This post was originally published on here
All living things must survive in environments that are constantly changing. Seasons shift from summer to winter, and weather patterns can swing from floods one year to drought the next. Populations of plants and animals are always dealing with new pressures, explains University of Vermont scientist Csenge Petak. What remains unclear is how this ongoing instability shapes evolution over time.
Petak wondered whether frequent environmental changes actually help populations adapt by preparing them for future challenges, or whether constant disruption slows progress. “Do populations benefit from lots of environmental fluctuations, making new generations more prepared to face future changes,” she asked, “or are they impaired, forced to readapt again and again, never reaching the heights of fitness that the same populations in a stable environment could achieve?”
Simulating Evolution Across Generations
To investigate this question, Petak teamed up with University of Vermont computer scientist Lapo Frati, along with two other UVM researchers and a collaborator from the University of Cambridge. Together, they designed a groundbreaking study using advanced computer simulations that followed thousands of generations of digital organisms.
The findings, published December 15 in the Proceedings of the National Academy of Sciences (PNAS), challenged simple assumptions about evolution. “We found remarkable variation in how populations evolved in variable environments,” the researchers reported. “In some cases, changing the environment helped populations find higher fitness peaks; in others, it hindered them.”
Impossible to Test in a Lab
Traditional evolutionary research often tracks a single population living under one set of conditions. Frati explains that this narrow focus can miss important patterns. “Researchers often watch the long-term trajectory of one population in a specific environment, says Frati. “We picked an array of environments and see how the specifics of each one influence the trajectory of many populations.”
To see why this broader approach matters, consider fruit flies living in very different parts of the world. A population in the United States might experience seasonal temperature swings, while another in Kenya alternates between long dry spells and heavy rainfall. These groups belong to the same species, yet they face very different challenges.
“Temperature fluctuations might promote better adaptation to both cold and warm seasons,” Petak explains. “But repeated cycling between dry and wet seasons might actually impede adaptation to drought, forcing the population to ‘restart’ evolution after they experience a long period of rainfall — leading to worse traits than in populations exposed only to drought.” As a result, one population may benefit from environmental shifts while another is held back by them.
Why History Matters in Evolution
Senior author Melissa Pespeni, a biology professor at UVM, says the study’s scale made these insights possible. “What’s exciting about this study is that we replayed evolution hundreds of times. This gave us a bird’s-eye view of how evolution played out across many different environments — something that would be impossible to test in the lab,” she said.
One major conclusion stood out. “The biggest takeaway for me is that starting point really matters. A population’s history shapes how high it can climb and how hard the path is to get there, which means we can’t assume one population represents an entire species.”
Why These Findings Matter Now
The results have important implications for real-world problems. Scientists need to know whether plants and animals can adapt quickly enough to survive accelerating climate change. At the same time, bacteria continually evolve resistance to antibiotics, posing a growing threat to human health.
Despite this complexity, research often focuses on just one population under one type of environmental stress. Broad conclusions are then drawn about how a species will respond to change. Petak argues that this approach can be misleading. “Computational models, like ours, can be used to formulate new hypotheses about real biological populations,” she says.
Testing Evolution in 105 Different Environments
In their simulations, the researchers created artificial organisms and exposed them to a wide range of shifting conditions. These digital environments reflected natural patterns such as temperature cycles and alternating periods of drought and rainfall.
“What is new in our work,” Petak explains, “is that instead of studying evolution in just one variable environment, we created 105 different variable environments. This allowed us to systematically compare how populations evolve across many distinct scenarios.”
AI Implications
The findings also extend beyond biology and may help inform research in artificial intelligence. Many AI systems struggle to learn new tasks without losing skills they already acquired. Co-author and UVM computer scientist Nick Cheney sees strong parallels between this challenge and evolutionary dynamics.
“AI systems have traditionally been built narrowly around solving one specific question,” Cheney says. Newer approaches aim to build systems that keep learning over time. A growing field known as online continual learning, he adds, “beautifully mirror the ideas explored in this paper around how evolution, learning, and development engage with — and benefit from — variable and dynamic environments.”
Learning How to Learn
For Frati, the broader message applies to learning systems of all kinds. “My research is about meta-learning, the capability of systems to learn to learn.” he says. Just as an AI cannot be evaluated based on a single task, evolution cannot be fully understood by studying one environment alone.
The study highlights the importance of testing systems across many comparable but distinct conditions when evaluating evolvability, which Frati describes as the ability of a system to evolve to evolve.
At its core, the research shows that evolution is shaped not just by change itself, but by the order, type, and history of those changes. As Petak puts it, “Our results show that the choice of variable environment,” she says, “can strongly influence the outcome.”
