How Smart Is the Octopus?
Bright enough to do the moving-rock trick.
By Carl Zimmer
Monday, June 23, 2008,
Aristotle didn't have a high opinion of the octopus. "The octopus is a stupid creature," he wrote, "for it will approach a man's hand if it be lowered in the water." Twenty-four centuries later, this "stupid" creature is enjoying a much better reputation. YouTube is loaded with evidence of what some might call octopus intelligence. One does an uncanny impression of a flounder. Another mimics coral before darting away from a pushy camera. A third slips its arms around a jar, unscrews it, and dines on the crab inside. Scientific journals publish research papers on octopus learning, octopus personality, octopus memory. Now the octopus has even made it into the pages of the journal Consciousness and Cognition (along with its fellow cephalopods the squid and the cuttlefish). The title: "Cephalopod consciousness: behavioral evidence."
So, is the octopus really all that smart? It depends on how you define intelligence. And if you've got a good definition, there are quite a few scientists who would love to hear it. Octopuses can learn, they can process complex information in their heads, and they can behave in equally complex ways. But it would be a mistake to try to give octopuses an IQ score. They are not intelligent in the way we are
-- not because they're dumb but because their behavior is the product of hundreds of millions of years of evolution under radically different conditions than the ones under which our own brains evolved.
You'd have to go back about 700 million years to find the moment in the history of life when humans and octopuses diverged. Our most recent common ancestor, scientists suspect, was a little wormlike creature with eyespots and little more. Since then, our lineage evolved bones; theirs evolved boneless bodies they control with water pressure. We've accumulated so many and such incredible differences over that time that 20th-century scientists were excited to discover a few deep similarities. In the 1950s, for example, biologists demonstrated for the first time that octopuses have massive brains.
Cephalopods belong to the same lineage that produced snails, clams, and other mollusks. A typical mollusk might have 20,000 neurons arranged in a diffuse net. The octopus has half a billion neurons.* The neurons in its head are massed into complex lobes, much the way our own brains are. In comparison with their body weight, octopuses have the biggest brains of all invertebrates. They're even bigger than the brains of fish and amphibians, putting them on par with those of birds and mammals.
In the late 1950s, Oxford biologist N.S. Sutherland decided to put the big brains of octopuses to the test. He would show them two shapes and reward them for touching one but not the other. They might learn to tell a rectangle in a horizontal position from the same rectangle rotated 90 degrees. And once they had figured out this test, the octopuses knew to select any horizontal rectangle they saw, no matter what its particular dimensions. They were learning what to learn.
Over the years, octopuses have shown many more signs of intelligence. They proved to have an excellent memory. They were clever and unpredictable. Jennifer Mather, a Canadian biologist, has tossed toys into octopus tanks and watched as the octopuses inspect them and puff them around with jets of water.* They are playing, she argues. Clams do not play. Humans do.
Mather is also the author of the new paper arguing for consciousness in octopuses. She does not claim that they have full-blown consciousness like we do but a simpler form known as primary consciousness. In other words, they can combine their perceptions with their memories to have a coherent feel for what's happening to them at any moment. Mather bases her claim not just on how octopuses behave but also on how their brains work.
For example, one sign of the complexity of the human brain is that we can be left-handed or right-handed. Our preference comes from one side of the brain dominating over the other--a sign of how the two sides of our brains are not identical. Instead, they divide up mental work and communicate with each other to create a unified sense of reality. Octopuses may not be left-handed (or left-armed), but Mather claims that they show similar kinds of specialization with their eyes. In a 2004 experiment, she and her colleagues found that when they looked out from their dens, some preferred to sit with their left eye facing out, others with their right.
But some octopus experts are skeptical of these bold claims. Many reports of weird octopus behavior come from casual observations in aquariums. Even some experiments have not held up to scrutiny. Last year, Jean Boal of Millersville University and her colleagues found fault with Mather's experiments on left- and right-brained octopuses. The problem was that the scientists had looked at too few octopuses. It was impossible to rule out the possibility that octopuses might not have any preference at all for either eye. The results of the experiments might simply have been a matter of chance.
After 50 years, in other words, we still don't know that much about what's going on in the heads of octopuses. Carefully designed experiments will be essential for finding out more, but so will a more octo-centric attitude. What we call intelligence is really just a set of behaviors and abilities that evolved in our ancestors as they adapted to a particular way of life. Octopuses evolved behaviors of their own, but they were adapting to a way of life that's hard for us to imagine--they were naked mollusks in a world of fish.
The earliest cephalopods, which lived about a half-billion years ago, had shells. Over the next 250 million years, they evolved into giant predators. They shot bursts of water out of siphons to swim--a prehistoric form of jet propulsion.* But their glory was cut short by fish with jaws--our ancestors. Fish could swim faster by bending their bodies than cephalopods could move by jetting. Today, only a single shelled cephalopod survives--the nautilus, which spends most of its life lurking deep underwater.
The other living cephalopods lost their shells. While they gave up a defense against predators, they were free to evolve new skills. Squids became fast swimmers. Octopuses instead moved to the sea floor, where they could use their shell-free bodies to explore cracks and crevices for prey. But in order to survive in this new niche, they had to become fast learners.
Jean Boal and her colleagues have done some experiments that show how good octopuses are at learning geography. Boal put the octopuses in tanks with an assortment of landmarks, such as plastic jugs, plates of pebbles, and clumps of algae. It took only a few trials for the octopuses to find the quickest route to a hidden exit in the bottom of the tank. What made Boal's results particularly impressive is that the octopuses were learning two completely different mazes at once. Boal would move them from one to the other after each trial. Somehow, the octopuses could keep track of two geographies concurrently. When octopuses are moving across new terrain, they can perhaps learn the best escape from predators.
Octopuses escape from predators not just by hiding quickly but by deceit. One of the most impressive examples of this deception is what marine biologist Roger Hanlon calls the moving-rock trick. An octopus morphs into the shape of a rock and then inches across an open space. Even though it's in plain view, predators don't attack it. They can't detect its motion because the octopus matches its speed to the motion of the light in the surrounding water.
For Hanlon, what makes this kind of behavior remarkable is that it's a creative combination of lots of behaviors, used to address a new situation. Similarly, when an octopus escapes an attack, it may puff up its body and turn white to scare a predator, shoot off puffs of ink to distract it, zigzag through the water, and then suddenly switch its skin to match the surrounding coral.
There's not much point in trying to pin this sort of behavior to some human-based scale of intelligence, because our behavior emerged as apes adapted to life spent on two legs, in groups, and using our hands to make tools. We'd fail pretty badly at an octopus-based test of intelligence, but surely we wouldn't hold it against ourselves.
Correction, June 25, 2008: The original version said that early cephalopods used bursts of air to propel themselves, and that octopuses used jets of water to push toys around a tank. In both cases, the animals used water, not air. (Return to the corrected sentence.) The original version also said the octopus has "half a million" neurons instead of "half a billion." (Return to the corrected sentence.)