats of apparent intelligence despite being completely brainless.

The slime mould Physarum polycephalum spends most of its life as a yellow mat, sliding among the leaf litter in its search for food, such as bacteria and fungal spores. This mat is a gigantic single cell, called a plasmodium, which forages by sending out dozens of tendrils from a central mass. The branches of this living network grow and shrink, emerge and vanish, according to what they encounter.

Without a brain, Physarum makes decisions by committee. The plasmodium is a single sac but it behaves like a colony. Every part rhythmically expands and contracts, pushing around the fluid inside. If one part of the plasmodium touches something attractive, like food, it pulses more quickly and widens. If another part meets something repulsive, like light, it pulses more slowly and shrinks. By adding up all of these effects, the plasmodium flows in the best possible direction without a single conscious thought. It is the ultimate in crowdsourcing.

Despite its lack of a nervous system, Physarum can still solve problems such as mazes, finding the shortest path between two exits baited with food. At first, it extends its tendril network throughout the entire labyrinth before trimming away the dead-end branches, leaving behind a single thick tube that covers the most direct path. If parts of the maze are bathed in unpleasant bursts of light, Physarum will find an alternative route that sticks to the shadows.

Here’s where miniature Tokyo comes in. Physarum is so good at finding quick routes between different places that Atsushi Tero from Hokkaido University wanted to see if it could match human town planners. He placed the mould in his food-based model of Tokyo, with patches of light representing impassable terrain. Sure enough, the mould filled the entire area before thinning out into selective connections between the food sources. These final networks were strikingly similar to Tokyo’s actual railway system, and comparable in terms of efficiency and resistance to problems.

Physarum accomplishes all of this without any forethought. It behaves like a town planner who cakes a city in rails before strengthening what works and demolishing what does not. It seems haphazard but this technique, honed by millennia of evolution, allows the giant cell to efficiently shunt nutrients from one end to another. The result is a biological network that matches the best man-made efforts.

Physarum is also a skilled decision-maker. At the University of Sydney, Tanya Latty and Madeleine Beekman found that it can weigh up different options to make the best possible choice. If it encounters several food morsels of varying nutritional quality, it devours them in the right proportions to get a balanced diet.

The slime mould is also vulnerable to problems that plague human decisions. Latty’s latest studyin the Proceedings of the Royal Society B: Biological Sciences, shows that when choosing between foods, the moulds that make the quickest choices also make more bad decisions, while those that take their time are more accurate.

Its preferences can also be swayed through simple marketing tactics. Physarum has no inclination towards either a hefty food chunk bathed in light or a medium-sized piece in shadow; both options have pros and cons. But Latty and Beekman changed the slime mould’s behaviour by giving it a third option that makes one of the originals seem more attractive – a small morsel in shadow, say. Human businesses use the same tactics to influence their customers too – unattractively basic products can make expensive ones seem more desirable, while even expensive products look like a steal if they’re placed next to even more costly alternatives (think vintage wine). Despite the gulf of intelligence that separates us, both humans and slime moulds like to compare our options, rather than paying attention to their absolute values.

Whether any of this actually counts as intelligence, though, is debatable. Physarum will not be challenging chess grandmasters any time soon but in a 2001 paper, Toshiyuki Nakagaki, who worked on the Tokyo and maze-solving studies wrote: “I consider that the Physarum’s maze-solving is smart or something like primitive intelligence.” But he added that, by these standards, “all biological systems must be rather smart”.

Andy Adamatzky from the University of West England agrees. “Physarum’s intelligence is not higher than intelligence of a stone rolling down a hill (the stone “chooses” a shortest path downhill) or a plant orienting itself towards the sun,” he says. “Physarum just obeys physical, chemical and biological laws.”

Studying the slime mould may seem like a spot of trivial fun. Indeed, for his maze study, Nakagaki even won an Ignobel prize, an award given to research that makes you laugh, then think. But there is no denying that its simple behaviour can lead to complex feats.

Some scientists are trying to tap into this ability for a variety of practical uses. Take the Tokyo experiment. Tero thinks that the slime mould’s abilities could help planners to design better networks by using biological principles, and he has created a computer model to simulate the mould’s style of decision-making.

Adamtzky describes Physarum as a living computer, which is driven by a massively parallel core processor and can be controlled by food or light. These properties can be used to integrate the slime mould into machines. Klaus-Peter Zauner from the University of Southampton managed to control a six-legged robot using a slime mould grown in the shape of a six-pointed star. He used beams of light to move each of the mould’s branches, and a computer translated these movements to the robot’s own legs. It was the first time that a robot had been driven by living cells. Other groups have used moulds as sensors to sniff out chemicals, engines to drive miniature boats with rhythmic pulses, or conveyor belts to transport liquids and small particles.

Adamtzky also sees Physarum as an inspiration for artificial robots, which would be very different from conventional designs with electronic hardware driven by central programmes. These new robots would be able to change their shape and run off the collective actions of their various parts, just like a living slime mould. Adamatzky calls them “biological amorphous robots” or, more affectionately, “bloboids” or “plasmobots”. They are like a team of employees who work together for the common good without the direction of a manager. Despite any central intelligence, they would still produce efficient behaviour, which would give them one up over many governments.

Ed Yong blogs at Not Exactly Rocket Science

Read More