The telephone rings. A man hurries through his apartment, picks up the handset and says hello. Yet the ringing continues--because the sound came from the mans pet parrot. The owner shoots the bird a nasty glare as he hangs up, mutters about being fooled again and stalks out of the room.
Scenes like this, used in cartoons and comedies, are based on the fascinating ability of parrots to closely mimic common sounds and human voices. But some cognitive scientists who have spent years working with parrots are convinced that these birds, and others, are capable of much more. According to the experts, the animals can not only understand what we say and respond sensibly, but they can also grasp higher concepts such as "same or different" and the continued existence--or permanence--of objects that are shown and then hidden from view. The implication is that parrots and other avian groups, despite their simple-looking brains, may in some ways be as intelligent as primates and aquatic standouts such as dolphins.
More Than Imitation
For decades, biologists thought instinct alone controlled the behavior of parrots. This view was supported by characteristics of the birds' brains, which, outwardly, appear much simpler than those of mammals [see illustration on opposite page]. The animals' talent for producing speech was explained as pure imitation, devoid of any understanding of meaning. In the late 1970s Irene Pepperberg of Purdue University began a long series of experiments that she has continued to this day, now at Brandeis University. Her goal has been to determine, concretely, just how intelligent these birds are. The results are surprising.
On supporting science journalism
If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.
Pepperberg first had to solve a fundamental problem: How could she evaluate her feathered subjects? Their ability to mimic the human voice seemed to provide a solution. Human sounds are indeed foreign to the interparrot repertoire, yet it might be possible for the birds to use the sounds to communicate with humans deliberately.
Language acquisition by humans and song learning by birds have much in common. Both are based on the child or chick imitating the behavior of elders, and in both cases the learning requires many repetitions. Pepperberg adapted a training method developed by Dietmar Todt--the Model/Rival Technique--that mirrors the learning behavior of people and birds. A typical session might work in the following way: Two trainers sit on opposite sides of a small table in front of a parrot. Various objects are on the table. The first trainer picks up an item, shows it to the other person and asks her, "What is this?" The partner answers, "Ball. That is a ball." Trainer one then praises trainer two and hands her the ball as a reward. Sometimes, however, trainer two purposely answers incorrectly by saying, for example, "That is a clothespin." Trainer one scolds her and removes the object from view for a few minutes.
The parrot is also queried and is praised or chided according to its answers. If it voices the correct label, it gets to play with the object for a while. In this scheme, trainer two is both a model for the parrot and its rival for attention from trainer one. Once the parrot learns a few words, it can sometimes take over the role of trainer two to teach other birds.
After three decades of lessons, Pepperbergs first subject, a grey parrot named Alex, has acquired a respectable vocabulary and is still going strong (some parrot species can live 60 years). Alex knows labels for about 50 objects, seven colors, five shapes, seven materials, six numbers and even a few verbs.
Knowing What Different Means
A comprehensive vocabulary, however, is by itself no evidence for higher cognitive functions--just for a good memory. The bigger question is whether parrots understand what they are saying. To find out, Pepperberg has been subjecting Alex, and three other grey parrots that have learned speech in the lab, to numerous experiments. The results collected thus far are clear: the birds understand the meaning of individual words and show higher cognitive abilities.
Pepperbergs parrots, without exception, respond accurately to standard questions such as "What is this?" as well as to questions about an objects shape or color. Alex and his pals, however, have also garnered an understanding of conceptual categories such as color, shape and material. Such an insight is achieved only if the animal realizes that red, green and blue are possible variants of one category of object. Pepperbergs parrots do. Furthermore, they have learned the meaning of "same or different." Show a red triangle and a red circle and ask, "What is same?" and the parrot will respond, "Color." Ask, "What is different?" and the reply will be "Shape." These are remarkable achievements, because the bird must first interpret the question correctly, then identify the right category, and reply with an acoustic signal not used by other parrots. Alex also replies "none" if nothing is the same or different, demonstrating understanding of the concept of absence.
In addition, Alex seems to understand the notion of "and" as well as how to employ it. Given the question "What is rectangular and red?" he can pick the one object with both characteristics from a selection of items. His performance in comparing items is impressive. He can solve questions such as "What color bigger?" (which means "What color is the bigger object?"). The correct reaction makes it obvious that Alex understands the idea of relativity--among a group, which item is bigger.
Quantities are manageable, too. If Alex sees a display with four red balls and three green balls plus five red blocks and six green blocks arranged randomly, he can answer "How many red blocks?" correctly with "Five." Recent data suggest that Alex is counting the larger sets of objects one by one and recognizing the smaller numbers intuitively, at a glance, the way people can.
The grey parrots can even use verbs and combine them with different objects. For example, one will say, "Want apple" or "Want go chair." The trainers grant these wishes so that the birds experience the consequences of their utterances. Here, too, the parrots evidently understand what they say. If Pepperberg hands over the wrong object, the parrot will reject it by saying, "No," and then repeat the original request. If brought to the wrong place, the bird will refuse to hop off the trainers arm and will reiterate the demand.
Another example of abstract thinking is the so-called object permanence problem--the realization that an object continues to exist even when it changes position or can no longer be seen. This basic concept in understanding ones environment is not as straightforward as it may seem. Among humans, this ability develops only gradually during the first year of a babys life.
Adult dogs, cats and pigeons possess a rudimentary idea of object permanence, but it is much less powerful than for parrots, apes or humans. They fail, for example, in a shell game. After watching a trainer hide a ball under one of three cups and then move the cups around, the animals are no more successful than chance at indicating which cup still contains the ball. Some adult primates and parrots, however, do about as well as mature humans. Studies of young parrots reveal that object permanence develops in stages tied to the development of specific brain regions.
Give Me a Banerry
Over the years the researchers have been continually surprised by their subjects abilities. Alex, who has lived the longest in Pepperbergs lab, is the master. Several times he has invented useful new terms to describe objects--such as when he was supposed to learn "apple." At that time Alex had already learned the names of several edible fruits, including banana, cherry and grape. He had tasted apples, too, but had never been taught anything about them. One day the trainer picked up an apple and asked, "What is this?" Alex answered, "Banerry," and bit into the fruit. The trainer sought to correct him and repeated "apple" several times, but Alex insisted on "banerry," using the same careful diction the trainer always used when introducing a new word.
From then on, Alex stubbornly applied this term to every apple he saw. Pepperberg will never know for certain what happened inside Alexs brain when he coined the term, but she thinks he combined two words he knew--for banana and cherry. Perhaps apples taste something like bananas to him, and any fruit with red skin would look like a large cherry.
When Alex saw his reflection for the first time, he suddenly asked a trainer, "What color?" while pointing with his head toward the mirror. Alex had been asked this question, though only in reference to colored objects in the lab. The query was not just evidence for his understanding of the concept of color; Alex was transferring the question from a familiar setting to a new one. After the trainer overcame her initial surprise, she told him, "Grey. You are a grey parrot." Alex asked the same question five more times and got the same answer each time. From that day on, "grey" was part of his vocabulary.
These examples are clear evidence that parrots--and probably several other types of bird--are not merely bound by instinct. They possess splendid memories and understand complex relations. They can learn an intricate communication system, enjoy robust social lives and be impressively curious. Their intelligence may at least match that of primates and dolphins--which have heretofore been considered the smartest animals--and in some situations may surpass them. Pepperbergs parrots have already passed tests that are considerably more demanding than those completed by mammals.
The study of Alex and his friends is far from concluded. Right now the birds are learning to sound out letters and even to use a specially designed computer. Perhaps someday they will get their own home page or blog, and we will be able to chat with them online.