It’s a complicated question. We still don’t know exactly why humans dream, or why dreams might be important. Studying animal dreams is even harder; dogs can’t tell us what made them whine or run during a snooze.
Depending on how one defines them, animal dreams could have intriguing implications. (Read more about the secrets of sleep.)
“I think dreaming gives us a way of extending a number of cognitive capacities to animals; that includes things like emotion, memory, and even imagination,” says David M. Peña-Guzmán, who studies the philosophy of science at San Francisco State University and recently authored When Animals Dream: The Hidden World of Animal Consciousness.
We know primates have emotions, but consider spiders, which a recent study suggests may experience REM-like sleep and even visual dreams. The thought of spider dreams sounds outlandish, but it may be true.
“We have this idea of dreams being a confabulatory narrative with kind of crazy, vivid elements to it,” says Matthew Wilson, a neurobiologist at MIT. “But when we look into animal models we’re simply trying to understand what goes on during sleep that might influence learning, memory, and behavior.”
What cats dream about
Domestic cats were some of the first animals subjected to dream research. Michel Jouvet, a pioneer of sleep studies, uncovered evidence of feline dreaming in the 1960s when he observed cats’ behavior while they slept and then altered it dramatically.
In REM sleep, human muscles don’t move much despite the intense mental activity that powers our dreams. This state of atonia ensures the body doesn’t act out our dreams no matter how real they seem. Jouvet learned that, in cats, a brainstem structure called the pons seemed to regulate REM sleep and produce partial paralysis. (Read how scientists are trying to see our dreams.)
By removing parts of the pons, however, Jouvet caused a dramatic change in behavior. With their brains deep in REM sleep the cats began to move as if awake, hunting, jumping, grooming and aggressively defending themselves against invisible threats.
Jouvet called this period paradoxical sleep, when the body is still but the mind remains fully active. This provided a window into what was happening in the cats’ sleeping brains.
“The cats performed behaviors that are very easy to interpret as mapping onto a waking experience,” says Peña-Guzmán.
Rats recall maze memories
After rats run a maze during the day, they can rerun the same course while asleep, research has shown. When awake, a rat’s hippocampus, a part of the brain responsible for making and storing memories, remembers the neuron pattern of navigating the maze. Later when asleep, the brain reproduces the identical pattern, suggesting the rat remembers or relearns the maze all over again.
This 2001 finding was one of the first to suggest that animals had complex dreams. And it was just the beginning, says co-author Wilson, the MIT neurobiologist.
“We’ve done other studies that suggest the ways memories of past experience get brought up again in sleep might be similar to what we would experience as dreams."
Those rat brain studies show that when maze memories occur during sleep, the visual imagery that went along with them is also reactivated, meaning the sleeping rodents saw what they had seen in the maze while awake. The same has been found for auditory and even emotional areas that reactivate when the rat reruns the maze during REM sleep. (Learn why rats avoid harming other rats.)
“A lot of things suggest there is an extensive re-experiencing of the awake state during sleep,” Wilson says. “If we want to call that dreams, I’m perfectly comfortable with that description. The interesting part is, if that’s what’s going on, what does it mean?”
Zebra finches remember songs
Despite being known for their lyrical songs, zebra finches aren’t born singers. The birds must learn by listening, practicing, and, perhaps, by dreaming.
In 2000 researchers learned that neurons in the birds’ forebrains fire with a distinct pattern while they sing a song—one that scientists can recreate note by note. While the birds sleep, their brain reproduces this same pattern, replicating the song they heard and sang that day, suggesting the birds remember and practice songs in their sleep.
The study authors suspect the songbirds dream of singing. Do dreaming birds relive their waking experiences? Or are singing dreams more like algorithms running without conscious awareness? Scientists may be getting closer to finding out.
After further research over two decades, finches were the first non-mammals found to have a similar sleep structure as humans, including REM sleep. More recent work shows that the birds also move their vocal muscles to match the music in their brains, and can be prompted to actually sing a song played to them in their sleep.
Sleeping finches also produce variations on their songs, suggesting that they gather sensory information while awake and create adaptive changes by improvising new versions to promote learning in a dreamlike state.
Sleeping deeply with the fishes
Zebrafish also experience REM-like sleep, according to Stanford neurobiologist Philippe Mourrain. While sleeping, these fish lose muscle tone, develop arrhythmic heartbeats, and show brain activity that looks like that of an awake fish. One notable difference from humans, though not all other animals, was that the fish didn’t move their eyes. (Nor, lacking eyelids, did they close them.)
The finding suggests that REM sleep, the state when most dreams occur, may have evolved at least 450 million years ago—before land and aquatic animals diverged in their evolution.
“Twenty years ago, people would tell me fish don’t even sleep,” says Mourrain. “Now we see… those behavioral features are conserved from insects to spiders and vertebrates. And in REM sleep, you lose control of your most vital regulatory systems. Evolution would not have conserved such a fragile state if it didn’t matter.”
But why does dreaming matter? Does REM sleep conservation across evolution mean that even fish might dream? (Read how REM sleep may serve as "overnight therapy" in people.)
That depends on your definition of dreaming. For Mourrain, dreaming is best explained as simply the shuffling of synapses, or in other words, a reset of the neural connections that prepares our nervous system for the coming day through processes like memory consolidation and cognition optimization.
“I would not be surprised if actual dreams were found in animals, and I think eventually we’ll be able to show that scientifically,” he says.
“You did something in the daytime, and your brain will replay it, integrate it, and mix it with other experiences. We’re not the only species capable of remembering and learning.”