What really makes the caged bird sing

The sweet music of a songbird is actually the product of powerful vocal cords and a sophisticated brain, says Tim Birkhead - and it could point the way forward for important neurological research

This weekend, the National Cage Bird Show will take place in Birmingham. Many consider this event an anachronism, an unpleasant reminder of the past, when small birds were more widely abused. But for the thousands of enthusiasts who will visit it, the exhibition is part of a valiant effort to maintain a historic tradition that reached its peak in the latter part of the 19th century. Since then, bird-keeping has been in decline because it is considered cruel, and recent television documentaries have portrayed bird fanciers as sad and stupid people – bird-brained, in fact. But as a professional zoologist, I would disagree.

Birds have fascinated people for a long time. In the past they were caught for two things: for food or for their song. Plump and tender birds like larks, wheatears and ortolans were eaten – usually by the wealthy – as they still are in Italy. Birds with a good voice, like the nightingale, were kept for their song. If they were good looking and sang half-decently, like the goldfinch, they were also kept – but only the males, for only males sing; females were eaten.

For aeons the nightingale was the songbird par excellence. It was plain in appearance, but its voice was unsurpassed. But after the canary first appeared as a cage bird on mainland Europe in the late 1400s, it nudged the nightingale off its premier perch. Nightingales were difficult to keep, and sang only for a few weeks each year, whereas the canary (which was equally dull to look at, since the wild bird is streaky green, not yellow) had a voice nearly as good, was easy to keep and would sing through much of the year.

The people who kept canaries and other songbirds were of three types: those who simply liked their sound – birdsong was like having Radio 1 on in the background; those who used their charges to enhance their own status through singing contests; and thirdly, those (a minority) who actually studied them.

Singing for status was an all-consuming passion, and fanciers often competed using wild-caught goldfinches and chaffinches. They also used the canary – but the canary was a different case because, as it was a costly import, great efforts were made to rear it in captivity, and once it was breeding, it could be artificially selected. The Germans led the way, and focused on producing superb songsters which, because of their rolling song, were to become known as roller canaries. From these Teutonic beginnings, all the 70 or so canary breeds that are now in existence developed.

The Germans also led the way when it came to science. The key player was the almost unheard-of Baron von Pernau (1660-1731), a wealthy politician whose observations of both wild and captive birds anticipated the work of Konrad Lorenz, the Nobel prize-winning zoologist, by almost two centuries. While English intellectuals were still writing banal descriptions of what birds looked like, Pernau was describing how they behaved, and using ingenious methods to draw remarkably prescient conclusions. He was especially interested in how birds acquire their song, and he was the first to recognise that songs weren't hard-wired in the brain, as most people believed, but that learning was the key to a bird's singing success.

Seventy years later in England, the Honourable Daines Barrington (the naturalist Gilbert White's correspondent) took these investigations a step further. Unaware of Pernau's work, Barrington claimed his own as the first scientific study of birdsong. His investigations took the form of two main activities. In 1773 he compiled a set of criteria to rank the quality of song of difference species. On the basis of "compass" and "sprightliness", the nightingale came top, with 90 out of a possible 100 points, followed by the linnet with 70 points: he didn't score the canary, but a later study put the canary in second place.

For the second part of Barrington's study, he asked the pioneer surgeon John Hunter to dissect some birds and to examine their vocal cords. The species with the best song was also found to have the best developed larynx. "Mr Hunter found the muscles of the larynx to be stronger in the nightingale, and in all those instances where he dissected both cock and hen that the same muscles were stronger in the cock," Barrington wrote.

Research completed in the last few years has found that those sexual differences in the larynx (now referred to as the syrinx – the songbirds' voice box) are also reflected in the brain. The region of a birds' brain responsible for song, the higher vocal centre (HVC) is up to four times bigger in males than in females. Moreover, because song-learning requires considerable brain power, the HVC is also bigger in those species which sing more complex songs. Female birds can manage with a smaller HVC, because processing male song and deciding who sings well or indifferently is less demanding than producing the song in the first place.

If they could see us today, Pernau and Barrington would be amazed to discover how the study of song-learning in little birds now provides a general model for how the brain – including our own – works. Through trying to understand how songbirds sing, Fernando Nottebohm at the Rockefeller University in New York has discovered something that has shaken the very foundations of brain research: the canary's brain is capable of regrowing neurones, once considered an impossibility. Nottebohm found that if he gave a female canary a shot of testosterone, she would start to sing like a male – precisely because her HVC had grown new neurones and increased in size.

The second thing that he found was that the seasonal occurrence of birdsong matched the seasonal growth and degeneration of the HVC. In both cases, the changes were almost unbelievable, because brains simply were not supposed to do that. The established view, based on studies of rats, monkeys and humans, was that we are born with all the neurones we will ever get, and if we lose any, say, through a brain injury, there are no replacements and we have to manage with what we have left. The take-home message from Nottebohm's research is that if we can find out how birds do it, it may be possible to persuade human brain cells to regrow – and this may ultimately provide a cure for degenerative brain diseases such as Parkinson's.

So if you go to the National Cage Bird exhibition in Birmingham next weekend, spare a thought for the birds that are singing away – and remember that neither they, nor the bird fanciers, are as daft as some people think.

Professor TR Birkhead is in the Department of Animal and Plant Sciences at the University of Sheffield