Pain, Pleasure, Fear, Disgust! Making Sense of the Faces that Mice Make

Machine learning shows that mice show characteristic expressions in response to emotional events, just as people do. Image reprinted with permission from Dolensek et al., Science. 2020 Apr 3;368(6486):89-94.

Published August 12, 2020

When a person feels disgust, fear or pain, there’s a good chance an observer can identify those emotions from the facial expressions of the individual they are watching. But what about animals like mice? Do they make recognizable facial expressions in response to stimuli – something bitter or sweet tasting, for example – that evoke strong emotions? And if they do, can humans make any sense of those expressions at all? A recent study says “yes,” on both counts.

Research led by Nadine Gogolla, Max Planck Institute of Neurobiology, Munich, Germany, now reveals that mice, just like people, make stereotypical facial expressions in response to emotional events. These expressions, which were indicative of disgust, pleasure, pain, malaise or fear, could be identified using machine learning, an area of study where a computer algorithm becomes “smarter” based on experience.

The investigators also used a technique that allowed them to visualize the activity of neurons in the brain in response to the different stimuli. With this approach, the group discovered that the activity of so-called “face neurons” in a specific region of the brain, called the insular cortex, was associated with the different facial expressions.

“This is a great study, because the authors did not only develop the algorithm that identified facial features in response to emotionally-linked stimuli, but rather went deeper to gain some mechanistic insight behind this process,” said Jeffrey Mogil, McGill University, Montreal, Canada, referring to some of the workings of the nervous system that underlie the researchers’ findings. Mogil was not involved with the recent work.

The research was published April 3, 2020, in the journal Science.

A mouse and its facial expressions
The path to the new study was not planned, according to Nejc Dolensek, a PhD student at Max Planck Institute of Neurobiology, Munich, Germany, and the first author of the study.

“I have to admit that it is a bit random how the story came about. I wanted to study the neuronal activity of different brain regions, but needed something to correlate the neuronal activity to, so I first looked at the response of the pupils of the eyes of a mouse to different stimuli,” Dolensek explained. “At some point I zoomed out of the image and noticed that the mice had responded with very slight but distinct facial expressions to the different stimuli.”

To test whether mice truly change their facial expressions in response to different stimuli that provoke strong emotions, as is the case in people, Dolensek and colleagues exposed the animals to a diverse set of stimuli, assumed to trigger changes in the emotional state of the animals similar to the changes seen in humans.

These stimuli included electric shocks to the tails of the animals, a sugar (sucrose) solution, a bitter-tasting quinine solution, lithium chloride, which causes malaise, as well as stimuli that evoke fear. The group monitored the facial expressions of the animals by using a video camera, both before and in response to these stimuli.

The authors found that the animals responded to the various stimuli with observable facial movements. But it was difficult for the investigators to interpret what those movements meant to the animals – what emotion underlay them.

So, to learn more, the scientists used a machine learning approach in which a computer algorithm could learn to distinguish facial features, by “training” the algorithm with images of the facial expressions of the mice.

With this, the team saw that their learning algorithm could classify the facial expressions into distinct emotional events resulting from the various stimuli. These included pain (in response to the electric shock) pleasure (from the sucrose), disgust (from the quinine), malaise (from the lithium chloride); and fear (indicated if the animals tried to escape or froze in place).

It turned out the algorithm could predict the underlying emotion in response to the stimuli solely from facial expressions across different mice – with more than 90% accuracy.

The group also showed that, the majority of the time, the facial expressions were specific to the emotion that the different stimuli caused. For example, facial expressions of pain in response to tail shock were different from facial expressions seen in response to the other stimuli.

Are the expressions of emotions real?
The researchers next asked whether the facial expressions were merely reflex reactions to the various stimuli or real expressions of emotions. To do this, the team tested features such as the intensity of the emotion, whether the animal experiences an emotion as good or bad, and the flexibility of the emotion – does the emotion change? – depending upon the animal’s internal state.

The results showed that the facial expressions indeed reflected true expressions of emotions. For instance, when the group varied the strength of a stimulus, by increasing the strength of the tail shocks or of the concentration of sucrose or quinine, for example, the facial expressions of the animals became more intense.

With regard to flexibility of the emotions, the authors manipulated the internal state of the animals without changing how strong the stimulus was and saw what effect that had on the facial expressions. Here, mice showed stronger expressions of pleasure in response to water or to sucrose when they were thirsty, compared to when they were not thirsty.

The researchers also wanted to test how learning might affect the facial expressions. So they exposed the mice to a sucrose solution and then injected the animals with lithium chloride, in order for the animals to learn to dislike the sugar solution instead of feeling pleasure from it.

Before this experimental manipulation, the animals’ facial expressions were indicative of pleasure in response to sucrose. But after the manipulation, when the animals now associated sucrose with the malaise that lithium chloride causes, the sugar now elicited facial expressions of disgust.

Looking into the nervous system
To dig deeper, the researchers used optogenetics, a method that uses light to change the activity of specific neurons in the brain. Here, they wanted to see if they could evoke different facial expressions by using light to activate neurons in an area of the brain called the ventral pallidum. This is a region responsible for the reward experienced from something pleasant.

“I was really surprised how well this experiment worked,” said Dolensek. “The light activation of the ventral pallidum caused the animals to exhibit pleasure-like facial-features similar to what we had seen when the animals consumed the sugar water.”

Next, the researchers applied the different stimuli while making video recordings of the facial expressions. They used this approach in combination with an imaging technique that allowed them to identify neurons that were activated by specific facial expressions.

This led to the discovery of “face neurons,” in a brain region called the insular cortex, whose activity reflected specific features of the facial expressions, including, for instance, how long the expressions lasted. They also saw that most face neurons responded to just one single emotion.

Why do mice do this at all?
Why do mice even bother to show facial expressions in the first place? The authors note that emotions are often considered particularly important for social interactions. But in this case, the mice were not socially interacting with other mice. So what gives? The researchers speculate that perhaps facial expressions set the animals on a path to taking action with movements and other behaviors.

In any event, the new research builds the case that when a mouse makes a facial expression, it reflects something meaningful about what the animal may be experiencing – yet another way in which rodents resemble human beings far more than we might appreciate.

Francie Moehring, PhD, is a freelance writer based in Milwaukee, US.

This story first appeared on the IASP Pain Research Forum and has been lightly edited for RELIEF.