Male Birds' Ability To Learn Song Affects Female Mating Response

According to the scientists, the finding offers broader insight into the role that traits learned by males play in sexual success.

In an article in the September 22, 2002, Proceedings of the Royal Society of London: Biological Sciences (now online), biologists led by Duke University Professor of Biology Stephen Nowicki reported studies in which they tested the mating response of female song sparrows to songs of captive-raised males. Importantly, the scientists had analyzed the males' songs in detail to determine the degree of accuracy with which the males copied songs they attempted to learn. They found that the females preferred those songs that came closest to wild-type songs they heard when young and presumably learned as models. The scientists' research was sponsored by the National Science Foundation.

According to Nowicki, he and his colleagues in the field have long theorized that female songbirds pay attention to male song as an indicator of fitness.

"We've developed experimental evidence that there is a link between early stress, male brain development and song-learning," he said.

"But until now, experimental and field observations showing that females were interested in song only contrasted the presence or absence of song, or relatively gross features of song, like the size of the repertoire. This is the first study to explicitly demonstrate that females care about song-learning quality," he said.

To test the effects of fine differences in song quality on female response, the researchers trained captive-reared male song sparrows to sing by exposing them to the recorded songs of wild birds. To induce variation in stress among the birds, some were placed on a restricted diet during development. Using spectrographic analysis, the researchers rated the captive-reared birds on two measures of song quality

-- how much of the wild-bird song they copied versus how much they invented, a practice common among song sparrows. Those birds who did invent more song elements also tended not to copy well those elements they did copy.

-- how close the males had come to actually matching just the wild-bird song elements they were attempting to copy

To determine the effects of song quality, the researchers exposed wild-caught adult females -- presumably experienced in listening to male songs -- to the captive-reared males' songs. The scientists measured female response to the songs by the amount they performed characteristic and distinctive female mating presentation display -- which includes a shivering of the wings, the lifting of the tail and a characteristic call.

As a control, the scientists exposed the wild-caught females to what the scientists had judged as well-learned male songs, as well as the digitally recorded wild songs. The female birds responded equally to both.

However, when the scientists exposed the females to the captive-reared males' songs, they found the females responded more strongly to male songs that had been better learned by both of the scientists' measures.

"The females showed a strong preference for songs that had been copied well, as opposed to songs that had been copied poorly," said Nowicki. "And by our measures, the males got points taken off for originality. That seems to make sense because we would argue that males that deviate from original song haven't learned the song as well." In addition to insight into bird song, said Nowicki, such studies can give basic insight into the evolution of animal signals in general.

"We know sexual selection is a very powerful evolutionary force that has led to phenomena such as the evolution of extravagant displays and the evolution of size differences between sexes. I believe that this work demonstrates that sexual selection might not be acting directly on the obvious trait that is expressed, but on the mechanisms that underlie the expression of that trait. In the case of bird song, a male's song reflects the birds' developmental history, and song expression is only the trait that the female can gain access to for information about -- in this case -- brain mechanisms."

Also, said Nowicki, the discovery that females assess song quality emphasizes the importance of studying the neurobiology of song expression and placing it in an evolutionary context.

While the current studies show clearly that females prefer well-learned songs, among the next research steps, said Nowicki, will be to determine how females learn to judge song quality.

"There is only very thin evidence that females learn song, so it's a major scientific question whether females are learning something about the population that they're living in, and using that as a way of assessing males," he said. Such female studies also will reveal whether the female's ability to distinguish good songs from bad reflects the birds' fitness and influences evolution, said Nowicki.