Mus musculus mice self-organize existing behaviors into structured bouts as they become expert hunters.
This study demonstrates that lab mice, without food restriction, learn to forage for hidden resources in a complex, dynamic environment by reorganizing their existing behaviors into structured, high-quality bouts of sampling and site-checking, thereby establishing a novel approach for investigating the neural and evolutionary basis of naturalistic decision-making.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer
In the wild, a mouse is a master of multitasking. It must navigate a shifting landscape, remember where food was found, listen for danger, and decide when to move or stay still, all while balancing its own hunger and safety. Scientists have long wanted to understand how the brain handles this kind of complex, real-world decision-making. For decades, the standard way to study animal behavior in a lab has been to train them to perform the same simple action hundreds of times in a row, like pressing a lever for a treat. While this method has taught us a great deal, it strips away the messy, self-paced nature of life in the wild. Researchers have been searching for a way to watch animals make choices in a dynamic environment that feels more like the real world, without forcing them into a rigid, repetitive routine.
A team of scientists at the Janelia Research Campus has now created such a setting, revealing how mice learn to hunt without ever being forced to do so. They built a large, hexagonal arena made of 157 modular tiles, creating a maze-like space where mice could roam freely. Hidden within this arena were special tiles that could release a live cricket, a natural food source for the mice. However, the mice could not see the crickets, nor could they smell them from a distance. The only clue was a sound. A speaker hidden inside the correct tile would play a noise, but only if a mouse sat perfectly still for half a second. If the mouse kept moving, the sound remained silent. To find the food, the mouse had to stop, listen, and then decide whether to investigate that spot or move on.
At first, the mice were clumsy. They wandered around, checking many different tiles, often sitting still and hearing a sound, only to find nothing there. But as they gained experience over several days, something remarkable happened. They became efficient hunters, finding the crickets much faster and with far fewer mistakes. The researchers expected that the mice might simply learn to sit still more often or check more tiles to increase their chances of finding the sound. They watched the mice closely, tracking their every move with high-speed cameras and using computer programs to analyze their body positions. They looked for an increase in specific actions, like standing up on their hind legs to listen better or grooming themselves less so they could focus.
The data showed that the mice did not actually do these specific actions more often. They did not stand up more frequently, nor did they check more tiles per hour as they became experts. The frequency of these individual behaviors remained roughly the same whether the mouse was a beginner or a master. Instead, the mice changed how they organized their existing behaviors. They began to chain together moments of standing still and checking tiles into tight, focused sequences. The researchers called these sequences "engaged bouts." In these short bursts of time, a mouse would stand up, listen, check a tile, move a few steps, stand up again, and listen once more, all without stopping to groom or eat. These bouts were like a focused search pattern that the mice switched on only when they were close to solving the puzzle.
This shift in strategy was the key to their success. The mice did not need to learn new tricks or work harder; they simply learned to group their old, familiar actions into a more effective pattern. When the researchers removed the sound cues, the mice immediately lost their efficiency, proving that the sound was essential. When the mice were food-deprived, it did not significantly affect their success metrics, showing that their performance was robust even without caloric restriction. The study demonstrates that learning in a complex, natural setting is not always about doing more of the same thing. It is about rearranging what you already know how to do, weaving your existing skills into a new structure that fits the challenge at hand.
The researchers designed this experiment to be flexible and open to future study. They used a system that could track the mice and control the environment in real time, allowing them to record brain activity or test specific brain circuits later on. This setup offers a new window into how the brain handles the kind of fluid, self-directed decision-making that happens in nature. By watching mice solve a puzzle in a space that feels like the wild, scientists can now ask deeper questions about the neural machinery behind learning. The mice showed that they could adapt to a changing world not by becoming different animals, but by becoming better at organizing the behaviors they already possessed.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.