Sound research

How do we learn to talk?

Robert J. Dooling has spent 30 years looking for answers in an unlikely place: the ears of birds.


Dooling, a University of Maryland psychologist, has tested the hearing of dozens of avians. He has tried to decipher their songs into language and piped their calls through tiny speakers in an effort to probe their intelligence.

One of his graduate students has even put tiny headphones on parakeets to see how their neural circuits work when they get mixed audio signals.


"If you want to study speech in humans, birds make ideal subjects," Dooling said.

And in his latest work, Dooling has used his knowledge of the bird's inner ear to develop a theory about the hearing range of its ancient ancestor.

"Birds are descendants of dinosaurs, and their ears are scale models of each other," said Dooling, who will present his findings next month at a conference in Salt Lake City.

The structure of the dinosaur's inner ear, he adds, indicates that its hearing is a bit like that of an elephant, which can detect low-pitched sounds but has a harder time with higher frequencies, Dooling says.

Dooling, one of a handful of avian experts around the world who chart the hearing ranges of birds, rewards them with food when they learn to peck at bars in response to a song or chirp. Over the years, he and colleagues have documented the hearing of up to 50 species. "We study how their auditory system guides the sounds they make," Dooling said.

In his lab, cages filled with brightly colored parakeets, zebra finches and canaries line two rooms. Researchers shuttle the birds into cages in soundproof booths, where tiny speakers emit recorded bird songs. Microphones pick up the live birds' responses and television monitors capture every detail.

Now, about those headphones: Doctoral candidate Michael Osmanski slipped tiny headphones over birds' ears to see how they respond when they're trained to produce one sound, but hear something different as they sing.

If a human tries to recite a phrase and hears a recorded version delayed by a quarter-second, he will begin to stutter or get confused. Osmanski found that birds do, too.


Dooling's unique approach makes him something of a legend in avian academic circles. "In a way, you can think of him as a national treasure because he runs the only lab in the country interested in things like that," said Richard Fay, a psychologist who studies the auditory systems of fish at Loyola University Chicago.

Each of the three avian species Dooling studies has a different acoustic profile. For example, zebra finches learn their songs within two months of hatching and keep them throughout their lives. But parakeets adjust their songs to fit in with any new flock that comes along.

"Parakeets are more plastic throughout their lives. In terms of what they can learn, they're like us," he said.

Canaries, meanwhile, are seasonal songbirds. Male canaries vocalize more in spring and summer when their hormone levels are up and they're trying to attract a mate.

Dooling has been working with birds since his days as a graduate student at St. Louis University in the 1970s, when he started to investigate the effects of excessive noise levels on humans. In those days, researchers were still trying to measure how much noise exposure led to human hearing loss.

As part of that effort, Dooling tested excessive noise on chinchillas, mammals with hearing abilities that are about the same as humans. That done, he wanted to learn how humans and other animals learn to make the sounds they produce - and the role that hearing plays in the process. For that he needed a new animal.


"I began to think, what if I began to work with something with a really different ear," he said.

He turned to birds because he liked them. As a boy, he raised homing pigeons in his parents' garage in Omaha, Neb., fascinated by their ability to return home - even when a neighborhood traveling salesman took them on trips and released them hundreds of miles away.

"I was maybe 8 or 9 years old, and I got a big kick out of it," he said.

Although he has kept birds as pets in the past, these days he has a dog and two cats in his Darnestown home.

Dooling, 61, completed his postdoctoral work with birds at Rockefeller University and signed on with the University of Maryland, College Park in 1981.

He is now co-director of the Center for Comparative and Evolutionary Biology of Hearing, an international group of psychologists, biologists, engineers, and hearing and speech experts that receives National Institutes of Health funding.


"All the work is aimed at understanding the human hearing system and how it works," Dooling said.

Birds make good subjects because when it comes to producing sounds, many birds, like humans, have to learn the sounds they eventually produce.

Birds that are deaf at birth will produce far different sounds than birds that can hear. And birds that go deaf as adults will lose the ability to reproduce the so-called "template" sounds they developed in the nest.

Within a few months of hatching, sparrows and finches are at the peak of their ability to learn new sounds, but the window "sort of shuts" as they mature, Dooling said.

The pattern doesn't hold for all birds: A parrot can learn new words and phrases throughout much of its adult life, he said.

Humans can have better low- and high-frequency hearing than birds, Dooling said, "and we're as good or better at the middle-pitch ranges."


Dooling and other researchers have found that the size of a structure in a bird's inner ear, known as the basilar papilla, correlates with the range of its hearing. The smaller the structure, the better the bird hears in high frequencies; the larger it is, the more limited its hearing is to low-frequency sounds.

Dooling and a pair of German colleagues have also published findings showing that because birds are related to dinosaurs, the same relationship between the inner ear and hearing ranges likely holds true for at least three types of dinosaur: the brachiosaurus, the allosaurus and the archeoptryx.

From analyzing fossilized dinosaur remains, Dooling, Otto Gleich of the University of Regensburg and Geoffrey Manley of the Technical University of Munich concluded that larger dinosaurs, such as brachiosaurus and allosaurus, heard best at low frequencies and could not have heard sounds above 3,000 hertz, or cycles per second.

A parakeet, by contrast, hears best between 2,000 and 4,000 Hz, while humans have a much wider hearing range. A healthy youth can hear frequencies of up to 20,000 Hz while most middle-aged adults have an upper limit of about 12,000 Hz.

Animals, who use sounds to attract mates, escape predators and defend territory, generally produce sounds in ranges their own species can hear. "If they use vocalization for communications, they hear the sounds they produce," Dooling said.

The fossil records show dinosaurs were good at hearing at least one important sound in terms of survival: the footsteps of other dinosaurs.


Dooling will discuss his findings June 5 at a meeting of the Acoustical Society of America. His research, published in the German journal Natural Sciences, could help scientists unravel mysteries about the world inhabited by dinosaurs before they went extinct 65 million years ago, experts say.

"It will tell you the kind of things in the environment they responded to, how they interacted with their environment and things about the other animals around them," said Neil Todd, a neuroscientist who studies crocodile sounds at the University of Manchester in England.

What we hear

Sound travels in waves, and the number of vibrations per second determines its frequency.

Frequency is measured in hertz (Hz). A low-frequency sound, such as a heartbeat (1-2 Hz), will have a low pitch, while a high-frequency sound, such as a dog whistle (above 20,000 Hz), will have a high pitch.


The fundamental frequency of an adult male voice is about 150 to 180 Hz, while a woman's voice is pitched at 200 to 300 Hz.

Some of the sounds we make, such as the consonant "s" sound, the "th" sound and the "ch" sound, have high energy behind them and can reach up to 6,000 Hz.

Most of a parakeet's vocalizations fall within the same range it can hear best: 2,000 to 4,000 Hz.

Source: Robert Dooling

More information

To learn more about bird sounds, visit Robert J. Dooling's Web site at