When biologist Roland Anderson of the Seattle Aquarium pulled back the tank's lid, I wasn't sure whether it was to let me get a look at Steve or to let Steve get a look at me. Clearly, Steve was looking—his big hooded eye followed me, and a single five-foot-long arm reached out to the hand I held above the water's surface. The arm inched up past my wrist to my shoulder, its suckers momentarily attaching and releasing like cold kisses. I couldn't help feeling as if I was being tasted, and I was, by tens of thousands of chemoreceptors. And I couldn't help feeling as if I were being studied, that a measuring intelligence lay behind that intent eye and exploring arm.
Finally, when the arm's fingerlike tip reached my neck, it shot back like a snapped rubber band. Steve curled into a tight, defensive ball in the corner of the tank. His skin texture changed from glassy smooth to a fissured moonscape; his color changed from mottled brown to livid red—which seemed to signal anger—and he squinted at me. Had something alarmed or offended him? Perhaps we were both a great mystery to each other.
Three-year-old Pandora, a giant Pacific octopus at the Seattle Aquarium, weighs about 45 pounds. Each of her eight arms is about five feet long and lined with 200 suckers, which she uses not only to feel but to taste. "This octopus prefers native littleneck clams to Manila clams [a variety that was introduced into Puget Sound]," says Roland Anderson, a biologist at the aquarium. An octopus also has sharp eyesight.
Octopuses and their cephalopod cousins the cuttlefish and the squid are evolutionary oxymorons: big-brained invertebrates that display many cognitive, behavioral, and affective traits once considered exclusive to the higher vertebrates. They challenge the deep-seated notion that intelligence advanced from fish and amphibians to reptiles, birds, mammals, early primates, and finally humans. These are mollusks, after all—cousins to brainless clams and oysters, passive filter feeders that get along just fine, thank you, with a few ganglia for central nervous systems. Genetic studies show that mollusk ancestors split from the vertebrates around 1.2 billion years ago, making humans at least as closely related to shrimps, starfish, and earthworms as to octopuses. And so questions loom: How could asocial invertebrates with short life spans develop signs of intelligence? And why?
Although biologists are just beginning to probe these questions, those who observe the creatures in their natural haunts have long extolled their intelligence. "Mischief and craft are plainly seen to be the characteristics of this creature," the Roman natural historian Claudius Aelianus wrote at the turn of the third century A.D. Today's divers marvel at the elaborate trails the eight-leggers follow along the seafloor, and at their irrepressible curiosity: Instead of fleeing, some octopuses examine divers the way Steve checked me out, tugging at their masks and air regulators. Researchers and aquarium attendants tell tales of octopuses that have tormented and outwitted them. Some captive octopuses lie in ambush and spit in their keepers' faces. Others dismantle pumps and block drains, causing costly floods, or flex their arms in order to pop locked lids. Some have been caught sneaking from their tanks at night into other exhibits, gobbling up fish, then sneaking back to their tanks, damp trails along walls and floors giving them away.
That Steve was named Steve was also revealing: Octopuses are the only animals, other than mammals like cuddly seals, that aquarium workers bother to name. So Anderson, Seattle's lead invertebrate biologist, began to wonder: If keepers recognize octopuses as individuals, how much difference is there among individual octopuses? Might these bizarre-looking mollusks have personalities? And if so, how else might their evolution have converged with ours across a billion-year chasm?
Meanwhile, in the waters off Bermuda, Canadian comparative psychologist Jennifer Mather was asking similar questions. Mather had observed an Octopus vulgaris, the common Atlantic octopus, catch several crabs and return to its rock den to eat them. Afterward it emerged, gathered four stones, propped these at the den entrance and, thus shielded, took a safe siesta. The strategy suggested qualities that weren't supposed to occur in the lower orders: foresight, planning, perhaps even tool use.
When Mather and Anderson met at a conference, they discovered they had stumbled onto similar phenomena and began collaborating. Other scientists had already tested the ability of octopuses to solve mazes, learn cues, and remember solutions. They had found that octopuses solve readily, learn quickly and, in the short term, remember what they have learned. Mather and Anderson delved deeper, documenting a range of qualities and activities closely associated with intelligence but previously known only in advanced vertebrates. Some of their work has been controversial, and some of their conclusions have been disputed. But other researchers are now confirming their key points and logging even more startling findings.
Anatomy confirms what behavior reveals: Octopuses and cuttlefish have larger brains, relative to body weight, than most fish and reptiles, larger on average than any animals save birds and mammals. Although an octopus brain differs from a typical vertebrate's brain—it wraps around the esophagus instead of resting in a cranium—it also shares key features such as folded lobes, a hallmark of complexity, and distinct visual and tactile memory centers. It even generates similar electrical patterns. Electroencephalograms of other invertebrates show spiky static—"like bacon frying," says neurophysiologist Ted Bullock of the University of California at San Diego, who nonetheless found vertebratelike slow waves in octopuses and cuttlefish. The pattern, he says, is "similar to but weaker than that of a dog, a dolphin, a human."
Researchers at the Konrad Lorenz Institute for Evolution and Cognition Research in Austria recently found one more telling indicator: Octopuses, which rely on monocular vision, favor one eye over the other. Such lateralization, corresponding to our right- and left-handedness, suggests specialization in the brain's hemispheres, which is believed to improve its efficiency and which was first considered an exclusively human, then an exclusively vertebrate, attribute.
The mystery deepens. According to the social theory of intelligence articulated by N. K. Humphrey and Jane Goodall, complex brains blossom in complex social settings; chimps and dolphins have to be smart to read the intentions of other chimps and dolphins. Moreover, such smarts arise in long-lived animals: Extended childhoods and parental instruction enable them to learn, and longevity justifies the investment in big brains. But many cephalopods live less than a year, and the giant Pacific octopus, which has one of the longest documented life spans, survives for only four years. Their social lives are simple to nonexistent: Squid form schools, but they don't seem to establish individual relationships. Cuttlefish gather while young and later on to mate, but they don't form social structures. Octopuses are solitary; they breed once, then waste away and die. Females tend their eggs, but the tiny hatchlings are on their own. As cephalopod-respiration expert Ron O'Dor of Dalhousie University in Nova Scotia wonders, "Why would you bother to get so smart when you're so short-lived?"
For Jennifer Mather, pursuing those questions marks a convergence of childhood and adult passions. Mather grew up in Victoria, British Columbia, along a biologically rich shoreline. "I got fascinated with intertidal life," she recalls. "I always thought I'd study mollusks." In college, she took an animal-behavior class and had an epiphany: "Most people in comparative psychology compare humans and other primates," she observes, which leaves the field wide open for studies of mollusk behavior and cognition. "And if you talk about mollusk behavior, you're talking about cephalopods."
Mather landed in an unlikely spot for marine research: at the University of Lethbridge in landlocked Alberta, which hasn't had any cephalopods since the Devonian Period. But in the 1980s academic jobs were scarce. Mather then found a lab with Anderson in Seattle and a field base near a secluded coral reef off Bonaire, an island in the Netherlands Antilles. There she leads an international investigation of communication and interactions among Caribbean reef squid—the first long-term study of a wild cephalopod population.
Billye, a giant Pacific octopus, has a brain the size of a walnut. But she can solve puzzles she would never encounter in the wild, such as opening a glass jar to remove the fish inside. At right, (1) Billye is offered the jar. (2) She grasps the jar and, in (3) and (4), manipulates it into position and twists. (5) The top is released. Biologist Roland Anderson says he was not surprised that Billye could perform such a complex task. "Back in the '50s," he says, "Jacques Cousteau found that octopuses could open a corked bottle. He would put the cork in the bottle with a little hole in the top, and the octopus would reach inside and remove the cork."
In Seattle, Mather and Anderson have pursued octopuses. Perhaps their most startling and controversial finding is that individuals show distinct personality traits, the first ever measured in an invertebrate. They found that octopuses confronted with the same threat alerts and food stimuli react in different ways. One might flee, but another might fight or show curiosity. That sets them apart from other invertebrates, says Shelley Adamo, a psychologist at Dalhousie who has studied both cephalopods and insects. For example, individual crickets may behave differently at different times—singing today and silent tomorrow. But they don't have consistent patterns that set one cricket apart from another.
Personality can be a controversial word. Some behavioralists call such labels anthropomorphic, while others contend that it's anthropocentric to presume other animals cannot have personalities. Some of Mather and Anderson's peers feel more comfortable with the findings than the terminology. "They do good work and ask interesting questions," says cephalopod researcher John Cigliano of Cedar Crest College in Allentown, Pennsylvania. "But I'm not entirely convinced. It's a tricky business just coming up with a definition of personality." David Sinn, a graduate student at Portland State University, followed up Mather and Anderson's personality work with a more extensive study that they coauthored. That study avoided the "p" word, charting the emergence of key "temperamental traits" in 73 lab-bred California octopuses. It found considerable temperamental variation and distinct developmental stages. Like mammals, Sinn's octopuses were more active and aggressive when young and grew more alert to danger as they matured—evidence that their behavior was learned.
Previous researchers tested octopuses in artificial mazes; Mather and Anderson found ways to observe learning and cognition in more natural circumstances. They charted the efficiency and flexibility with which giant Pacific octopuses switch strategies to open different shellfish—smashing thin mussels, prying open clams, drilling tougher-shelled clams with their rasplike radulae. When served clams sealed with steel wire, for example, octopuses deftly switched from prying to drilling.
Tool use was once commonly invoked as uniquely human. Scientists know better now, but they still cite it as evidence of distinguishing intelligence in chimpanzees, elephants, and crows. Mather describes several ways octopuses use their water jets as tools: to clean their dens, push away rocks and other debris, and drive off pesky scavenger fish.
In 1999 she and Anderson published an even more sensational claim: that octopuses engage in play, the deliberate, repeated, outwardly useless activity through which smarter animals explore their world and refine their skills. Amateur aquarists were the first to suspect that octopuses played. While still in high school, James Wood, now a marine biologist at the University of Texas's marine lab in Galveston, watched his pet octopus grab, submerge, and release her tank's floating hydrometer as if she were a toddler with a bath toy. She also spread her mantle and "bubble surfed" the tank's aerator jets.
Anderson tested for play by presenting eight giant Pacific octopuses with floating pill bottles in varying colors and textures twice a day for five days. Six octopuses examined the bottles and lost interest, but two blew them repeatedly into their tanks' jets. One propelled a bottle at an angle so it circled the tank; the other shot it so it rebounded quickly—and on three occasions shot it back at least 20 times, as if it were bouncing a ball.
One respected cephalopod expert isn't convinced. Jean Boal, an animal behaviorist at Millersville University in Pennsylvania, is acutely aware of the dangers of getting carried away when studying these charismatic megamollusks. She previously worked at the Zoological Station in Naples, a wellspring of cephalopod research. In 1992 Graziano Fiorito, a researcher at that lab, announced a bombshell: Octopuses could learn by watching other octopuses. Such observational learning, a hallmark of intelligent social animals, seemed impossible. And it probably was. Other researchers, including Boal, have been unable to reproduce Fiorito's results. Some questioned his methodology, and for a year or two the controversy cast a pall on research into octopus learning.
Boal subsequently withdrew her own initial findings of complex learning by octopuses. She has since carved herself a niche as the field's designated skeptic, often questioning conclusions and urging more rigor. "My bias is to build a case slowly, with careful science," Boal says quietly. "That's not the case with all cephalopod biologists." She doesn't rule out the possibility that octopuses play, but she questions whether the bottle-jetters did: "It could reflect boredom, like a cat pacing."
One authority on play behavior, psychologist Gordon Burghardt at the University of Tennessee in Knoxville, says that as Anderson and Mather describe it, the bottle-jetting would qualify as play. Boredom, he says, can be "a trigger for play." And other confirmation is emerging. Doubting the Seattle findings, Ulrike Griebel of the Lorenz Institute recently conducted more extensive trials. She offered common octopuses varied objects, from Lego assemblies to floating bottles on strings (a favorite). Some octopuses took toys into their nests and toted them along while fetching food—acquisitive behavior that Griebel says "might be an early stage of object play."
Meanwhile, Anderson has been investigating another phenomenon little-noted in invertebrates: sleep. Until recently, only vertebrates were believed to sleep in the full metabolic sense. But Anderson has observed that octopuses, ordinarily hypervigilant, may sleep deeply. Their eyes glaze over, their breathing turns slow and shallow, they don't respond to light taps, and a male will let his delicate ligula—the sex organ at the tip of one arm—dangle perilously.
Stephen Duntley, a sleep specialist at Washington University Medical School in St. Louis, has videotaped similar slumber in cuttlefish, with a twist: Sleeping cuttlefish lie still, their skin a dull brown, for 10- to 15-minute stretches, then flash bold colored patterns and twitch their tentacles for briefer intervals. After viewing Duntley's footage, Anderson suggests the cuttlefish might merely be waking to check for threats. But Duntley says the cycling resembles the rapid eye movement sleep of birds and mammals, when humans dream. If invertebrates undergo a similar cycle, Duntley argues, it would affirm "that REM sleep is very important to learning." Would it also suggest that cuttlefish and octopuses dream? "That's the ultimate question," Duntley responds.
The ultimate question, with octopuses as with other sentient creatures, may be how we should treat them. In 2001 Mather argued in The Journal of Applied Welfare Science that people should err on the humane side, since some octopuses "very likely have the capacity for pain and suffering and, perhaps, mental suffering." If captive cephalopods suffer mentally—or even get "bored," as Boal puts it—then they should benefit from enrichment: amenities and activities that replicate elements of their natural environment. Mather, Anderson, and Wood have urged enriched environments but have no experimental evidence that it makes a difference. Recently that evidence came from a French study that even the skeptical Boal calls "beautiful work." Ludovic Dickel, a neuroethologist at the University of Caen, found that cuttlefish raised in groups and in tanks with sand, rocks, and plastic seaweed grew faster, learned faster, and retained more of what they learned than those raised alone in bare tanks. Performance rose in animals transferred midway from impoverished to enriched conditions and declined in those transferred to solitary confinement.
"Octopuses can bend their arms in every direction," says biologist Jeff Christiansen of the Seattle Aquarium. "In order to move, they have to work muscles against themselves because they lack bones for the muscles to pull on."
Other evidence suggests that solitary octopuses, like solitary orangutans, may communicate more with others of their species than researchers previously realized. Cigliano found that California octopuses that were kept together quickly established hierarchies and avoided wasteful, dangerous confrontations; the weaker animals seemed to recognize and yield to the stronger ones, even when the latter were hidden in their dens. The flip side of communication is deception, another hallmark of intelligence. And some octopuses and cuttlefish practice it. Male cuttlefish adopt female coloring, patterns, and shape—to mate surreptitiously with females guarded by larger rivals. And the Indonesian mimic octopus fools predators by impersonating poisonous soles and venomous lionfish, sea snakes, and possibly jellyfish and sea anemones.
And so, piece by piece, Mather, Anderson, and other researchers fill in the puzzle. A picture emerges of convergent evolution across a billion-year gap. One after another, these precocious invertebrates display what were supposed to be special traits of advanced vertebrates. But one question nags: Why would short-lived, solitary creatures acquire so many of the cognitive and affective features of long-lived, social vertebrates?
Mather proposes "a foraging theory of intelligence." She says that animals like octopuses (or humans) that pursue varied food sources in changeable, perilous habitats must develop a wide range of hunting and defensive strategies. That takes brainpower. "If you find yourself foraging in a complex environment, where you have to deal with many kinds of prey and predators," she says, "it makes sense to invest more in cognition." Temperamental variation—call it personality—also helps a species survive in a volatile, supercompetitive milieu by ensuring that different individuals respond differently to changing conditions, so some will thrive. Even semelparity, the live-fast-die-young strategy of growing quickly and throwing everything into one breeding blast, may serve that end by assuring rapid turnover and regeneration.
Although cephalopods are an ancient order, shell-less cephalopods are relatively recent arrivals—about 200 million years old, like mammals and teleost, or bony, fishes. Before that, ammonites and other shelled cephalopods ruled the seas, but competition from the nimble, fast-swimming teleosts wiped out all but the relic nautilus. The cephalopods that survived were the zoological counterrevolutionaries that turned the vertebrates' weapons against them. They shed their shells and became speedy, like squid, or they became clever and elusive, like octopuses and cuttlefish. Octopuses, naked and vulnerable, took to dens, as early humans took to caves. Like humans, they became versatile foragers, using a wide repertoire of stalking and killing techniques. To avoid exposure, they developed spatial sense and learned to cover their hunting grounds methodically and efficiently. Mather and O'Dor found that the Bermudan O. vulgaris spends just 7 percent of its time hunting; Australian giant cuttlefish spend 3 percent.
In short, octopuses came to resemble us. Their hunting done, they huddle safely in their dens, a bit like early humans around campfires. "You have to wonder what they think about while they're tucked away," says O'Dor. Do they muse on the cruel turns of evolution, which have left them all dressed up with big brains but with no place to go and little time to use them?
See the online article "Octopuses Are Smart Suckers" by Roland Anderson and Jennifer Mather: is.dal.ca/~ceph/TCP/smarts.html.
James Wood's Cephalopod page has scientific articles, a wealth of information about different species, and excellent FAQ pages: www.dal.ca/~ceph/TCP.
For more hard-core cephalo-science, go to CephBase at www.cephbase.utmb.edu.