They have eight arms lined with suckers that can latch on to just about anything, yet octopuses never seem to get tangled up in their own limbs. That’s because a self-recognition mechanism in the skin of the common octopus prevents its suckers from sticking to the animal, according to a new study published today in the journal Current Biology.
No one’s really looked at this nonstick phenomenon before, says Frank W. Grasso, director of the BioMimetics and Cognitive Robotics Laboratory at Brooklyn College, City University of New York. The mechanism behind it, though, is a surprisingly simple way to control the fairly complex system that is an octopus’s arm—and it has piqued the interest of researchers in the field of robotics.
“Two-thirds of [an octopus'] nerves are not in its brain, but in its arms,” says James Wood, a marine biologist affiliated with the University of Hawaii on Oahu. Those nerves enable the arms to quickly change shape, color, and texture and allow the suckers to grasp and taste objects, he says.
Octopuses have a neural network that is distributed throughout their body yet ultimately controlled by the brain, says Wood, who was not involved in the study. “We don’t really know how the whole system works,” he adds. But the new study is giving researchers a glimpse into how the common octopus (Octopus vulgaris) controls such a complicated system as its arm.
When an octopus’s sucker contacts a surface, “a local reflex triggers attachment,” says study co-author Grasso. Even if that arm is amputated—and it’s not uncommon for octopuses to lose limbs in the wild—the arm remains active and able to move and grasp objects for about an hour.
When Grasso and colleagues looked at the behavior of amputated octopus arms in the laboratory, the researchers found that the suckers didn’t latch on to the arm itself or to other octopus arms covered in skin.
If they removed the skin from the arms, the suckers on the amputated limb attached themselves to the skinless arm. The scientists also covered half of a petri dish with octopus skin and found that suckers on amputated arms grabbed the uncovered half of the dish while avoiding the half covered in skin.
When the scientists offered the amputated arm to intact octopuses, the creatures either treated the limb like food (common octopuses are cannibalistic), didn’t touch the arm at all, or stuck one end in their mouth and carried it around.
These results suggested some kind self-recognition mechanism that prevented the suckers from attaching to octopus skin, says Grasso. Some of the octopus behavior also suggested that the brain could override that mechanism. Researchers aren’t sure what kind of mechanism is at work, though.
Building Better ‘Bots
“It’s a very interesting finding,” says Cecilia Laschi, a biorobotics professor at the Scuola Superiore Sant’Anna in Pisa, Italy, who was not involved in the study. Not only does it answer an intriguing biological question, but it could also help researchers build better robots, she says.
Traditionally, scientists have gotten a robot to move by programming the machine’s “brain” to solve all the calculations needed to move an appendage from point A to point B, says Grasso. “This made for really slow and impractical robots.”
In an attempt to simplify things and produce more realistic movements, researchers now try to pattern their robots after the way animals move. As long as you understand the basic principles of animal movements, you should be able to simplify the directions you give to a robot, Grasso says.
Octopuses are interesting, he says, because their arms can move in a nearly infinite number of ways. “Controlling it is sort of a nightmare from a computational point of view,” Grasso says. “[So] we look for simplifying principles.”
The self-recognition ability in octopus arms is one such simplifying principle. Applying it to a robot should be possible, says Laschi, although it would probably be a mechanical system rather than a chemical one, as is likely the case in the octopus.
Laschi says findings such as this could contribute significantly to the field of so-called soft robotics, machines that are able to change their shape as needed in the course of executing tasks.
Follow Jane J. Lee on Twitter.