It has long been known that mice can be trained to perform simple tasks in exchange for a reward. By giving a mouthful of food to a hungry mouse or a drop of water to a thirsty mouse, you can entice the mouse to move through a maze or click a certain button. . However, in some cases, your mouse may not behave as expected and you may not be able to complete the task at hand. Often, researchers have dismissed these behaviors as simple mistakes resulting from carelessness or carelessness. still, the study Published in a magazine on April 26th current biology Even when rats understand the rules of a task, they can deviate from their behavior, potentially testing their own hypotheses and attempting to learn more about their environment.
“These mice have richer internal lives than we realize. These are not just stimulus-response machines. They may have something like a strategy.”
The decisions mice make during behavioral tests appear to be more complex than just the basic choice for reward. During human-imposed trials in the laboratory, mice may be continually exploring and retesting the rules of their environment and performing experiments. Own A small experiment.
[ Related: Mice may be able to recognize their own reflections ]
The discovery expands our understanding of what’s going on inside rodent brains and suggests that mice and other nonverbal animals may know more than we realize. is showing. This research could also ultimately help uncover the neurological basis of human behavior. “These mice have richer internal lives than we thought,” says Dr. Kishore Kuchibhotla, senior study author and assistant professor of neuroscience at Johns Hopkins University. “They’re not just stimulus-response machines. They may have something like a strategy,” he added.
Mouse holding the steering wheel
The work will be built on things like Previous research This tested mice with a simple licking task and added a level of complexity with a two-choice test to analyze mouse motivation.Kuchibhotla and lead author of the study Zu ZiyiThe neuroscience PhD student restrained thirsty mice in place and trained them to turn a wheel in a specific direction with their front legs in response to sound. One tone corresponds to turning the wheel to the right, a second tone corresponds to turning the wheel to the left. If the mouse responds to one of the sounds with the correct behavior, it will receive a small cup of water. If the wheel spins in the wrong direction or doesn’t spin at all, nothing happens.
Through thousands of experiments with 13 mice, the researchers tracked the mice’s choices, reaction speed, and accuracy and noticed several patterns. First, the mice seemed to become more accurate in their decision-making as the trials progressed, indicating that they were mastering the task at hand. Individual mice also seem to have quirks and preferences when it comes to selecting wheel direction. And even if a mouse reaches an expert level of wheel steering ability, it will still exhibit erroneous responses for short periods of time, repeatedly spinning the wheel in the same direction regardless of which sound is played. It happened often.
[ Related: How video game tech, AI, and computer vision help decode animal pain and behavior ]
To better understand what was happening during these matches, Kuchibhotla and Zhu conducted a “survey” trial in which they temporarily stopped rewarding the mice for correct answers. Very quickly, the mice changed course, stopped exploring, and began responding more accurately to left and right sound cues, following a trained pattern. This shows that the mice understood what to do to get the cup, ingested large amounts of water, and were intentionally forgoing the reward.
“As behavioral neuroscientists working with animal models, we have a responsibility to come up with smarter and richer ways to extract meaning from nonverbal animals.” [actions]” Dr. Brian Swais He is a neuroscientist and psychiatrist at Mount Sinai who studies animal behavior but was not involved in the new study. “I think this paper did a really good job in that regard…It was a beautiful deep dive into behavioral analysis,” he added, noting in particular the follow-up of the initial trial data and the ways in which the researchers changed the experiment. .
“Animals may seem like they’re making a lot of mistakes, but while they’re making those mistakes, they’re actually getting smarter.”
Zhu and Kuchibhotla used computational models to understand how the results of each trial relate to those before and after it, and what factors might be influencing the mice’s behavior. We evaluated whether it would be possible. They found that reward played a big role, but so did the bias to rotate the wheel in the preferred direction, which varied from mouse to mouse. However, this bias was not corrected. Over the course of many trials, the mice flipped a switch and rotated to either side, and when researchers presented the mice with audio prompts indicating only their preferred direction, the mice indicated longer times to rotate the wheel. Their undesirable aspects. Taken together, these observations point to a dynamic selection bias that researchers hypothesize is a learning strategy.
learn without language
“Mice surprisingly use higher-order approaches to learn even simple tasks that seem non-adaptive. Although it may seem like this animal is making a lot of mistakes, , while making those mistakes, you’re actually getting smarter,” Kuchibhotla says. “We put animals in these strange situations. We don’t know when the environment will change. They don’t know when we will change the rules. This kind of continuous exploration… has value.”
Whereas humans can rely on language to understand tasks, non-verbal animals have to find out for themselves what the rules are for a particular situation. Kuchibhotla suggests that this difference may explain why mice take such a constantly changing approach to a task. “Verbal or written instructions disrupt the mental search space. Once you know what to do, you don’t need to explore. This is one of the hypotheses we have. If there is no human [also] Engage in continuous exploration. ” He is now following up with human behavioral tests to determine whether that is true.
Finding lessons from mouse mistakes
Other follow-up work includes tracking neural activity in mice engaged in wheel-spinning tasks, training and testing mice on multiple tasks at once to see how strategies change, and ultimately These include having cognitively impaired mice perform similar tests that may reveal underlying patterns. Even in human neurological diseases like Alzheimer’s disease.
There are limits to what this single study can prove. Despite all the caution researchers are taking, Sweis suggests that something simpler than strategy may still be at play. For example, the mouse may have changed the direction in which it rotates its wheel from time to time because its front legs got tired. “I don’t think that negates anything. [the study authors are] That’s what’s shown here, but physical factors may be a factor,” Switzer says. “We need to understand Brian in his whole-body context.”
Still, he explains, examining the process of choice can “give us insight into how the brain works” and reveal what’s going on when things go awry. . He suggests that another follow-up project could investigate how aging affects exploration processes and whether task flexibility changes or decreases with age. There are many possible ramifications to this interesting study, which “serves as a rich foundation for understanding biology a little bit more.”
This is a reconstruction of behavioral outcomes in rodents from decades in which errors were dismissed as uninteresting failures. “Animals need to make mistakes in order to learn,” says Swais. And there is much we can learn from them.
