Slime molds are yellow, oozing, amoeba-like organisms often found on decaying logs and in moist areas. They have no neurons of any kind, not to mention a brain. Each organism consists of just a single, giant cell that can extend for meters. Yet, they are capable of surprisingly complicated, almost intelligent behaviors.
In fact, during some amazing experiments, scientists found that slime molds like Physarum polycephalum can even rival human engineers.
During one such experiment, researchers in Japan arranged oat flakes on a dish to match the positions of cities around the Japanese capital. They placed the slime mold P. polycephalum at the point representing Tokyo. Then they watched as the organism spread outward, touched the food, and slowly morphed itself.
Thin tendrils appeared. Some thickened into tubes while others vanished. After about a day, the living network that remained looked uncannily like the rail system around one of the world’s largest cities.
Approximately 20 million passengers use Tokyo’s extended, interconnected railway system daily, making it the busiest in the world. The network spans the Tokyo Metropolis and neighboring prefectures, with 13 subway lines and vast commuter lines. Some of the best human minds have worked on this network to make it as efficient as possible, which makes it all the more amazing that a brainless blob arrived at the same solution for such a complex optimization problem.
A Creature that Can Act Intelligently Without the Ability To Think
Slime molds are highly efficient at exploring their environment and making use of the resources they find there. Researchers have harnessed this ability to solve mazes and other problems under controlled conditions. Credit: Audrey Dussutour, CNRS.
At its smallest, Physarum exists as microscopic cells that actively swim around. When they swarm together, they merge into a single giant cell called a plasmodium, which can extend for meters. It moves at a top speed of four centimeters per hour by extending tendrils in any direction. A single plasmodium can tear itself into fully functioning pieces, and those pieces can fuse right back together.
Every part of the slime mold rhythmically expands and contracts, pushing the fluid inside around. If one tendril touches something attractive, like food, it pulses faster and widens. If it bumps into something repulsive, like bright light, it shrinks and pulses slower. By adding up all of these localized effects, the organism crowdsources its way toward the best possible direction that balances finding food with avoiding threats without a single conscious thought.
This simple pulsing mechanism yields extraordinary results. When Physarum finds scattered food sources, it engulfs them. To keep exploring efficiently, it transforms its sprawling, Byzantine web of tendrils into a streamlined network of tubes. The paths carrying the most nutrients expand, while the unused links gradually contract and disappear.
×
Thank you! One more thing…
Please check your inbox and confirm your subscription.
The Tokyo Rail Network Test
Tokyo rail network formation with Physarum polycephalum
In a 2010 study published in the journal Science, researchers led by Toshiyuki Nakagaki and Atsushi Tero from Hokkaido University put the slime mold’s biological optimization to the test. They presented the slime mold with a real-world problem that requires balancing competing objectives.
“The planning is very difficult because of the tradeoffs,” says cell biologist Mark Fricker of the University of Oxford, who was also involved in the research. Connecting all cities with the shortest possible track forces some travelers to take highly indirect routes, and a single deviation from the most optimal network graph could isolate a large part of the network. Building in more redundancy makes the system more resilient, but at a much higher cost. Human engineers spend countless hours trying to balance these priorities.
The researchers placed oat flakes on agar plates in a pattern mimicking the cities surrounding Tokyo, and dropped the slime mold in the center. They even used illumination to represent geographical constraints like mountains and lakes, which the light-avoidant mold steered clear of.
Initially, the slime mold dispersed evenly, exploring its new territory. But over the next day, it began refining its pattern, strengthening the tunnels between the oat flakes while letting redundant links fade.
Engineering by Amoeba
The final result was astonishing. The organism had constructed a network of interconnected tubes that looked almost identical to the finely engineered rail system surrounding Tokyo.
Every day, the Tokyo rail network ferries millions of people quickly and reliably. “In contrast, the slime mold has no central brain or indeed any awareness of the overall problem it is trying to solve, but manages to produce a structure with similar properties to the real rail network,” Fricker says. “There is a remarkable degree of overlap between the two systems,” he adds.
The researchers didn’t stop at admiring the mold’s handiwork. They extracted the simple rules dictating its behavior to create a biologically inspired mathematical model.
Biochemist Wolfgang Marwan of Otto von Guericke University notes that the slime mold’s behavior “is really difficult to capture by words.” “You see they optimize themselves somehow, but how do you describe that?” he asks. This research, he wrote, “provides a simple mathematical model for a complex biological phenomenon.”
Watch slime mold form a map of the Tokyo-area railway system.
Mathematical Slime
This adaptable model could guide the design of future technological networks that need to adjust on the fly. For instance, short-range wireless sensor networks used for early flood or fire warnings could self-organize. Because these sensors are often destroyed when disaster strikes, the network needs to quickly reroute information. Decentralized networks would also benefit swarms of robots navigating hazardous environments or soldiers on a battlefield.
“The idea would be that, if one put it into a new context, a system using these rules would build a network that ought to have respectable properties,” says Fricker.
But not everyone is ready to hand complicated infrastructure engineering duties over to a blob. Melanie Mitchell, a computer scientist at Portland State University, calls the work “a very interesting example of how biology can inspire new methods in technological design.” However, she adds a note of caution: “This paper uses only one relatively simple example,” she says. “It’s not obvious that similar experiments would work as well for matching other transport networks.”
Researchers using slime mould to compare motorway systems found the US network to be among the least efficient. Photograph: Andrew Adamatzky and Jeff Jones
Yet, the mold’s mapping skills have been repeatedly tested. Andrew Adamatzky from the University of West England later performed similar experiments mapping other countries, including the U.S., the U.K., and China. He noted that the mold had some similar ideas to us humans.
“We found that the slime mold approximated almost all interstates,” Adamatzky wrote in the New York Times.
The Brainless Majority
In experiments conducted by Dussutour’s team, disks of yellow slime mold (at bottom) can eat plates of oatmeal (at top) — but only if they cross gelatinous bridges (at center) laced with noxious but harmless compounds. Here, the middle slime mold sample has learned to disregard the chemicals, a process called habituation. Credit: Audrey Dussutour, CNRS.
The slime mold’s architectural prowess forces us to reconsider what it means to be intelligent. If anything, it’s a reminder that centralized brains aren’t the only way to process information.
In fact, Audrey Dussutour, a researcher at the CNRS in France, has shown that slime molds can even learn. By forcing them to cross bridges laced with repellents like salt or coffee to reach food, she found that the organisms gradually habituated to the deterrents. They got used to the chemicals, ignored them, and moved faster.
Incredibly, if a “naïve” slime mold merges with a habituated one, the resulting fused organism retains the memory and ignores the chemical barriers.
“Most people thought that it was impossible for a cell to learn,” Dussutour told The Atlantic. “But we’ve tried this now with more than 2,000 slime molds. It can’t be an accident.”
How a single cell stores these memories without neurons remains a mystery. “We’ve talked to a lot of neuroscientists and they have no idea,” she says.
But perhaps we shouldn’t be so shocked. Biologist Michael Levin of Tufts University previously showed that flatworms can retain memories even after regrowing a decapitated head, proving memory doesn’t strictly depend on neurons. “On the one hand, it’s not terribly surprising,” Levin told The Atlantic. “It has to be encoded in some biophysical change in cells; something different and perduring has to occur as a result of experience, otherwise memory wouldn’t work. Whatever that medium is, inside of cells, why wouldn’t it be transferable?”
By studying Physarum, we are peering into the deep evolutionary roots of cognition and network design, honed over countless cycles of evolutionary selection pressure.
“I think we’re beginning to realize that brains are not prerequisites for complex and interesting behavior,” says Tanya Latty from the University of Sydney. “The majority of life forms on Earth are brainless, but we know very little about how this brainless majority are able to adapt their behavior in changing environments.”