What Pets and Animals Actually Dream About: What Brain Science and REM Sleep Research Reveals

In 2001, neuroscientists Matthew Wilson and Kenway Louie at MIT wired electrodes into the hippocampi of rats and let them run a maze during the day. At night, while the rats slept, Wilson and Louie watched the same neurons fire, in the same sequence, at the same relative timing, as they had during the run. The correlation was strong enough that the researchers could tell, from the sleeping brain alone, exactly where in the maze the rat had been. The rats were not just resting. They were replaying.
This finding cracked open a question animal lovers had always asked informally and scientists had long avoided: do animals dream? The answer, for mammals at least, is almost certainly yes, and the mechanism is the same one that produces human dreams.
REM, rapid eye movement, sleep is the phase when the brain is most electrically active. In humans, it's when vivid dreaming happens. Neurons in the cortex and hippocampus fire in patterns that closely mirror waking activity. The brainstem simultaneously sends signals that paralyse voluntary muscles, which is why you don't physically act out your dreams.
All mammals studied so far show REM sleep. So do birds. The platypus, one of the oldest surviving mammal lineages, spends more time in REM per sleep cycle than almost any other species, up to eight hours in young animals, and shows the most intense rapid eye movements of any creature measured. If REM duration is any guide to dream intensity, the platypus is having a very busy night.

What varies across species is the ratio of REM to non-REM sleep, the length of cycles, and how deeply the muscle paralysis sets in. Dogs cycle through sleep stages faster than humans, roughly every 20 minutes versus every 90, which may explain why a sleeping Labrador seems to twitch, whimper, and paddle its paws so often. Small dogs appear to dream more frequently than large ones, though large-breed dogs seem to have longer individual dream episodes.
The behavioral evidence in pets is hard to dismiss. A dog in deep sleep will sometimes bark softly, move its legs in a running motion, or growl at nothing. Cats do the same, whisker flicks, paw flexes, the occasional chirp directed at an invisible bird. These aren't random motor discharges. They occur specifically during REM, when the brain is most active and the muscle paralysis is, in dogs and cats, incomplete enough to let small movements through.

Stanley Coren, a psychologist at the University of British Columbia who has written extensively on dog cognition, has argued that dogs almost certainly dream about dog things: smells they followed that morning, the owner who came home, the squirrel that got away. This is consistent with Wilson and Louie's rat data. The sleeping brain doesn't invent abstract content, it replays and recombines what the waking brain experienced. For your pet, that means the walk, the meal, the afternoon nap in a patch of sunlight.
In 1998, neurobiologist Dave Margoliash at the University of Chicago found that song-related neurons in zebra finches fire during sleep in patterns that match the birds' daytime singing. The finches were, in some sense, practising their songs in their sleep. This isn't metaphor, the same neurons, the same sequences.

The octopus complicates the picture further. In 2021, researchers at the Federal University of Rio Grande do Norte in Brazil observed octopuses cycling through phases of active sleep during which their skin flickered rapidly through colour and texture changes, the same chromatophore activity that produces camouflage and communication displays while awake. The researchers, led by Sidarta Ribeiro, suggested the animals might be experiencing something analogous to dreaming, replaying visual scenes from their waking hours. Octopuses have no evolutionary relationship to mammals. Their complex nervous systems evolved independently. If they are dreaming, consciousness arrived at the same solution twice, by entirely different routes.
Fish show something resembling REM sleep at the neurological level, though whether this maps to subjective experience remains genuinely open. Invertebrates below a certain complexity threshold almost certainly do not dream in any meaningful sense.
The scientific interest in animal dreaming is not really about whether your dog sees you in its sleep. The deeper question is what dreaming reveals about the nature of consciousness itself. For a brain to replay experience during sleep, it must have encoded that experience as something worth replaying. It must have a model of the world detailed enough to simulate. That's a form of inner life, not identical to human consciousness, but not absent either.
For pet owners, the practical implication is simpler. A dog that twitches through a dream is processing its day. Waking it abruptly disrupts that consolidation. Most animal behaviourists recommend letting sleeping dogs lie, not as folk wisdom, but because sleep is when the brain cements what it learned while awake.
The rat in Wilson and Louie's lab ran the maze once with its body and once more with its neurons, alone in the dark. Every animal that dreams is doing the same thing: living the day a second time, in a brain that never fully switches off.