by Dan Ariely
While Alexa Warburton was researching her senior thesis at Middlebury College’s newly created octopus lab, “every day,” she said, “was a disaster.”
She was working with two species: the California two-spot, with a head the size of a clementine, and the smaller Florida species Octopus joubini. Her objective was to study the octopuses’ behavior in a T-shaped maze. But her study subjects were constantly thwarting her.
The first problem was keeping the octopuses alive. The 400-gallon tank was divided into separate compartments for each animal. But even though students hammered in dividers, the octopuses found ways to dig beneath them—and eat each other. Or they’d mate, which is equally lethal. Octopuses die after mating and laying eggs, but first they go senile, acting like people with dementia. “They swim loop-the-loop in the tank, they look all googly-eyed, they won’t look you in the eye or attack prey,” Warburton said. One senile octopus crawled out of the tank, squeezed into a crack in the wall, dried up, and died.
It seemed to Warburton that some of the octopuses were purposely uncooperative. To run the T-maze, the preveterinary student had to scoop an animal from its tank with a net and transfer it to a bucket. With bucket firmly covered, octopus and researcher would take the elevator down to the room with the maze. Some octopuses did not like being removed from their tanks. They would hide. They would squeeze into a corner where they couldn’t be pried out. They would hold on to some object with their arms and not let go.
Some would let themselves be captured, only to use the net as a trampoline. They’d leap off the mesh and onto the floor—and then run for it. Yes, run. “You’d chase them under the tank, back and forth, like you were chasing a cat,” Warburton said. “It’s so weird!”
Octopuses in captivity actually escape their watery enclosures with alarming frequency. While on the move, they have been discovered on carpets, along bookshelves, in a teapot, and inside the aquarium tanks of other fish—upon whom they have usually been dining.
Even though the Middlebury octopuses were disaster prone, Warburton liked certain individuals very much. Some, she said, “would lift their arms out of the water the way dogs jump up to greet you.” Though in their research papers the students refer to each octopus by a number, the students named them all. One of the joubini was such a problem they named her the Bitch. “Catching her for the maze always took twenty minutes,” Warburton said. “She’d grip onto something and not let go. Once she got stuck in a filter and we couldn’t get her out. It was awful!”
Then there was Wendy. Warburton used Wendy as part of her thesis presentation, a formal event that was videotaped. First Wendy squirted salt water at her, drenching her nice suit. Then, as Warburton tried to show how octopuses use the T-maze, Wendy scurried to the bottom of the tank and hid in the sand. Warburton says the whole debacle occurred because the octopus realized in advance what was going to happen. “Wendy,” she said, “just didn’t feel like being caught in the net.”
Data from Warburton’s experiments showed that the California two-spots quickly learned which side of a T-maze offered a terra-cotta pot to hide in. But Warburton learned far more than her experiments revealed. “Science,” she says, “can only say so much. I know they watched me. I know they sometimes followed me. But they are so different from anything we normally study. How do you prove the intelligence of someone so different?”
Measuring the minds of other creatures is a perplexing problem. One yardstick scientists use is brain size, since humans have big brains. But size doesn’t always match smarts. As is well known in electronics, anything can be miniaturized. Small brain size was the evidence once used to argue that birds were stupid—before some birds were proven intelligent enough to compose music, invent dance steps, ask questions, and do math.
Octopuses have the largest brains of any invertebrate. Athena’s is the size of a walnut—as big as the brain of the famous African gray parrot Alex, who learned to use more than one hundred spoken words meaningfully. That’s proportionally bigger than the brains of most of the largest dinosaurs.
Another measure of intelligence: you can count neurons. The common octopus has about 130 million of them in its brain. A human has 100 billion. But this is where things get weird. Three-fifths of an octopus’s neurons are not in the brain; they’re in its arms.
“It is as if each arm has a mind of its own,” says Peter Godfrey-Smith, a diver, professor of philosophy at the Graduate Center of the City University of New York, and an admirer of octopuses. For example, researchers who cut off an octopus’s arm (which the octopus can regrow) discovered that not only does the arm crawl away on its own, but if the arm meets a food item, it seizes it—and tries to pass it to where the mouth would be if the arm were still connected to its body.
“Meeting an octopus,” writes Godfrey-Smith, “is like meeting an intelligent alien.” Their intelligence sometimes even involves changing colors and shapes. One video online shows a mimic octopus alternately morphing into a flatfish, several sea snakes, and a lionfish by changing color, altering the texture of its skin, and shifting the position of its body. Another video shows an octopus materializing from a clump of algae. Its skin exactly matches the algae from which it seems to bloom—until it swims away.
For its color palette, the octopus uses three layers of three different types of cells near the skin’s surface. The deepest layer passively reflects background light. The topmost may contain the colors yellow, red, brown, and black. The middle layer shows an array of glittering blues, greens, and golds. But how does an octopus decide what animal to mimic, what colors to turn? Scientists have no idea, especially given that octopuses are likely colorblind.
But new evidence suggests a breathtaking possibility. Woods Hole Marine Biological Laboratory and University of Washington researchers found that the skin of the cuttlefish Sepia officinalis, a color-changing cousin of octopuses, contains gene sequences usually expressed only in the light-sensing retina of the eye. In other words, cephalopods—octopuses, cuttlefish, and squid—may be able to see with their skin.
The American philosopher Thomas Nagel once wrote a famous paper titled “What Is It Like to Be a Bat?” Bats can see with sound. Like dolphins, they can locate their prey using echoes. Nagel concluded that it was impossible to know what it’s like to be a bat. And a bat is a fellow mammal like us—not someone who tastes with its suckers, sees with its skin, and whose severed arms can wander about, each with a mind of its own. Nevertheless, there are researchers still working diligently to understand what it’s like to be an octopus.
Jennifer Mather spent most of her time in Bermuda floating facedown on the surface of the water at the edge of the sea. Breathing through a snorkel, she was watching Octopus vulgaris—the common octopus. Although indeed common (they are found in tropical and temperate waters worldwide), at the time of her study in the mid-1980s, “nobody knew what they were doing.”
In a relay with other students from six-thirty in the morning till six-thirty at night, Mather worked to find out. Sometimes she’d see an octopus hunting. A hunting expedition could take five minutes or three hours. The octopus would capture something, inject it with venom, and carry it home to eat. “Home,” Mather found, is where octopuses spend most of their time. A home, or den, which an octopus may occupy for only a few days before switching to a new one, is a place where the shell-less octopus can safely hide: a hole in a rock, a discarded shell, or a cubbyhole in a sunken ship. One species, the Pacific red octopus, particularly likes to den in stubby brown glass beer bottles.
One octopus Mather was watching had just returned home and was cleaning the front of the den with its arms when, suddenly, it left the den, crawled a meter away, picked up one particular rock, and placed the rock in front of the den. Two minutes later, the octopus ventured forth to select a second rock. Then it chose a third. Attaching suckers to each rock, the octopus carried the load home, slid through the den opening, and carefully arranged the three objects in front. Then it went to
sleep. What the octopus was thinking seemed obvious: “Three rocks are enough. Good night!”
The scene has stayed with Mather. The octopus “must have had some concept,” she said, “of what it wanted to make itself feel safe enough to go to sleep.” And the octopus knew how to get what it wanted: by employing foresight, planning—and perhaps even tool use. Mather is the lead author of Octopus: The Ocean’s Intelligent Invertebrate, which includes observations of octopuses who dismantle Lego sets and open screw-top jars. Coauthor Roland Anderson reports that octopuses even learned to open the childproof caps on Extra Strength Tylenol pill bottles—a feat that eludes many humans with university degrees.
In another experiment, Anderson gave octopuses plastic pill bottles painted different shades and with different textures to see which ones evoked more interest. Usually each octopus would grasp a bottle to see if it was edible and then cast it off. But to his astonishment, Anderson saw one of the octopuses doing something striking: she was blowing carefully modulated jets of water from her funnel to send the bottle to the other end of her aquarium, where the water flow sent it back to her. She repeated the action twenty times. By the eighteenth time, Anderson was already on the phone with Mather with the news: “She’s bouncing the ball!”
This octopus wasn’t the only one to use the bottle as a toy. Another octopus in the study also shot water at the bottle, sending it back and forth across the water’s surface, rather than circling the tank. Anderson’s observations were reported in the Journal of Comparative Psychology. “This fit all the criteria for play behavior,” said Anderson. “Only intelligent animals play—animals like crows and chimps, dogs and humans.”
Aquarists who care for octopuses feel that not only can these animals play with toys, but they may need to play with toys. An Octopus Enrichment Handbook has been developed by Cincinnati’s Newport Aquarium, with ideas of how to keep these creatures entertained. One suggestion is to hide food inside Mr. Potato Head and let your octopus dismantle it. At the Seattle Aquarium, giant Pacific octopuses play with a baseball-sized plastic ball that can be screwed together by twisting the two halves. Sometimes the mollusks screw the halves back together after eating the prey inside.
At the New England Aquarium, it took an engineer who worked on the design of cubic zirconium to devise a puzzle worthy of a brain like Athena’s. Wilson Menashi, who began volunteering at the aquarium weekly after retiring from the Arthur D. Little Corporation sixteen years ago, devised a series of three Plexiglas cubes, each with a different latch. The smallest cube has a sliding latch that twists to lock down, like the bolt on a horse stall. Aquarist Bill Murphy puts a crab inside the clear cube and leaves the lid open. Later he lets the octopus lift the lid. Finally he locks the lid, and invariably the octopus figures out how to open it.
Next he locks the first cube within a second one. The new latch slides counterclockwise to catch on a bracket. The third box is the largest, with two different locks: a bolt that slides into position to lock down and a second one like a lever arm, sealing the lid much like the top of an old-fashioned glass canning jar.
All the octopuses Murphy has known learned fast. They typically master a box within two or three once-a-week tries. “Once they ‘get it,’” he says, “they can open it very fast”—within three or four minutes. But each may use a different strategy.
George, a calm octopus, opened the boxes methodically. The impetuous Gwenevere squeezed the second-largest box so hard she broke it, leaving a hole two inches wide. Truman, Murphy said, was “an opportunist.” One day, Murphy put two crabs inside the smaller of the two boxes, and the crabs started to fight. Truman was too excited to bother with locks. He poured his seven-foot-long body through the two-inch hole Gwenevere had made, and visitors looked into his exhibit to find the giant octopus squeezed, suckers flattened, into the tiny space between the walls of the fourteen-cubic-inch box outside and the six-cubic-inch one inside it. Truman stayed inside for half an hour. He never opened the inner box—probably he was too cramped.
Three weeks after I first met Athena, I returned to the aquarium to meet the man who had designed the cubes. Menashi, a quiet grandfather with a dark mustache, volunteers every Tuesday. “He has a real way with octopuses,” Dowd and Murphy told me. I was eager to see how Athena behaved with him.
Murphy opened the lid of her tank, and Athena rose to the surface eagerly. A bucket with a handful of fish sat nearby. Did she rise so eagerly sensing the food? Or was it the sight of her friend that attracted her? “She knows me,” Menashi answered softly.
Anderson’s experiments with giant Pacific octopuses in Seattle prove Menashi was right. The study exposed eight octopuses to two unfamiliar humans, dressed identically in blue aquarium shirts. One person consistently fed a particular octopus, and another always touched it with a bristly stick. Within a week, at first sight of the people, most octopuses moved toward the feeders and away from the irritators, at whom they occasionally aimed their water-shooting funnels.
Upon seeing Menashi, Athena reached up gently and grasped his hands and arms. She flipped upside down, and he placed a capelin in some of the suckers near her mouth, at the center of her arms. The fish vanished. After she had eaten, Athena floated in the tank upside down, like a puppy asking for a belly rub. Her arms twisted lazily. I took one in my hand to feel the suckers—did that arm know it had hold of a different person than the other arms did? Her grip felt calm, relaxed. With me earlier, she had seemed playful, exploratory, excited. The way she held Menashi with her suckers seemed to me like the way a long-married couple holds hands at the movies.
I leaned over the tank to look again into her eyes, and she bobbed up to return my gaze. “She has eyelids like a person does,” Menashi said. He gently slid his hand near one of her eyes, causing her to slowly wink.
Biologists have long noted the similarities between the eyes of an octopus and the eyes of a human. The Canadian zoologist N. J. Berrill called it “the single most startling feature of the whole animal kingdom” that these organs are nearly identical: both animals’ eyes have transparent corneas, regulate light with iris diaphragms, and focus lenses with a ring of muscle.
Scientists are currently debating whether we and octopuses evolved eyes separately or whether a common ancestor had the makings of the eye. But intelligence is another matter. “The same thing that got them their smarts isn’t the same thing that got us our smarts,” says Mather, “because our two ancestors didn’t have any smarts.” Half a billion years ago, the brainiest thing on the planet had only a few neurons. Octopus and human intelligence evolved independently.
“Octopuses,” writes philosopher Godfrey-Smith, “are a separate experiment in the evolution of the mind.” And that, he feels, is what makes the study of the octopus mind so philosophically interesting.
The octopus mind and the human mind probably evolved for different reasons. Humans—like other vertebrates whose intelligence we recognize (parrots, elephants, and whales)—are long-lived, social beings. Most scientists agree that an important event that drove the flowering of our intelligence was our ancestors’ beginning to live in social groups. Decoding and developing the many subtle relationships among our fellows, and keeping track of these changing relationships over the course of the many decades of a typical human life span, was surely a major force shaping our minds.
But octopuses are neither long-lived nor social. Athena, to my sorrow, may live only a few more months: the natural life span of a giant Pacific octopus is only three years. If the aquarium added another octopus to her tank, one might eat the other. Except to mate, most octopuses have little to do with others of their kind.
So why is the octopus so intelligent? What is its mind for? Mather thinks she has the answer. She believes the event driving the octopus toward intelligence was the loss of the ancestral shell. Losing the shell freed the octopus for mobility. Now they didn’t need to wait for food to find them; they could hunt like tigers. And while most octopuses love crab best, t
hey hunt and eat dozens of other species—each of which demands a different hunting strategy. Each animal you hunt may demand a different skill set. Will you camouflage yourself for a stalk-and-ambush attack? Shoot through the sea for a fast chase? Or crawl out of the water to capture escaping prey?
Losing the protective shell was a tradeoff. Just about anything big enough to eat an octopus will do so. Each species of predator also demands a different evasion strategy—from flashing warning coloration if your attacker is vulnerable to venom, to changing color and shape for camouflage, to fortifying the door to your home with rocks.
Such intelligence is not always evident in the laboratory. “In the lab, you give the animals this situation, and they react,” points out Mather. But in the wild, “the octopus is actively discovering his environment, not waiting for it to hit him. The animal makes the decision to go out and get information, figures out how to get the information, gathers it, uses it, stores it. This has a great deal to do with consciousness.”
So what does it feel like to be an octopus? Philosopher Godfrey-Smith has given this a great deal of thought, especially when he meets octopuses and their relatives, giant cuttlefish, on dives in his native Australia. “They come forward and look at you. They reach out to touch you with their arms,” he said. “It’s remarkable how little is known about them . . . but I could see it turning out that we have to change the way we think of the nature of the mind itself to take into account minds with less of a centralized self.”
“I think consciousness comes in different flavors,” agrees Mather. “Some may have consciousness in a way we may not be able to imagine.”
In May I visited Athena a third time. I wanted to see if she recognized me. But how would I be able to tell? Scott Dowd opened the top of her tank for me. Athena had been in a back corner but floated immediately to the top, arms outstretched, upside down.