In the world of animal research, there has long been a clear hierarchy of intelligence. Monkeys are at the top, naturally. Then come rats, the workhorses of the psychology lab. And down at the bottom are mice, cute and fluffy but not all that bright. For at least a hundred years researchers have used rats in their psychology experiments, assuming that they were the smarter of the two lab rodents. But new research shows that that might not be true.
Tony Zador and his lab at Cold Spring Harbor recently published a paper calling into question that hundred-year-old assumption. In their work, they show that mice can perform decision-making tasks in the lab just as well as rats can. “The paper that we had here, that basically solidified an intuition that we had developed over the last five years,” Zador told me, “which is that anything we could train a rat to do we could train a mouse to do as well.”
This finding isn’t trivial. Being able to use mice in experiments instead of rats could open up all kinds of new research options. For one thing, scientists have been able to manipulate a mouse’s genome in really useful ways, silencing certain genes to figure out what role they play. There are mouse models for everything from Alzheimer’s to Parkinson’s. Being able to put those mice through the paces of a psychology experiment could help researchers connect diseases with the behaviors they impact.
The size of mice is an advantage here too. Mice are cheaper and easier to store in the lab. And using certain techniques, their brains are easier to image. Being able to use mice in psychology experiments to figure out how the brain receives information and chooses what to do with it is huge.
So where did this idea that rats are smarter than mice come from, anyway?
Zador says it’s a historical bias. Psychology researchers started out using rats. And if you plop a mouse into an experiment designed for a rat, you might come to the same conclusion they did: that the mice just aren’t as smart. Mice in a rat-based experiment don't perform very well, and researchers took that to mean not that their experiments were designed poorly for mice, but that mice were simply dumb.
“There was 100 years of practice in training rats. And basically when people tried to treat the mice in exactly the way they treated the rats, the rats seemed smarter," Zador says. In other words, "over the course of 100 years people had figured out how to train rats, and that mice aren’t rats.”
In fact, it took Zador and his team years to figure out how to adapt their experiments to suit mice. Some of those changes were physical—mice are far smaller than rats, so they had to manipulate the setup to accommodate them. Other changes involved the training techniques. “There was no big epiphany, there was no aha! moment,” Zador says. “But it was just a series of small optimizations and we got mice to perform pretty much as well as the rats.”
As an evolutionary aside: You might think that mice and rats would be basically the same when it comes to these kinds of things, but Zador points out that mice and rats diverged somewhere between 12 and 24 million years ago. For comparison, humans and chimpanzees split somewhere between 5 and 7 million years ago. So it's no surprise that mice behave differently than rats, and that that difference impacts their training in the lab.
Now, after years of work, Zador and his team have figured out how to train mice just as well as rats. They can train a mouse in their setup in a few days, and over 90 percent of their little furry subjects successfully complete the training—a figure on par with rats.
Zador points out that this isn’t really about overall intelligence. There are things—like processing primate faces, for example—that monkeys are certainly better at than rats or mice. And primates are likely to outperform rodents on complex reasoning tasks. But when it comes to the kinds of research that many labs do about basic processing and decision making, rodents can do just as well as monkeys—and mice can do just as well as rats.
“I’m open to the possibility that there are ways in which rats are smarter than mice, but the things that we use in my lab don’t challenge that,” Zador says.
This exercise isn’t purely to prove people wrong (although that’s certainly nice too). Zador’s co-author Santiago Jaramillo, who was a post-doc when the study was done, has just started his own lab. He’s using mice as the basis of his experiments.
This is in some ways poetic. When Zador started his lab, he based his work on rats rather than monkeys. It was an unpopular decision. “The monkey people, they were pretty skeptical that we would be able to train rats to perform the kinds of sophisticated tasks that they were training monkeys to do, and that turned out just fine,” he told me.
Now, Jaramillo is setting off to do the same thing but with mice, and Zador hopes that their work will quell some of the skepticism he might face. “This paper is going to make people revisit their bias that mice are dumber,” Zador says. Jaramillo’s new lab might do the same.
We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.
