A biologist rejects an idea he studied for years that quality birdsong is a sign of bird smarts
After some 20 years of theorizing, a scientist is publicly renouncing the “beautiful hypothesis” that male birds’ sexy songs could indicate the quality of their brains.
Behavioral ecologist Steve Nowicki of Duke University called birdsong “unreliable” as a clue for choosy females seeking a smart mate, in a paper published in the March 2018 Animal Behaviour. He will also soon publish another critique based on male songbirds that failed to score consistently on learning tests. And in what he calls a “public service announcement,” Nowicki summarized the negative results of those tests on January 4 at the annual meeting of the Society for Integrative and Comparative Biology in Tampa, Fla.
“This was a beautiful hypothesis that got beaten up by data,” he says.
Knowing that something about male singing matters to a female songbird, Nowicki and other researchers once proposed that the quality of singing might indicate a bird’s brainpower. The idea was that, because songbirds need to learn their songs, females could select males with the best brain development by selecting those singing the most precisely copied songs. A brainier male might be better at hunting baby food or spotting predators, thus helping more chicks to survive. Or braininess might signal an indirect benefit, such as contributing good genes to chicks.
The first evidence for the notion that birdsong indicates bird smarts came from Neeltje Boogert at the University of Exeter in England, whose research suggested female zebra finches preferred smarter males with more complex songs. But subsequent studies have found evidence both supporting and contradicting the theory. To try to settle the matter, Nowicki and collaborators hand-raised 19 male song sparrows in the lab, controlling which songs the little birds heard as examples to copy so that it was clear how well each youngster learned each song.
To judge the birds’ mental capacity separately from their song learning, the researchers administered five learning challenges, such as learning which colored container lids to flip open for food. We “found a hodgepodge” of results, Nowicki says. A bird might have done well on some tests and flubbed others, and the puzzle solving results still didn’t match the bird’s song learning ability.
“Maybe they weren't cognitively good at stuff because they lived their life in a cage,” Nowicki wondered. To resolve that question, he and colleagues turned to wild swamp sparrows in a Pennsylvania marsh.
It’s trickier to judge the precision of birds’ mimicry in the wild, where youngsters hear multiple adult songs to copy. So a coauthor had a computer sort recordings from the marsh into a few “typical” song types. The closer a newly musical sparrow’s song came to one of those forms, the higher the researchers ranked the bird’s learning. Researchers also put these new singers through the cognitive tests. But again, the results were a hodgepodge, Nowicki says, finally beating his old hypothesis about birdsong to pieces.
Songbirds may not have a general bird “IQ” in the sense of how people use the term, says Rindy Anderson of Florida Atlantic University in Davie, who has worked with Nowicki. Instead, bird intelligence may be more modular, she says, with the birds good at certain learning tasks, but lousy at others.
Even if male songbirds don’t betray their smarts through song, some scientists are testing a different group of birds, parrots, to see if females prefer smarter mates. The colorful birds may be the closest living relatives to songbirds on the genealogical tree.
Among flirting budgerigars, a kind of parrot, females preferred males that performed food-finding tricks that they had learned, researchers in China and the Netherlands reported in the January 11 Science. A parrot researcher who wasn’t involved in that study, Tim Wright at New Mexico State University in Las Cruces, is finishing up a paper on different way of testing whether budgerigars’ sex appeal is tied to their smarts. In spite of Nowicki’s warnings about unreliable song, Wright isn’t ready to give up on the basic notion that a male’s smarts matter to his mate.
S. Nowicki et al. Song is not a reliable signal of general cognitive ability in a songbird. Society for Comparative & Integrative Biology annual meeting, Tampa, Fla., January 4, 2019.
A.L. DuBois et al. Song is not a reliable signal of general cognitive ability in a songbird. Animal Behaviour. Vol. 137, March 2018, p. 205. doi: 10.1016/j.anbehav.2018.01.020.
J. Chen et al. Problem-solving males become more attractive to female budgerigars. Science. Vol. 363, January 11, 2019, p. 166. doi: 10.1126/science.aau8181.
R.A. Anderson et al. Song learning and cognitive ability are not consistently related in a songbird. Animal Cognition. Vol. 20, March 2017, p. 309. doi: 10.1007/s10071-016-1053-7.
S. Milius. Hearing awful or great singing changes birds’ choice. Science News Online, October 17, 2014.