Tuning in to songbirds and their songs.

Tuning in to songbirds and their songs

Next to humans, songbirds have perhapsthe most varied language repertoire of any animal. Recent studies of their brains and behavior are revealing singing secrets that may help scientists understand how birds--and humans--learn and use the melodies they make.

In the last decade, scientists havelinked the size of certain regions of a bird's brain with its ability to sing. For example, one brain region in male canaries appears to grow during breeding season, when songs are used to attract mates and stake out territories from other males.

Recently, Sarah W. Bottjer at the Universityof Southern California in Los Angeles and her colleagues demonstrated that in the course of learning their species' song, baby male zebra finches show growth in one brain region while another region is diminished. Specifically, the caudal nucleus of the ventral hyperstriatum (HVc) increases its number of neurons by 50 percent during the 70-day maturing period; the magnocellular nucleus of the anterior neostriatum (MAN) loses half of its cells. According to Bottjer, this is the first demonstration in any animal species that one brain region grows at what appears to be the expense of another.

The loss of neurons in the MAN suggeststo Bottjer that zebra finches are born with a wide capacity for possible notes and that later, once they've learned the species' songs, they discard the cells for notes they no longer need. This idea, which the researchers are now testing, is supported by other scientists' findings that baby zebra finches raised by different species learn the other species' songs and ignore zebra finch songs later in life. In addition, adult zebra finches appear incapable of learning new songs, according to Bottjer.

This trait is somewhat similar to thebehavior of humans, in that the human capacity to learn languages diminishes considerably after puberty. Because of such similarities, Bottjer says she would like to examine human brains at postmortem to see if there is any evidence that the region involved in human vocal development gets smaller as children reach puberty.

There is evidence, however, that earlyin the 70-day maturing period, the MAN "is important for vocal learning,' Bottjer says. She and her co-workers have found that when they damaged the MAN early in a zebra finch's development, its later vocal repertoire was diminished and the sounds it made were abnormal. But MAN lesions in older juveniles and adult birds had no effect.

"This suggests to us that there may besome [very early] function carried out in the MAN region,' she says, "such as taking in auditory information or programming motor information with respect to vocal behavior.' It appears, she adds, that HVc may be taking control of vocal behavior as the bird ages.

In the March JOURNAL OF NEUROBIOLOGY,Bottjer and her co-workers also report that early in a zebra finch's vocal development, MAN neurons don't accumulate the male hormone testosterone very effectively. Toward the end of development, however, when the MAN region appears no longer needed for song learning, its nerve cells are full of testosterone.

"We see this as a very intriguing finding,'says Bottjer. One possibility is that testosterone "causes certain neural circuits to become hardwired in the brain. And once those circuits are wired in, they seem to lose their capacity for forming new kinds of behavior.' In particular, she posits that the increase of testosterone directs the MAN to send certain signals to another section of the brain, which permanently stores the bird's songs.

Another approach to learning aboutbird songs is to monitor the singing behavior of birds in the field. In the February THE CONDOR and in upcoming issues of ANIMAL BEHAVIOR, Stephen I. Rothstein at the University of California at Santa Barbara and Robert C. Fleischer at the University of Hawaii in Honolulu report on the "flight whistles'--songs that flying male birds use to communicate with males and females over long distances--of brown-headed cowbirds in the eastern Sierra Nevada.

The researchers found that groups ofcowbirds separated by several kilometers have very distinct flight-whistle dialects. For example, the flight whistle of one group near Mammoth Lake in California contains three syllables, or continuous sounds, while a group to the south has a flight whistle made up of four syllables, three of which are identical to the whistle of its northern neighbors. Rothstein says there are dozens, if not hundreds, of dialects within a 300-kilometer band.

"Even though these flight-whistle dialectsare only now being described,' he says, "we feel that they're one of the most clear-cut examples of dialects in song-birds.' Unlike other kinds of songs, flight whistles are simple and short, so researchers have little difficulty identifying different dialects.

"The big question in dialects, aboutwhich there's been a lot of controversy, is how they are maintained--what is making all the birds within one population converge to one type of vocalization,' says Rothstein. He and Fleischer propose a new theory called "honest convergence,' in which female cowbirds judge the suitability of a male caller by its whistle; males who do not know the proper dialect are either newcomers or yearlings, who are too young to mate.

Rothstein thinks the honest convergencetheory is better than the three existing hypotheses because it is the only one that adequately explains how a large population of birds can maintain one dialect while still having a large amount of gene flow and exchange between dialect groups.