National Geographic

Why Octopus Arms Don’t Get Tangled

If you cut off an octopus’s arm, the severed limb will still move about for at least an hour. That’s because each arm has its own control system—a network of around 400,000 neurons that can guide its movements without any command from the creature’s brain.

The hundreds of suckers along each arm can also behave independently. If a sucker touches an object, it will change its shape to form a tight seal, and contract its muscles to create a powerful suction. It grabs and sucks, by reflex.

This setup allows the octopus to control its astonishing appendages without overly taxing its brain. Your arm has a small number of joints and can bend in a limited number of ways. But an octopus’s arm can create as many joints as it wants, in any direction, anywhere along its length. It can also extend, contract, and reshape itself. To control such infinitely flexible limbs, it needs to outsource control to the limbs themselves.

But what happens if one arm brushes past another? If the suckers grab objects on reflex, why aren’t octopuses constantly grabbing themselves by mistake?

To find out, octopus arm expert Benny Hochner teamed up with octopus sucker expert Frank Grasso.“Octopus suckers are undervalued in terms of their complexity,” says Grasso. “I’m one of their proponents. They’re really exquisite manipulation devices.”

Together with Nir Nesher and Guy Levy, the duo noticed that the suckers on a freshly amputated arm will never attach to another arm. Sure, they’ll grab skinned parts of an amputated arm or the bare flesh at the point of amputation, but not the arm itself. They’ll grab Petri dishes, but not those that are covered with octopus skin.

Common octopus. Credit: Pseudopanax.

Common octopus. Credit: Pseudopanax.

Octopuses clearly have some kind of sucker-proof coating on their own skin.  The team confirmed this idea by extracting chemicals from the skins of both fishes and octopuses, and applying these cocktails onto Petri dishes. They found that the octopus extract could block a sucker’s grabbing reflex but the fish extract could not.

“We all knew that octopuses are very dependent on chemical sensing but we haven’t done much research on this,” says Jennifer Mather from the University of Lethbridge, who studies octopus behaviour. “This paper will probably kick start it.”

Whatever the mystery chemical, it’s clear that octopuses can override its influence. The team showed that that living animals will occasionally grab amputated arms, even by the skin. Their brains can veto the reflexes of their suckers.

They can even tell if an amputated arm belonged to them or to another octopus. If they sensed another individual’s severed arm, they would often explore it, grab it, and hold it in their beaks in an unusual posture that the team called “spaghetti holding”. (Common octopuses will cannibalise their own kind, so a floating arm is fair game.) But when they sense their own severed limbs, they typically avoided it, and only rarely treated it like food.

“This gives us some idea of how octopuses might generate a sense of self—not by vision, which would be hopeless given their changeable appearance, but by chemical cues,” says Mather.

The octopus’s self-avoiding arms are a great example of embodied cognition—the idea that an animal’s body can influence its behaviour independently of its brain. As Andrew Wilson and Sabrina Golonka explain, “the brain is not the sole resource we have available to us to solve problems. Our bodies… do much of the work required to achieve our goals.”

The octopus… well… embodies this idea. Its brain governs many of its decisions and exerts control upon its arms, but the arms can do their own thing, including getting out of each others’ way. The animal doesn’t need to know the location of each of its arms to avoid embarrassing entanglements. It can let its arms do the work of evading each other.

This concept might be useful for designing robots. A typical robot, like Honda’s ASIMO, relies on top-down programs that control his every action. He can pull off pre-programmed feats like dancing or running, but he trips over even minor obstacles. He’s inflexible and inefficient. By contrast, Boston Dynamics’ Big Dog relies on embodied cognition. His springy legs are designed to react to rough terrain without needing new instructions from his central processor. (Thanks again to Wilson and Golonka for the examples.)

By studying the arms of octopuses, scientists may one day be able to design soft versions of Big Dog, pairing its flexible movements with an equally flexible chassis. Big Octopus, perhaps.

Reference: Nesher, Levy, Grasso & Hochner. 2014. Self-Recognition Mechanism between Skin and Suckers Prevents Octopus Arms from Interfering with Each Other. Current Biology. http://dx.doi.org/10.1016/j.cub.2014.04.024

More on octopuses:

There are 15 Comments. Add Yours.

  1. Mike Hansen
    May 15, 2014

    What if you do the octopus experiment with amputated octopus arms on squid arms, or squid skin? Does that make for a different outcome?

  2. Megan
    May 15, 2014

    This is absolutely fascinating. I love octopuses, easily some of the most amazing creatures on the planet.

  3. Tonia
    May 16, 2014

    It’s Octopuses’ with JUST an apostrophe! English 101! Journalism is dead. You only hire idiots anymore…

    [Octopuses are more intelligent than linguistic prescriptivists. - Ed]

    • Megan
      May 16, 2014

      No, he got it right. He used “an octopus’s arm”, a singular possessive. Even if a word ends in s, you use ‘s for singular possessive. Damn, grammar nazism is dead.

  4. check before you troll
    May 16, 2014

    Firstly, I really enjoyed reading about this, it sounds like elegant experimental design. Octopuses are amazing creatures.
    @Tonia – “Octopus’s” is correct as a possessive form of a noun ending in “s”. “Octopuses’ ” would refer to something belonging to more than one octopus. http://www.crockford.com/wrrrld/style2.html

  5. Joel
    May 16, 2014

    Fascinating article on amazing creatures. Their sophistication is only starting to become apparent.

    And Tonia: don’t be a hater! National Geo is doing amazing work. There’s no journalism organisation in the world that doesn’t have an occasional typo. And the one you’ve pointed out is actually debatable depending on what country you’re from.

  6. wbond
    May 16, 2014

    I would prefer “octopi,” but octopuses is OK.

    Tonia, why not the Latin genitive for the possessive rather than “octopus'”? (Although that would be “octopi” for the singular, which might confuse). Oh well. Latin is dead.

    • Megan
      May 16, 2014

      Technically, Latin took the word “octopus” from the Greek, so the correct pluralization would be octopodes, not octopi. We’re speaking English, however, so octopuses is what looks best to me.

  7. Robert C Brooke
    May 16, 2014

    Very interesting.Do other cephalopods have similar abilities?

  8. John
    May 16, 2014

    Tonia. I think It’s so sad that instead of being amazed by an interesting article about the complexities of newly discovered autonomous limbs of the octopus, you scour the ad for the tiniest little dot that is missing, point it out and call all the workers at NG idiots.

    I bet your a rage at all the happening parties.

  9. wbond
    May 17, 2014

    Megan, you are correct, and I a fool. Here, I was trying to make a light-hearted pedantry joke, and I got it all wrong. Speed kills.

    Of course, it’s “New Latin,” from Greek roots that would be octopous. Nonetheless, the “hypercorrect” octopi is much more common than the octopodes form. Greek also has an evidently rare dual number, which would be octopode for two of these beautiful creatures.

    Thanks for not correcting my hypercorrect genitive which is simply a mess. Should have been octopodis or octopodos (Gk).

    My point, however, shared by others, was: use whichever darn plural you want, and put the apostrophe wherever you please. Who cares. Neat story.

  10. zackoz
    May 18, 2014

    On the sense of self angle, I assume octopuses have been subjected to the mirror test?

    I don’t understand the point about not using vision to to test an octopus’s sense of self.

    Might it not recognise itself in a mirror, if it’s as smart as everyone says?

  11. Surya Koturan
    May 18, 2014

    Fascinating! I have often wondered this about octopuses. There could be room for scientific inquiry to why an octopus would eat floating bits of other octopuses and not its own…Just a thought. I would want to know the reason.

  12. heidi B
    May 18, 2014

    Oh Dear Lord…. Could we all just let the grammar go and focus on the science…

  13. Andrew D Wilson
    May 25, 2014

    @zackoz:

    The mirror test is highly over rated, as is ‘sense of self’. BF Skinner got pigeons to pass the mirror test just by analysing the relevant behavioural contingencies: http://www.robertlanza.com/self-awareness-in-the-pigeon/

Add Your Comments

All fields required.

Related Posts