Editor's Note: Members of the crow family, known as the corvids, are among the smartest birds in the world. Some are capable of using tools, playing tricks, teaching each other new things, even holding “funerals.” And yet there’s still much we don’t know about these fascinating, sometimes confounding creatures. What’s going on inside the mind of a corvid? Three leading scientists are finding answers.
Nicky Clayton | Eurasian Jays (below)
* * *
“I love you!” says Nicky. “I love you!”
"I love you!” says Lisbon.
Nicky Clayton has shoulder-length blonde hair and a posture that reflects her background in dancing. She is a scientist. She is very smart. Lisbon is a bird, a Eurasian Jay. He’s pretty smart, too. Like most Eurasian Jays, especially the males, Lisbon is also a good mimic. So it’s not clear whether he really loves Nicky, although he certainly likes it when she gives him a worm.
If he loves anyone, it’s probably Rome, his longtime mate. Lisbon and Rome, both eight years old, have been together since they were just two. They share a wired enclosure out here at the edge of Madingley, a peaceful, manicured English village a few miles west of Cambridge.
Clayton, 53, moved to Cambridge University about 16 years ago, around the time when she was becoming an international science superstar for her investigations into avian intelligence. As part of the deal, the university agreed to construct several aviaries at its Madingley annex according to Clayton’s specifications. They’re not fancy, but the birdcages include plenty of space for the captives to fly around, play, and mate, as well as special compartments where they collaborate with Clayton in state-of-the-art bird cognition experiments. Today the aviaries house about 70 birds, including Eurasian Jays, Western Scrub-Jays, and Rooks, all members of the corvid family. At night, the caws and kuks can be heard over much of the village.
On this particular day Clayton is calling on Lisbon and Rome to show off their brainpower in a demonstration exercise. She puts a yellowish-white wax worm on top of a wooden beam that runs across the wire mesh of their cage, then another worm, and another. Lisbon takes one of the wax worms and hides it in the gravel at the bottom of the enclosure. He then picks up two stones with his beak and places them, one by one, on top to mark the spot. (Like many corvids, Eurasian Jays cache their food to eat at another date. Astonishingly, they can faithfully recover hundreds of food items—everything from seeds and berries to young rodents—as long as several months later, during the winter, say, when food is scarce.) Next Lisbon takes another wax worm and flies over to where Rome is sitting on her perch. Lisbon offers her the larva; at first she acts as if she couldn’t care less. (So far no one has been able to explain such feigned indifference.) Eventually, though, she relents and accepts the gift, allowing Lisbon to poke his beak inside hers to deposit the worm. Victorious, Lisbon flies back to collect more wax worms.
Recently Lisbon got a chance to put his skills to a new test. Clayton and her colleagues designed some experiments to see if male Eurasian Jays can anticipate what their female partners want to eat. In the trials Lisbon and six other males watched through the wire mesh as Clayton’s students and post-docs served the females larvae from either wax worms or mealworm beetles. Then it was the males’ turn to feed their partners. Considering the options—worm or beetle—they invariably chose the one that their partners had not already gorged on. For us humans, at least when we’re being thoughtful, intuitively understanding what another person might want is everyday business—will she like sushi or Thai food, Indian or Chinese? But for a bird, it’s a really big deal.
Scientists have long known that birds—especially those in the crow family—are amazingly clever. In fact, some say they may be among the most intelligent animals on Earth, able to accomplish a number of impressive feats that were once thought to be uniquely human—like solving problems, fashioning tools, considering future events, and maybe, as Lisbon demonstrated, pondering another’s state of mind. Yes, a bird can do all that. And Clayton’s research has played a major role in these discoveries. “She is a remarkable person who achieves a huge amount in a short time,” says Peter Slater, a birdsong researcher at the University of St. Andrews in Scotland and Clayton’s former Ph.D. supervisor. “I don’t think she sleeps! She had, and obviously still has, a great knack for identifying interesting questions and designing experiments to address them.”
Nicky Clayton’s interest in birds goes way back. The only child of an organic chemist father and a schoolteacher mother, she grew up in Blackpool, a seaside resort in northwestern England. As early as she can remember, she wanted to fly like a bird. She loved the way birds moved through the sky, how they lived “in such a complex, three-dimensional world.” She spent much of her holiday and vacation time at Leighton Moss, a nature reserve near Blackpool run by the Royal Society for the Protection of Birds. Creeping through the lush reeds, she would train her binoculars on the bitterns, Bearded Tits, and Marsh Harriers the sanctuary is famous for.
In the early 1980s Clayton was accepted to Oxford University’s zoology department to study with John Krebs. An expert in bird behavioral ecology, Krebs would later become known for his discovery that food-storing birds have a relatively large hippocampus, the brain region largely responsible for consolidating memories. During Clayton’s time with Krebs, she zeroed in on how the left and right eyes of Marsh Tits work together when the birds are creating mental treasure maps of where their food is located. Her experiments demonstrated the key role of spatial memory in food-caching birds, and led to that Ph.D. at St. Andrews with Slater.
Decades of earlier research using tape recordings of singing adult birds had clearly shown that young birds learn their songs from older males. But there was a big debate about which males—their fathers, or neighboring birds? Clayton, working with Zebra Finches, showed that both answers could be right, depending on the situation. If the young birds were placed in the same cage with the adult tutors, they would learn from the individual that was most aggressive toward them. But if they were separated from the instructors by wire mesh—which meant they couldn’t be attacked—they would learn from the adult that sounded most like their own fathers, bully or not.
Around the time Clayton’s work was really getting into gear, other scientists were generating their own evidence that avian intelligence had been seriously underestimated. In the 1990s Gavin Hunt, now at the University of Auckland in New Zealand, upset the conventional wisdom about bird brains when he showed that New Caledonian Crows can manufacture tools out of twigs and leaves and use them to forage for insects hiding in logs and trees. Hunt’s work shook up animal behavior researchers nearly as much as Jane Goodall’s did some 30 years earlier when she observed chimpanzees in Tanzania fashioning tools out of blades of grass and using them to fish termites out of the ground. Until then most scientists thought only humans made tools, an assumption commonly referred to as “Man the Toolmaker.” The finding that birds could do so, too—along with, according to more recent research, dolphins, elephants, and even fish—has forever put that prejudice to rest.
But Clayton’s work goes a big step beyond tool use—to the very crux of the bird-brain question. Lisbon’s ability to predict what Rome will want to eat illustrates that he and his fellow jays may share a sophisticated cognitive ability that might relate to “theory of mind”—the ability to attribute mental states, such as pretending, knowledge, desires, intents, and beliefs, to oneself and others, and to understand that others’ mental states might be different than one’s own. This is the definition of empathy. And this could be evidence that it exists in birds.
Her experiments have also revealed that members of the crow family appear to engage in another previously assumed human-only ability called “mental time travel,” the capacity to remember what happened in the past and to plan for the future—like when a child sees an astronaut blast off into space and then decides that he or she wants to grow up to be a space traveler, too. While it is very unlikely that mental time travel in birds is so grandiose, Clayton and her colleagues have shown that it might exist in at least a rudimentary form.
“Calling someone ‘birdbrained’ used to be an insult,” says evolutionary biologist Russell Gray, director of the Max Planck Institute for the Science of Human History in Jena, Germany, who later demonstrated with Gavin Hunt that tool use in New Caledonian Crows was a combination of inherited talent and learned behavior. “But the clever experiments Nicky has conducted have fundamentally changed our view of avian cognition.”
Clayton first encountered Western Scrub-Jays in the mid-1990s, not under the gloomy skies of England but on the green lawns of sunny California, where she had accepted a position on the faculty of the University of California-Davis. She was recruited to the university by Peter Marler, a pioneer in the field of animal behavior most famous for his investigations of the origins of birdsong and his research with Jane Goodall to better understand how gorillas and chimpanzees communicate. Clayton would ponder the topic for her research project while eating lunch outside on the spacious campus. Watching Western Scrub-Jays hide food on the lawn, she began to wonder whether, in addition to spatial memory, the scrub-jays also had something called “episodic memory”—the ability to remember things that had happened in the past, along with when and where.
Episodic memory is a key ingredient in mental time travel. In what would become a turning point in her career, Clayton soon chanced upon Cambridge’s Anthony Dickinson, an expert in cognitive and behavioral neuroscience, at a meeting in Montreal, and the two got to talking. “Tony said that animals don’t have episodic memories—they don’t need it,” Clayton recalls. Dickinson has a similar recollection: “At the time I knew of no examples, nor could I think of a function for episodic memory. All animals require is knowledge about their environment, not memory for specific episodes. Or so I thought at the time.”
While Clayton had a hunch that corvids did indeed have this memory function, proving it was another matter. But she soon figured out a way to do it. First she provided several opportunities for the jays to decide whether to cache fresh wax worms or dry peanuts in different-colored plastic ice cube trays filled with sand. The jays learned quickly that the wax worms rotted over several days and were no longer tasty, while the peanuts had a very long shelf life. Next the birds were allowed to cache wax worms on one side of the tray and peanuts on the other side. (One or the other side of the tray was covered with clear plastic, so they had no choice where to cache each item.) Then they recovered the food, after either four hours or five days. Following the four-hour period, the jays greatly preferred the wax worms, but after five days, they went almost exclusively for the peanuts, even though they favored the worms.
Clayton wanted to be sure that her experiment was airtight—that the birds were relying strictly on memory and not on the smell of spoiled worms to make their choices. So her team performed a second experiment in which, after the birds had cached the food, the researchers took all of the food out of the trays and replaced it with fresh sand. Then they counted how many times the jays poked their bills into the sand on either side of the tray, looking for the food. The jays remembered not only what food they had hidden and where they had hidden it, but also when they had done so.
That work led to yet another landmark study—which came to be known as the “planning for breakfast” experiment—that showed scrub-jays were able to plan ahead. Clayton and her team housed the birds in three compartments, or “rooms,” arranged in a row, and taught them that they would be able to find food in either one or the other end room each morning. In the evening they were treated to pine nuts in the central room, and allowed to hide them away in special trays. The birds learned quickly. They cached three times as many pine nuts in the “breakfast” room than in the other room. (The position of the room varied in each trial of the experiment so that the birds would not identify it simply by its location.) Many researchers hailed these results as evidence that scrub-jays could plan for future situations, which, together with episodic memory, is the other key feature of mental time travel.
Michael Corballis, a psychologist at the University of Auckland who had been a leading advocate of the view that mental time travel was a uniquely human ability, says that he has now “come round to Nicky’s view,” due to her “elegant experiments,” as well as more recent research on rats suggesting that they can replay in their brains the twists and turns of a maze.
But before Clayton left UC Davis to take up her permanent post at the University of Cambridge, another important event occurred, one that would profoundly affect both her personal and professional life: She met her future husband, Nathan Emery, a neuroscientist doing his post-doc, at a party. But it’s what he did after they met that mattered. Some guys send flowers. Emery sent her his paper about the role of eye gaze—the ability to look into one another’s eyes—in primate social life. “Nathan thought that primates were the bee’s knees because they could do this and that,” Clayton recalls. “So I sent him something that showed birds did it, too. Jays really look at you, and they recognize people.”
Clayton and Emery married, and later published a key study they call the “it takes a thief to know a thief paper,” about theory of mind. They found that if one scrub-jay was being watched by another while it was hiding food, it would re-hide it someplace else when the second jay was no longer around. But the only jays that “re-cached” food in this way were those that were already known to steal food from others. It seemed that a jay had to know its own scheming mind to realize that it needed to protect its food from other thieves. “The moment you are both the protector of your own caches and an active thief of others’, you have to see it from different points of view,” Clayton says.
While most animal cognition researchers agree that bird smarts were considerably underestimated before Clayton crashed onto the research scene, the field is more divided about what her findings mean, and in particular whether corvids—with brains the size of a walnut—really do have anything approaching human-like theory of mind. One of the sharpest critics has been Daniel Povinelli, an anthropologist at the University of Louisiana at Lafayette. Povinelli acknowledges that these birds are very smart, in some ways even smarter than chimpanzees and other primates, and that they are capable of a rudimentary form of abstract reasoning. But he insists that the evidence falls short of proving that they can actually understand what is in another bird’s mind. In Povinelli’s view, their seemingly sophisticated behavior can be explained more simply by what they actually observe in the real world than by an ability to understand how another bird thinks.
Thomas Suddendorf, a psychologist at the University of Queensland in Brisbane, Australia, says that to really understand what is going on in those bird brains, Clayton and other researchers need to publish more studies to provide evidence of both the birds’ cleverness and their intellectual boundaries. “It is much more exciting to learn that an animal can do something we thought it could not,” Suddendorf says. “But for a comprehensive picture, we need to also find out about their limits. Negative results can be very informative about the underlying processes driving behavior.”
Probably the most direct challenge to Clayton’s interpretations was mounted by Elske van der Vaart, a researcher now at the University of Reading in the United Kingdom. Van der Vaart worked in Clayton’s lab for six months while a Ph.D. student, and then published a series of computer simulations suggesting that at least some of what seemed to be theory of mind could be duplicated by “virtual birds”—that is, fictional birds whose behavior was generated entirely in a computer—that had nothing to go on but observations of the behavior of other birds. No mind reading was required. Soon after, though, Clayton’s group published its own paper countering that at least one of van der Vaart’s key assumptions about what the birds were doing was wrong.
Van der Vaart says that Clayton was “totally fine” with her attempts to come up with alternative explanations. “I never got the impression that Nicky has any doubts that scrub-jays are exactly as smart as they appear to be. But she definitely sees the value of models that challenge that hypothesis, if only because they sharpen our understanding of exactly how they are smart.
Although her experiments have convinced many researchers that corvids have something at least akin to theory of mind and mental time travel, Clayton says she is not just interested in whether they have it or not. What she cares about most is “how they do it, and whether they do it in the same ways that humans do.”
The brain of a corvid may be small, but Clayton points out that this does not tell the whole story about avian intelligence. Relative to body size, corvid brains are as large as those of chimpanzees and gorillas. Conversely, the brain of a pigeon, which has roughly the same body size as a typical corvid, is much smaller. Clayton and Emery, who is now at Queen Mary University of London, have argued in recent publications that the brains of birds and humans are much closer in the way they function than previously realized, even if the two animal groups went their separate evolutionary ways about 300 million years ago. Although the bird brain has a very different structure—in contrast to the human prefrontal cortex, birds do most of their “thinking” in a structure called the nidopallium caudale—Clayton says that “all of the patterns of connectivity and communication channels are similar. There’s only so many ways you can make a thinking machine.”
During the past few years Clayton and her graduate students and post-docs have taken two main approaches to comparing and contrasting how birds and humans think. Carrying on their work with Rome, Lisbon, and the gang, they have designed new experiments to figure out the limits of the birds’ ability to interpret their fellow corvids’ thoughts.
At the same time, Clayton’s team has begun working with kids ranging from about three to six years old, setting them up with tasks similar to those corvids can do successfully, to see at what age children develop these cognitive skills. In one set of experiments, similar in concept to those that Rome and Lisbon participated in, the team would ask one child to pack a lunch box for another kid of the same age after watching a video in which the second child was seen eating either crackers or apple slices. The idea was to see if the children would understand that the child in the video had already eaten plenty of one food and might prefer to have the other choice.
Another experiment resembled one Clayton performed years earlier with perishable wax worms and durable peanuts, except that in this case, the perishable item was a chocolate teddy bear and the durable one a hard tea biscuit. After the children were shown that the teddy bear, which was placed in a warm box, would melt over a short period of time—and told that they could not eat the chocolate if it was melted—they had to choose ahead of time which treat they would get after going away for either a shorter or a longer period.
So far these experiments, with designs similar to the food caching, planning ahead, and other situations used to test avian cognition, have shown that human children lag behind adult birds in the early toddler years, but that they are then able to perform as well as corvids when they get to be four or five years old.
The big question posed by the corvid and human experiments, Clayton says, is how these two different kinds of minds can produce such similar psychological functions. One way of looking at it, she suggests, is that the differences might be akin to those between Mac and PC computers—their operating systems are very different, but they end up doing a lot of the same things. While Clayton doesn’t speculate on which species is the Mac and which is the PC, she does hope her research will shed some light on a more fundamental question: “What does it mean to be a bird, and what does it mean to be a human?”