Alex the African Grey

That might be stretching things but, as an expert on animal cognition discusses, certain birds display startling abilities and intelligence


Irene Pepperberg is associate research professor at Brandeis University and the author of a new book, Alex and Me. She and Jonah Lehrer, the editor of Mind Matters, discuss what Alex and other African Grey Parrots can teach us about the evolution of intelligence and the concept of zero.

LEHRER: What first got you interesting in study avian intelligence? After all, to say someone has a “bird brain” is insulting.

PEPPERBERG: I had parakeets as pets as a child, and I knew they were quite smart. For instance, they could learn to say words and phrases in context. But I didn’t connect that to science at the time. I trained in chemistry at MIT and chemical physics at Harvard, not even knowing that a new field, animal cognition, was developing in psychology. It wasn’t until I saw the first NOVA programs, in 1974, on ape signing, dolphin intelligence and the one on “Why Do Birds Sing?” that I realized that one could look at animal-human communication and animal intelligence in a scientific way. That’s when I realized that no one was looking at parrots, which could actually talk. I decided to use their ability to produce human speech sounds to examine their cognitive processes.

LEHRER: Were you surprised by Alex’s talents?

PEPPERBERG: In general, no. But occasionally he would do something that was really impressive, jumping beyond the task at hand, transferring his knowledge unexpectedly from one domain to another. That’s when I’d get surprised.

LEHRER: What do you think was Alex’s most impressive cognitive feat?

PEPPERBERG: The work on the “zero-like” concept. He had shown that he could label the number of a subset of items in a heterogeneous mixture (for example, tell us the number of blue blocks in a mixture of red and blue balls and red and blue blocks), but we hadn’t tested his comprehension of number. That task was important, because young children, at a particular stage in number learning, can label a set but can’t, for example, remove a specific number of marbles from a big heap.


So we were testing him on number comprehension, again showing him heterogeneous mixtures of different numbers of objects of different colors (for instance, two blue keys, five purple keys, six green keys and asking, “What color is six?”). As was his wont, he was at about 90 percent accuracy on the first dozen or so trials, but we needed far more for statistical significance. The problem was that he just did not want to comply. He began to turn his back to us, throw the objects on the floor, or give us all the wrong answers and repeat the wrong answers so that, statistically, we knew he was avoiding the correct response. We started bribing him with candies and treats to get him to work. One day, in the midst of this, I’m testing him with a tray of three, four and six blocks of different colors, and I ask, “What color three?” He replies, “Five.” At first, I was puzzled: there was no set of five on the tray. We repeat this interaction several times, and he consistently says, “Five.” Finally, in frustration, I ask, “OK, what color five?” He says “none”! Not only had he transferred the use of “none” from a same-different task, where “none” was the response if nothing about two objects was indeed “same” or “different,” to the absence of a numerical set, but he had also figured out how to manipulate me into asking him the question he wanted to answer!

LEHRER: What can bird intelligence teach us about the evolution of human intelligence? Birds and primates parted ways a long time ago.

PEPPERBERG: Yes, primates and birds separated about 280 million years ago. But Alex’s abilities show us that it’s important to examine parallel evolution and to be willing to examine how a brain functions, not only how it looks. The cortical-like area of the parrot brain looks nothing like human cortex, but it is derived from the same pallial areas as is human cortex, functions in a similar manner and takes up roughly the same proportion of space. We also must examine the conditions that likely selected for intelligence in evolution. Grey parrots, for example, like nonhuman primates, are long-lived and exist in a complex ecological and social environment. Likely the same conditions that selected for intelligence in nonhuman primates were at work in the parrot lineage.

LEHRER: In your book, you describe repeated examples of scientists and journals ignoring and discounting your results. Why do you think people are so resistant to the idea of bird intelligence? And have things improved?

PEPPERBERG: When I started my research, very few scientists studied any bird other than the pigeon, and used any technique other than operant conditioning. Pigeons did not perform very well compared to other animals (such as rats and nonhuman primates), and were thus considered to be lacking in intelligence; scientists extrapolated their findings to all birds. At the time, scientists didn’t understand how the avian brain functioned, and thought it lacked any significant cortex. And, of course, when I began my research, some scientists started discounting much that had been done in the field of human-animal communication. So, when I started working with a parrot, and chose to use a nontraditional training method, few in the scientific community would give credit to Alex’s achievements.


Whether or not things have improved depends a lot upon whom you ask. Many scientists do appreciate what Alex did and have been inspired to further investigate the abilities of all birds—not only parrots and corvids, but also to perform new research with pigeons.

Other scientists, intent on proving the uniqueness of humans, tend to discount my research. Much of the work in avian cognition has shifted to Europe now, with large grants going to researchers in the U.K. (St. Andrews, Cambridge, Oxford) and other countries in the E.U. (such as Austria). Unfortunately, very little funding is available here in the U.S.

“Alex & Me,” Irene Pepperberg’s memoir of her 30-year scientific collaboration with an African gray parrot, was written for the legions of Alex’s fans, the (probably) millions whose lives he and she touched with their groundbreaking work on nonhuman communication. Alex — for anyone who missed the commemorations last year in The Economist, Nature, The New York Times and on the radio and TV — could label more than a hundred objects and understood the concept of categories, as well as bigger-smaller, same-different and absence. (The Guardian called Alex “smarter than the average U.S. president.”) To anyone who’s dreamed of talking with the animals, Dr. Doolittle style, Alex was a revelation.


For a technical analysis of his feats, you’ll want to read Pepperberg’s book “The Alex Studies,” published in 2000. The present book, in contrast, is largely celebratory — light on science, heavy on cute animal stories and heartwarming in its depiction of what was either a fruitful professional collaboration or a weirdly dependent friendship, or both. Still, it isn’t all billing and cooing: a strain of “I’ll show them” runs through the text. Accusations against the scientific establishment, which at first denied Pepperberg funding, publication, prestigious appointments and professional respect, propel the narrative.

Pepperberg’s basic biography ought to be rousing (nerdy girl abandons career in chemistry to pursue animal intelligence; is rejected by establishment; achieves international acclaim). But she tells her own story with far less emotion than she does Alex’s. There’s much that Pepperberg is unwilling to reveal — about her cold and controlling parents, her failed marriage and her relationships with colleagues. That’s the author’s prerogative, but it leaves a reader wondering how she ended up in her 50s, alone and jobless, reduced to eating 14 tofu meals a week (to save money, not the earth). Her approach to herself is neither scientific nor humanistic: the woman remains an enigma.

Alex, on the other hand, is a delight — a one-pound, three-dimensional force of nature. Mischievous and cocky, he also gets bored and frustrated. (And who wouldn’t, when asked to repeat tasks 60 times to ensure statistical significance?) He shouts out correct answers when his colleagues (other birds) fail to produce them. If Pepperberg inadvertently greets another bird first in the morning, Alex sulks all day and refuses to cooperate. He demands food, toys, showers, a transfer to his gym.
This ornery reviewer tried to resist Alex’s charms on principle (the principle that says any author who keeps telling us how remarkable her subject is cannot possibly be right). But his achievements got the better of me. During one training session, Alex repeatedly asked for a nut, a request that Pepperberg refused (work comes first). Finally, Alex looked at her and said, slowly, “Want a nut. Nnn . . . uh . . . tuh.”

“I was stunned,” Pepperberg writes. “It was as if he were saying, ‘Hey, stupid, do I have to spell it out for you?’ ” Alex had leaped from phonemes to sound out a complete word — a major leap in cognitive processing. Perching near a harried accountant, Alex asks over and over if she wants a nut, wants corn, wants water. Frustrated by the noes, he asks, “Well, what do you want?” Mimicry? Maybe. Still, it made me laugh.
After performing major surgery on Alex, a doctor hands him, wrapped in a towel, to an overwrought Pepperberg. Alex “opened an eye, blinked, and said in a tremulous voice, ‘Wanna go back.’ ” It’s a phrase Alex routinely used to mean “I’m done with this, take me back to my cage.” The scene is both wrenching — Alex had been near death — and creepy, evoking the talking bundle in “Eraserhead.”

Pepperberg frames her story with Alex’s death: the sudden shock of it, and the emotional abyss into which she fell. Ever the scientist, she wonders why she felt so strongly. The answer she comes up with is both simple — her friend was dead — and complex. At long last, and buoyed by the outpouring of support from people around the world, she could express the emotions she’d kept in check for 30 years, the better to convince the scientific establishment that she was a serious researcher generating valid and groundbreaking data (some had called her claims about animal minds “vacuous”). When Alex died, that weight lifted.

BUT in revealing how Alex lived, and the day-to-day workings of her lab, Pepperberg may soon find herself open to fresh criticism. Her book raises an important question: why, if Alex has the cognitive skills of a young child, and even seems to grasp such concepts as love, would anyone confine him to a cage in a lab? Why run him through the drills, or scold him for getting answers “wrong”? (“You turkey,” he’d say, mimicking his trainers, or “Say better!”) During a stint at M.I.T.’s Media Lab, Pepperberg worked on a device, designed for gray parrot owners, that projects terrifying images of predators when their pets’ vocalizations “exceeded the desired level.” She doesn’t comment on the morality of either confining a highly intelligent creatureor scaring it into submission. She deals with the question of animal rights in just one sentence: while acknowledging it would be cruel to adopt a gray and leave it alone all day, “that doesn’t mean grays or other animals have wide-ranging political rights.”
Alex was a celebrity, and this book will surely please his legions of fans. Meanwhile, supporters of Pepperberg, who continues her research with other grays, will remind critics that we’d have no inkling of parrot intelligence without Alex’s sacrifice.
Elizabeth Royte, a frequent contributor to the Book Review, is the author of “Garbage Land: On the Secret Trail of Trash” and “Bottlemania: How Water Went on Sale and Why We Bought It.”


By ELIZABETH ROYTE (NYT, November 9, 2008)


THE last time Irene Pepperberg saw Alex she said goodnight as usual. “You be good,” said Alex. “I love you.” “I love you, too.” “You’ll be in tomorrow?” “Yes, I’ll be in tomorrow.” But Alex (his name supposedly an acronym of Avian Learning Experiment) died in his cage that night, bringing to an end a life spent learning complex tasks that, it had been originally thought, only primates could master.

In science as in most fields of endeavour, it is important to have the right tool for the job. Early studies of linguistic ability in apes concluded it was virtually non-existent. But researchers had made the elementary error of trying to teach their anthropoid subjects to speak. Chimpanzee vocal cords are simply not up to this—and it was not until someone had the idea of teaching chimps sign language that any progress was made.

Even then, the researchers remained human-centric. Their assumption was that chimps might be able to understand and use human sign language because they are humanity’s nearest living relatives. It took a brilliant insight to turn this human-centricity on its head and look at the capabilities of a species only distantly related to humanity, but which can, nevertheless, speak the words people speak: a parrot.

The insight in question came to Dr Pepperberg, then a 28-year-old theoretical chemist, in 1977. To follow it up, she bought a one-year-old African Grey parrot at random from a pet shop. Thus began one of the best-known double acts in the field of animal-behaviour science.

Dr Pepperberg and Alex last shared a common ancestor more than 300m years ago. But Alex, unlike any chimpanzee (with whom Dr Pepperberg’s most recent common ancestor lived a mere 4m years ago), learned to speak words easily. The question was, was Alex merely parroting Dr Pepperberg? Or would that pejorative term have to be redefined? Do parrots actually understand what they are saying?
Bird brained
Dr Pepperberg’s reason for suspecting that they might—and thus her second reason for picking a parrot—was that in the mid-1970s evolutionary explanations for behaviour were coming back into vogue.

A British researcher called Nicholas Humphrey had proposed that intelligence evolves in response to the social environment rather than the natural one. The more complex the society an animal lives in, the more wits it needs to prosper.
The reason why primates are intelligent, according to Dr Humphrey, is that they generally live in groups. And, just as group living promotes intelligence, so intelligence allows larger groups to function, providing a spur for the evolution of yet more intelligence. If Dr Humphrey is right, only social animals can be intelligent—and so far he has been borne out.

Flocks of, say, starlings or herds of wildebeest do not count as real societies. They are just protective agglomerations in which individuals do not have complex social relations with each other.

But parrots such as Alex live in societies in the wild, in the way that monkeys and apes do, and thus Dr Pepperberg reasoned, Alex might have evolved advanced cognitive abilities. Also like primates, parrots live long enough to make the time-consuming process of learning worthwhile.

Combined with his ability to speak (or at least “vocalise”) words, Alex looked a promising experimental subject.
And so it proved. Using a training technique now employed on children with learning difficulties, in which two adults handle and discuss an object, sometimes making deliberate mistakes, Dr Pepperberg and her collaborators at the University of Arizona began teaching Alex how to describe things, how to make his desires known and even how to ask questions.

By the end, said Dr Pepperberg, Alex had the intelligence of a five-year-old child and had not reached his full potential. He had a vocabulary of 150 words. He knew the names of 50 objects and could, in addition, describe their colours, shapes and the materials they were made from.
He could answer questions about objects’ properties, even when he had not seen that particular combination of properties before. He could ask for things—and would reject a proffered item and ask again if it was not what he wanted. He understood, and could discuss, the concepts of “bigger”, “smaller”, “same” and “different”. And he could count up to six, including the number zero (and was grappling with the concept of “seven” when he died). He even knew when and how to apologise if he annoyed Dr Pepperberg or her collaborators.
And the fact that there were a lot of collaborators, even strangers, involved in the project was crucial. Researchers in this area live in perpetual fear of the “Clever Hans” effect. This is named after a horse that seemed to count, but was actually reacting to unconscious cues from his trainer. Alex would talk to and perform for anyone, not just Dr Pepperberg.
There are still a few researchers who think Alex’s skills were the result of rote learning rather than abstract thought. Alex, though, convinced most in the field that birds as well as mammals can evolve complex and sophisticated cognition, and communicate the results to others. A shame, then, that he is now, in the words of Monty Python, an ex-parrot.

Sep 20th 2007
From The Economist print edition