AI Unravels the Evolution of Birdsong: A Symphony of Age, Movement, and Memory

The changing chorus: How movements and memories influence birdsong evolution

The researchers spent three years collecting over twenty thousand hours of sound recordings from a wild population of great tits (Parus major) in Oxfordshire, which has been studied for the past 77 years as part of the Wytham Great Tit study. Their aim was to investigate how the movement, age, and turnover of birds within a population influences the diversity and evolution of their songs -- including which songs become locally popular, which fade away, and how varied their song repertoires become. To achieve this, they used a new approach involving training an AI model to recognise individual birds based on their songs alone and measure song differences between individuals. This method allowed them to track variations in song repertoires across the population and uncover patterns in song evolution. The results showed that birds of similar age tend to have more similar repertoires, with mixed-age neighbourhoods having higher cultural diversity. Furthermore, the pace of song turnover within neighbourhoods is driven by individuals coming and going; when birds leave or die, many song types disappear with them and the young birds that replace them can speed up the adoption of new song types. At the same time age serves as a brake on change, as older birds continue to sing song types that are becoming less frequent in the population. In this way, older birds can function as 'cultural repositories' of older song types that younger birds may not know, just as grandparents might remember songs that today's teenagers have never heard. However, age is not the sole factor influencing song change. The study also found that when birds mix more -- through increased local dispersal and the arrival of immigrants -- they tend to adopt more common songs, which also slows the pace of song evolution. Furthermore, 'homegrown' songs tend to stay unique: Areas where birds stay close to their birthplace maintain more diverse and unique song cultures, similar to how isolated human communities often develop distinct dialects or musical styles. The results also indicated that newcomers tend to adapt but also enrich song diversity. Immigrant birds arriving from elsewhere typically adopt local songs rather than introducing entirely new tunes, however they tend to learn more songs overall, enriching the local 'musical scene.' Lead researcher Dr Nilo Merino Recalde (Department of Biology, University of Oxford) said: "Just as human communities develop distinct dialects and musical traditions, some birds also have local song cultures that evolve over time. Our study shows exactly how population dynamics -- the comings and goings of individual birds -- affect this cultural learning process, influencing both song diversity and the pace of change." The study is the first extensive test of the role of demography in driving cultural diversity and evolution at small scales in a wild animal population, using individual-level data and a very large dataset on song variation. This research not only provides insights into bird behaviour but also offers valuable perspectives on how demographic changes might affect cultural evolution across animal species -- with potential implications for conservation efforts. The complete dataset has been made publicly available for other researchers to explore. Professor Ben Sheldon (Department of Biology, University of Oxford), who leads the long-term bird study in Wytham Woods, commented: "Our work here shows, once again, that tracking individuals over their lives allows us to understand so much of the way that different processes interact in natural populations. It's thrilling to think that we can explain the acoustic landscape we hear in the woods each spring in terms of the result of the cumulative combination of individual movements and survival over many years."

Age and migration influence bird groups' song repertoires, study finds

Researchers used 20,000 hours of recordings of great tits in Oxford to see how culture changes among populations Which songs birds sing can - as with human music - be influenced by age, social interactions and migration, researchers have found. Not all birds learn songs, but among those that do, individuals, neighbourhoods and populations can produce different collections of tunes, akin to different music albums. Now researchers have found that changes in the makeup of a group of birds can influence factors including which songs they learn, how similar those songs are to each other and how quickly songs are replaced. Dr Nilo Merino Recalde, the first author of the study, from the University of Oxford, said: "This is very interesting, I think, partly because it shows that there are all these kind of common elements at play when it comes to shaping learned traits, [similar to] what happens with human languages and human music." But he said the parallels had their limits. "The function and the role of human music and language is very, very different to the function of birdsong," he said. "Birdsong is used to repel rivals, to protect territories, to entice mates, this kind of thing. And that also shapes songs." Writing in the journal Current Biology, Recalde and colleagues describe how they used physical tracking as well as artificial intelligence to match recorded songs to individual male great tits living in Wytham Woods in Oxford. In total, the study encompassed 20,000 hours of sound recordings and more than 100,000 songs, captured over three years. The researchers used their AI models to analyse the repertoires of individual birds, those within neighbourhoods and across the entire population to explore how similar the various songs were. As a result, the team were able to unpick how population turnover, immigration and age structure influenced the songs. The team found that within a neighbourhood of great tits, birds of a similar age tended to sing a similar selection of songs. Birds of a more advanced age tended to belt out older songs that were less common among the population as a whole - similar to how older people might hum more traditional tunes. Recalde said this was because great tits learn their songs in the first year of life. As a result, neighbourhoods with a greater mix of ages showed a greater diversity of songs. Movement also mattered. The team found new arrivals to a neighbourhood adopted local songs rather than introducing new ones - something Recalde said was down to dispersals typically happening when the birds were very young and hence still learning songs. But he added: "In the process, something interesting happens which we don't fully understand, which is that they seem to end up with a slightly larger repertoire." The researchers said neighbourhoods where birds tended to stay put had more unique "homegrown" tunes, possibly because they were less influenced by the songs sung elsewhere. By contrast, Recalde noted, greater movement meant repertoires became more similar because birds were more likely to sing "popular" tunes common throughout the population. Recalde said that as well as shedding light on how culture changed among the birds, the research raised new possibilities for conservation."[It] might let you say something about populations and their status and their structure and how well they're doing based on the songs themselves, without having to catch the birds," he said.

The changing chorus: How movements and memories influence birdsong evolution

New research from the University of Oxford has provided fresh insights into how bird songs evolve over time, revealing a significant role for population dynamics in shaping song diversity and change. The findings -- based on an analysis of over 100,000 bird songs -- have been published in the journal Current Biology. The researchers spent three years collecting over 20,000 hours of sound recordings from a wild population of great tits (Parus major) in Oxfordshire, which has been studied for the past 77 years as part of the Wytham Great Tit study. Their aim was to investigate how the movement, age, and turnover of birds within a population influences the diversity and evolution of their songs -- including which songs become locally popular, which fade away, and how varied their song repertoires become. To achieve this, they used a new approach involving training an AI model to recognize individual birds based on their songs alone and measure song differences between individuals. This method allowed them to track variations in song repertoires across the population and uncover patterns in song evolution. The results showed that birds of similar age tend to have more similar repertoires, with mixed-age neighborhoods having higher cultural diversity. Furthermore, the pace of song turnover within neighborhoods is driven by individuals coming and going; when birds leave or die, many song types disappear with them and the young birds that replace them can speed up the adoption of new song types. At the same time, age serves as a brake on change, as older birds continue to sing song types that are becoming less frequent in the population. In this way, older birds can function as 'cultural repositories' of older song types that younger birds may not know, just as grandparents might remember songs that today's teenagers have never heard. However, age is not the sole factor influencing song change. The study also found that when birds mix more -- through increased local dispersal and the arrival of immigrants -- they tend to adopt more common songs, which also slows the pace of song evolution. Furthermore, 'homegrown' songs tend to stay unique: Areas where birds stay close to their birthplace maintain more diverse and unique song cultures, similar to how isolated human communities often develop distinct dialects or musical styles. The results also indicated that newcomers tend to adapt but also enrich song diversity. Immigrant birds arriving from elsewhere typically adopt local songs rather than introducing entirely new tunes; however, they tend to learn more songs overall, enriching the local 'musical scene.' Lead researcher Dr. Nilo Merino Recalde (Department of Biology, University of Oxford) said, "Just as human communities develop distinct dialects and musical traditions, some birds also have local song cultures that evolve over time. Our study shows exactly how population dynamics -- the comings and goings of individual birds -- affect this cultural learning process, influencing both song diversity and the pace of change." The study is the first extensive test of the role of demography in driving cultural diversity and evolution at small scales in a wild animal population, using individual-level data and a very large dataset on song variation. This research not only provides insights into bird behavior but also offers valuable perspectives on how demographic changes might affect cultural evolution across animal species -- with potential implications for conservation efforts. The complete dataset has been made publicly available for other researchers to explore. Professor Ben Sheldon (Department of Biology, University of Oxford), who leads the long-term bird study in Wytham Woods, commented, "Our work here shows, once again, that tracking individuals over their lives allows us to understand so much of the way that different processes interact in natural populations. "It's thrilling to think that we can explain the acoustic landscape we hear in the woods each spring in terms of the result of the cumulative combination of individual movements and survival over many years."