For decades, scientists have thought that animals, including humans, use the best possible strategy to complete tasks like finding food or navigating a maze.
These “perfect” strategies are often complex and resource-intensive.
However, new research reveals that animals often use simpler, “good enough” strategies that get the job done without using too much brainpower.
A team of scientists at Janelia Research Campus explored different ways animals might solve a problem, going beyond the idea of a single perfect strategy.
Their research, published in Science Advances, shows there are many ways an animal can successfully complete a task, even if these ways are not perfect.
The researchers found that there are countless ways an animal can perform a simple task, like choosing between two options where the chance of getting a reward changes over time.
Instead of always using the best strategy, animals can use various other methods that work almost as well but require much less effort.
“As soon as you let go of the idea of being perfect, you find many ways to solve a problem,” says Tzuhsuan Ma, a postdoctoral researcher who led the study.
The team developed a theoretical framework to understand these different strategies. This framework helps explain why different animals might use different strategies, how these strategies might work together, and how they can be applied to other tasks.
This new understanding can help researchers study animal behavior in the real world.
Three years ago, Ma began exploring the different ways an animal could tackle a simple task. The researchers focused on strategies that were not the best but still effective. They looked at “small programs” that guide an animal’s actions based on past experiences.
Initially, the team found that the top-performing strategies were very similar to the optimal one.
This was disappointing because they were hoping to find different, yet effective, strategies. However, their persistence paid off when they discovered around 4,000 “good enough” strategies that were different from the optimal one.
To figure out which strategy an animal might be using, the team created a network to show the relationships between different strategies.
They developed a mathematical way to describe these relationships and an algorithm to show how one strategy could evolve into another. They found that small changes to an optimal strategy could lead to significant changes in behavior while still maintaining effectiveness.
This discovery suggests that animals, as generalists, might use these “good enough” strategies to solve various problems. This approach is different from the idea of animals being specialists that use one perfect strategy for a single task.
The research provides a new way to think about animal behavior, moving beyond the idea of a single optimal strategy.
The team is now focused on examining how these small programs can be applied to other tasks and designing experiments to determine which strategy an animal uses in real-time. They are also working with other researchers to test their framework.
“Understanding an animal’s behavior is essential for figuring out how the brain solves different problems,” says Ann Hermundstad, a group leader at Janelia. “Animals might be using very different strategies than we initially assume, and this work is helping us uncover that space of possibilities.”
This new perspective on “good enough” strategies not only sheds light on animal behavior but also opens up new avenues for understanding how the brain works and how it can solve problems efficiently.
Source: KSR.
