University of Minnesota
At top: 
Rats may not slap themselves on the head in a regret-inducing situation, but other responses are very similar to humans'.

Rat regret informs decision research

Rats that passed up a short wait at one feeding station only to encounter a much longer wait at the next showed evidence of regret.

June 19, 2014

When his graduate student Adam Steiner walked in and announced, “My rats are expressing regret,” neuroscience professor David Redish responded along the lines of "You've got to be kidding."

Redish uses rats to probe fundamental mechanisms of decision-making, and sees no reason other animals’ brains shouldn’t resemble humans’. But the idea of regret in rats still came as a surprise, and it required solid evidence—which he and Steiner have now supplied.

On June 8 the researchers reported in Nature Neuroscience that rats can recognize when they’ve made a boneheaded decision and change their behavior in response. That, says Redish, is the essence of regret. And it means rats may act as stand-ins for humans as researchers probe how the brain makes decisions.

“When we understand how rats make decisions, it tells us something about how humans do it,” says Redish. “The more we understand about how decision-making processes work, the more we can understand about how they go wrong and how to fix them.

“Addiction is one example. Addicts always say they’re going to stop, and then they go smoke ‘one last cigarette,’ which they regret.”

“In a normal situation, you’d expect that if you do something you really don’t like, you won’t do it again,” says Steiner. “Addicts realize its’ bad, but maybe that signal of regret isn’t strong enough. Before we can understand what’s going wrong with the regret mechanism we have to understand how it goes right and exerts its beneficial effect. It’s easier to see how something’s broken if you know how it works.”

Although findings will be tested in humans, the researchers say rats are valuable for teasing out the mechanisms of regret. Lacking language, rats don’t complicate experiments by, for example, hedging or lying about what they’re experiencing,

Restaurant Row

The researchers placed the rats on “Restaurant Row”: a square runway with corner spokes, each leading to a “restaurant” offering a food morsel in one of four flavors. As the rats reached each spoke, a tone told the rat how long the wait was (higher pitch=longer wait).

Rats had individual flavor preferences and showed a “threshold” delay they were willing to accept for each flavor.

“These furry little guys know what they like. For instance, they generally didn’t like the chocolate-flavored pellets,” says Steiner. “They make decisions in many cases based on individual preferences, and their needs of the moment.”

Because trials only lasted an hour, the rats were under time pressure to take only “good deals”—i.e., to wait for short delays for preferred morsels—and pass up bad deals (long delays for less preferred morsels). Once a rat had eaten its food or passed it up and gone to the next spoke, it couldn’t go back for a do-over.

Redish and Steiner recorded patterns of neural activity from two areas of the brain associated with regret in humans to see if regret-like processes modified the rats’ decision-making as they faced “stay or go” food choices.

Let’s make a deal

The rats faced regret-inducing situations whenever they passed up a good deal only to encounter a bad deal at the next spoke. That was akin to refusing a wait of 20 minutes at your favorite restaurant, then driving 10 minutes to your second favorite and finding a 30-minute wait.

In that case, the rats turned and looked backward toward the spoke where they had made the mistake. And their brain activity signaled that they were, mentally, back at that bad decision point. The rats also rushed through eating the “bad deal” morsels and hurried to the next spoke. 

None of this happened when the rats were merely “disappointed” by a bad deal. In a disappointment situation, the bad deal followed either another bad deal or a good deal they had taken, giving the rats no cause to rethink a decision.

In other words, rats’ behavior revealed that they could distinguish regret from disappointment.

“In the regret situation, it turned around and looked back. But it didn’t do that in a disappointment situation,” Redish explains. “In disappointment, the world is just bad—it’s not their fault. But in regret, the rat made a mistake. We now know they recognize that difference.”

“It was fairly surprising,” says Steiner, “but in retrospect, it makes sense. Regret keeps you from making the same mistake over and over.”

Next stops on the research path

Having found that rats make suitable subjects for research on regret, Redish looks forward to investigating new questions that this research opens up:

• Do rats avoid regret?

• Why do we avoid regret?  

• How does regret change decision-making, and how does it interact with other mental phenomena such as deliberation and habit?

Redish, along with colleagues, is in the planning stages for a human version of Restaurant Row. Those experiments will examine how brain function correlates with various problems, or lack of problems, in decision-making ability.

The research was supported by the National Institute on Drug Abuse.

Contact the writer at

Twin Cities Campus: