Nature’s Evolutionary Pulse: How AI Unlocked the Hidden History of Songbirds

For over a century, evolutionary biologists have operated under a foundational assumption: life does not progress at a steady, monotonous pace. Instead, theory has long suggested that the history of life on Earth is defined by "adaptive radiations"—explosive bursts of diversification where species rapidly evolve to occupy new ecological niches, separated by longer, slower periods of stasis. While the fossil record has hinted at this rhythmic pattern for decades, proving it across the vast, complex tree of life has remained an elusive goal.

Now, a groundbreaking study from the University of Michigan has provided the most compelling evidence to date for this "stop-and-go" model of evolution. By leveraging the power of artificial intelligence to analyze over 170,000 skeletal measurements from more than 2,000 species, researchers have reconstructed 45 million years of evolution in Passeriformes—the massive group of birds that includes most of the world’s songbirds. The findings, published in Nature Ecology & Evolution, reveal that these birds evolved in distinct pulses, often triggered by dramatic shifts in the Earth’s climate.

The Technological Leap: From Dust-Covered Bones to Digital Data

The scale of this research would have been impossible just a decade ago. To understand the evolutionary trajectory of songbirds, the University of Michigan team, led by postdoctoral fellow Jake Berv and associate professor Brian Weeks, needed to move beyond the study of single bones or isolated features. They needed a holistic view of the avian skeleton.

The catalyst for this project was "Skelevision," an innovative AI tool developed by the Weeks laboratory in collaboration with David Fouhey’s team at New York University. Skelevision acts as a high-throughput digitization engine. By photographing bird skeletons against a standardized grid, the AI can identify and measure 12 critical skeletal landmarks with surgical precision.

Over the course of a seven-year collaboration, the team processed more than 15,000 museum specimens, the vast majority of which were drawn from the U-M Museum of Zoology. Each specimen is scanned in a mere 45 seconds—a pace that transforms museum collections from static archives into dynamic, high-resolution datasets.

"It’s especially clear how important it is to invest in museums when you think about the scale of an analysis like this," says Brian Weeks. "It’s so far beyond the scope of what can be done using specimens contributed by an individual collector. It’s also fun to imagine what early collectors would make of how we’re using the specimens they collected—I imagine it would blow their minds to learn that a computer has analyzed a photograph of these specimens."

Chronology: A 45-Million-Year Evolutionary Map

To synthesize this massive influx of skeletal data, Jake Berv developed a novel statistical method dubbed "bifrost." Unlike traditional methods that treat individual bones as independent variables, bifrost analyzes the entire skeleton as an integrated, morphological unit. This allows researchers to track how body shapes evolved in response to environmental pressures over an immense 45-million-year timeline.

The Eocene-Oligocene Transition

The analysis highlighted a period of exceptionally rapid body-shape evolution occurring roughly 35 million years ago. This timing is significant, as it aligns perfectly with the Eocene-Oligocene transition—a geological epoch marked by intense global cooling and the formation of the Antarctic ice sheet. As the world grew colder, the ecological stage shifted, forcing songbirds to rapidly adapt their morphology to survive in new, harsher climates.

The 15-Million-Year Slowdown

Conversely, the data identified a distinct cluster of evolutionary slowdowns occurring approximately 15 million years ago. This cooling in the rate of change coincided with another major geological shift. According to the study, these periods represent times when lineages had successfully filled available ecological space and were temporarily "settled," awaiting the next major environmental disruption to spur further innovation.

Supporting Data: Geography as a Driver of Change

The research went beyond temporal analysis to examine how geography influences the pace of evolution today. By mapping their data against global bird distributions, the team discovered a clear correlation: birds residing in extreme latitudes—where seasonal temperature swings are most violent—tend to evolve at significantly faster rates than their counterparts living near the equator.

This geographic pattern mirrors the historical findings. It suggests that environmental variability, whether viewed through the lens of millions of years of climate change or across modern-day latitudinal gradients, acts as a primary "accelerator" for morphological evolution. When the environment changes, birds change with it; when the environment is stable, the pace of evolution plateaus.

Official Perspectives: Redefining Evolutionary Theory

The implications of this study reach deep into the heart of evolutionary biology. For decades, the "pulsed" model of evolution was a theoretical framework supported by fragmented evidence. Now, it has been empirically validated on a massive scale.

"This is really important for evolutionary theory because there’s a long history, going back 100 years, that predicts the emergence of new groups is often associated with an explosive burst of diversification," said Jake Berv. "This could be because of a new ecological opportunity, or it could be because a group dispersed to a new continent, resulting in dramatic accelerations in their rate of evolution."

Berv notes that the "bifrost" model successfully captures the sequence of changes necessary to explain the variation seen in modern birds. "The whole organism is an integrated, complex morphology, and each of the individual pieces is interrelated to every other part in the body," he explained.

Brian Weeks emphasizes that the study has fundamentally shifted his perception of how life evolves. "This pattern we found with rare, big increases in the rates of evolution and lots of small decreases is really consistent with a pattern where lineages are exploring new ecological space and changing rapidly to take advantage of that opportunity," Weeks said. "I hope our findings will inspire a new integration of rates of morphological change into other big areas of interest, things like the very well-known latitudinal gradients in biodiversity."

Implications: Preparing for a Changing Future

Perhaps the most pressing implication of this research is its relevance to the current climate crisis. As human activity drives rapid, unprecedented shifts in global temperatures, the question of how quickly species can adapt has become a matter of urgent scientific concern.

"Right now, we’re in this moment in human history where there’s dramatic global climate change, and we don’t know what’s going to happen over even a 10-year period, let alone over a 10-million-year period," Berv noted. "To have a chance of understanding the long-term impact of human activity on Earth, we have to study the relationship between events in Earth’s history and evolutionary transitions."

The study suggests that while evolution can be remarkably fast, it is also highly reactive to external forces. By understanding the "pulses" of the past, scientists may be better equipped to predict how biodiversity will respond to the rapid, human-induced environmental pressures of the future.

Furthermore, the study stands as a testament to the power of artificial intelligence in scientific discovery. By digitizing the past, AI is unlocking secrets that were previously buried in the drawers of museum cabinets. As the researchers continue to refine these methods, they hope to apply them to other branches of the tree of life, potentially rewriting the history of evolution one skeleton at a time.

This research, supported by the Michigan Institute for Data & AI in Society, the Natural Sciences and Engineering Research Council of Canada, the National Science Foundation, Schmidt Sciences, and the David and Lucile Packard Foundation, serves as a bridge between the deep history of our planet and the uncertain future of its inhabitants. It reminds us that while evolution is a slow, steady climb in the popular imagination, the reality is a vibrant, rhythmic dance—one that is as responsive to the climate as it is to the march of time itself.