Animals across a wide range of species appear to play with their food, from housecats batting a half-dead mouse across the kitchen floor to dolphins passing a pufferfish back and forth like a beach ball. The behavior looks frivolous, even cruel, but the motivations behind it turn out to be surprisingly varied. Some animals are managing fear, others are practicing skills they’ll need later, and at least one group may be getting intoxicated on purpose.
Why Your Cat Torments the Mouse
Domestic cats are probably the most familiar food-players, and they’re also among the best studied. Research on cat predation found that the tendency to bat, pounce on, and release prey before killing it is driven by an internal conflict between hunger and caution. When a cat is very hungry, it tends to kill quickly. When prey is small and easy to overpower, the kill also comes fast. But when those factors push in opposite directions, when a cat is less hungry or the prey is large enough to bite back, the cat is more likely to play with it before, after, or instead of killing it.
That pattern makes intuitive sense once you think about what “play” looks like in these encounters. The cat is repeatedly engaging and disengaging, testing the prey’s ability to fight back while staying at a safe distance. A rat can deliver a nasty bite, and even a cornered mouse will lunge at a cat’s face. The battering behavior lets the cat exhaust the prey and reduce the risk of injury before committing to a close-range kill. Far from being pointless cruelty, this is risk management.
Hunger level matters in a specific way: a very hungry cat overrides its caution because the need for calories outweighs the cost of a potential bite. A well-fed cat has no such urgency and can afford to be more cautious, which paradoxically makes it look more playful. This is why indoor housecats, which are never truly hungry, are the worst offenders when it comes to toying with prey. They have maximum caution and minimum motivation to finish the job quickly.
Dolphins and the Pufferfish Question
Bottlenose dolphins have been filmed nudging pufferfish between members of a group, gently mouthing them and passing them around before letting them go. The pufferfish are never eaten. This has led to a popular claim that dolphins use pufferfish to get high, and the science behind it is more interesting than the headline.
Pufferfish produce tetrodotoxin, a potent neurotoxin, primarily concentrated in their skin, especially in younger fish. When a pufferfish feels threatened, it releases some of this toxin into the surrounding water. At high doses, tetrodotoxin is lethal, but at the low doses a pufferfish releases during gentle handling, researchers have proposed it could induce a trance-like state. Observations note that dolphins behave differently after these encounters, appearing to drift near the surface in an altered state. A recent study examining the ion channels of dolphins in the genus Tursiops found genetic changes suggesting their sodium channels have a lower affinity for tetrodotoxin compared to other mammals, which could make them more resistant to the toxin at low exposures while still being affected by it.
Whether this qualifies as “playing with food” is debatable since the dolphins don’t eat the pufferfish. But the behavior fits the broader pattern of animals manipulating live prey for purposes beyond nutrition. If the pharmacological explanation holds up, dolphins would be one of very few animals that appear to manipulate another species for its chemical effects.
Birds That Impale, Fling, and Stash
Object play in birds, particularly parrots and corvids, often involves behaviors borrowed from their foraging toolkit. Lone birds will carry, fling, rip apart, and otherwise manipulate inanimate objects using the same motor patterns they employ when dealing with food. This suggests that at least some avian food play is practice, keeping predatory and foraging skills sharp in the absence of actual prey.
Shrikes take food manipulation to an extreme that looks especially bizarre. These small predatory songbirds are famous for impaling their prey, insects, lizards, even small birds, on thorns or barbed wire. What looks like gruesome play actually serves multiple practical purposes. Shrikes use impaled prey to mark territorial boundaries, to store food for periods of bad weather or breeding stress, and to divide labor between mating pairs. Perhaps most interesting, impaling lets certain prey items “age” on the spike while toxic defensive chemicals in the prey’s body decompose, making the food safe to eat later.
That last function is a reminder that what looks like playing with food is sometimes closer to cooking it. The shrike isn’t tossing the grasshopper around for fun; it’s sticking it on a thorn and waiting for the poison to break down, the way you might marinate tough meat. The behavior is practical processing, not play, despite looking unsettling to human observers.
Otters and Their Rock-Juggling Habit
Sea otters and Asian short-clawed otters are frequently observed juggling stones, rolling them between their paws, tossing them from hand to hand, and passing them over their chests. The behavior is endlessly charming on camera and has prompted the obvious question: is this practice for cracking open shellfish?
A study of captive otters tested this directly by measuring how often individual otters juggled rocks and then giving them food puzzles that required similar dexterity to solve. The result was clear: juggling frequency did not predict how well an otter solved the puzzles, and it didn’t predict how long they spent interacting with puzzles before cracking them.
So juggling doesn’t seem to be skill rehearsal, at least not in any straightforward way. A separate study of captive Asian short-clawed otters that were introduced to unfamiliar natural prey items, such as crabs and snails, found that the otters spent considerable time manipulating these prey with their paws, carrying them around the enclosure, and at times appearing to juggle them in the same way they juggle stones, before eventually trying to extract the meat.
This paints an interesting picture. Otters clearly apply the same manipulation behaviors to both rocks and food, but the rock juggling itself doesn’t make them better at food extraction. One interpretation is that juggling is a form of general exploration or entertainment that uses the same motor routines as foraging without being functional practice for it. Young and old otters juggle, and hunger doesn’t seem to increase it in a consistent way. Whatever is driving the behavior, it isn’t the simple “practice makes perfect” story that many people assume.
Bears, Salmon, and Selective Waste
Brown bears during salmon runs engage in a behavior that might not look like play at first glance but shares some features with it: they catch fish, take a bite or two, and then discard the rest. Sometimes they eat only the brain or the roe and leave the body on the riverbank. To a casual observer, it looks like the bear is being wasteful or toying with its catch.
Research on Alaskan brown bears found that this selective consumption tracks closely with how many fish are available on a given day. When daily local prey availability was high, bears were significantly more likely to discard partially eaten fish. At peak abundance, with over seven thousand live fish available in the immediate area, the probability of discarding reached about 30 percent for both males and females. Bears weren’t just being sloppy; they were selectively consuming the most calorie-dense parts of each fish, the brain and roe are fattier than muscle, and moving on to the next catch.
This behavior isn’t really play at all. It’s optimal foraging. When salmon are abundant, the energetic cost of catching a new fish is low, and it’s more efficient to eat only the richest parts and discard the rest than to finish every last morsel. When fish are scarce, bears eat more of each catch. The discarded carcasses end up fertilizing riverside vegetation, which in turn feeds the next generation of insects that feed the next generation of salmon. What looks like waste is part of a nutrient cycle.
The broader point here is that “playing with food” and “selectively processing food” can look identical from the outside. Bears flipping salmon around on the riverbank, eating a piece, dropping the rest, walking away, these behaviors don’t read as efficient predation to a human watching, but they are.
Octopuses and the Floating Cap
Play behavior in invertebrates is rare and hard to confirm, which makes a recent study of captive California two-spot octopuses especially striking. Researchers provided octopuses with test tubes containing shrimp, training them to unscrew the caps to get their food. After the octopuses mastered this task, something unexpected happened: all three individuals that had access to the free-floating cap began playing with it.
The play consisted of the octopus releasing the cap from its grasp so it floated upward into the water current, then reaching out to grab it as it drifted past, pulling it in close, and releasing it again. This sequence repeated multiple times and was clearly distinct from how the octopuses handled actual prey, which they captured and consumed immediately. In some cases, an octopus would hold the cap for several minutes before releasing it and resuming the play cycle. None of the octopuses showed any interest in playing with the test tube itself, despite it having similar floating properties, suggesting something specific about the cap’s size or feel made it appealing as a toy.
This is technically object play rather than food play, since the octopuses weren’t manipulating the shrimp. But the context matters: the cap had been part of the food-acquisition process, and the play behavior emerged directly from the feeding enrichment routine. The researchers scored this as level 4 object play on their intensity scale. For an animal with a brain organized nothing like a mammal’s, this kind of repeated, apparently purposeless manipulation of a non-food object is hard to explain as anything other than something like fun.
When Food Processing Gets Mistaken for Play
Not everything that looks like playing with food qualifies. Japanese macaques are famous for washing sweet potatoes in water, a behavior first observed in the 1950s that spread culturally through a troop on Koshima Island. More recent research has examined how these monkeys handle food contaminated with sand. Monkeys were sensitive to sand on their food and responded with increasing effort as contamination got heavier, spending more time brushing and washing food when it was dirtier. Dominant monkeys of both sexes showed a strong preference for dry brushing over washing in water.
To a casual observer, a monkey dipping a sweet potato in the ocean, rubbing it, dipping it again, and rubbing some more looks like it might be playing. But the behavior is calibrated to contamination level: clean food gets eaten immediately, lightly sandy food gets a quick brush, and heavily contaminated food gets extended washing. That scaling rules out play and points to deliberate food processing. The monkeys are cleaning their food, not amusing themselves with it.
This distinction matters because humans are prone to over-interpreting animal behavior as playful. When a raccoon dunks its food in water (their German name, Waschbär, literally means “wash bear”), it’s engaging tactile receptors in its paws that work better when wet. When a bird drops a snail on a rock repeatedly, it’s cracking the shell. These behaviors involve extended manipulation of food items but don’t meet the criteria researchers use to identify play, which typically include repetition without a functional outcome, voluntary engagement, and behavior that differs from the animal’s normal functional repertoire.
The Anthropomorphism Problem
Humans have a strong tendency to project our own emotional framework onto animal behavior. When a cat bats a mouse around, we see cruelty. When a dolphin passes a pufferfish to a friend, we see recreation. When a bear drops a half-eaten salmon, we see waste. Each of these interpretations carries emotional freight that may or may not map onto what the animal is actually experiencing.
The challenge for researchers is that play is genuinely difficult to define in animals. The standard criteria, that the behavior must be voluntary, repeated, apparently pleasurable, and not directly serving an immediate survival function, are all judgment calls. An otter juggling rocks looks like it meets every criterion, yet the same motor patterns appear when otters handle novel prey items, blurring the line between play and food exploration.
Researchers have also noted that captivity changes the picture dramatically. Animals in captive environments have their survival needs met, which frees up time and energy for behaviors that wild counterparts might rarely perform. The octopus cap-playing study was conducted in a laboratory. The otter juggling studies used captive populations. Wild animals of the same species might engage in these behaviors less often or not at all, simply because they have more pressing things to do. This doesn’t make the captive behaviors fake, but it does mean that viral videos of animals playing with their food disproportionately feature captive individuals in environments designed to promote enrichment-seeking behavior.
What the Behavior Accomplishes Across Species
Looking across the animal kingdom, the reasons animals manipulate food beyond what’s needed to eat it fall into a few broad categories. Risk management accounts for much of cat prey-play: the animal is testing whether the food can fight back. Skill development explains some bird and cetacean behaviors, where juveniles practice hunting techniques on prey they don’t intend to eat. Optimal foraging, as seen in bears, involves selectively consuming the most nutritious parts and discarding the rest when food is abundant enough to afford waste. Chemical processing, as shrikes demonstrate, uses time and exposure to render toxic prey edible. And at least some behavior, like otter rock juggling and octopus cap play, resists any clean functional explanation and may be something closer to what we’d call entertainment.
The category that tends to surprise people most is the pharmacological one. If dolphins genuinely manipulate pufferfish to experience low-dose tetrodotoxin exposure, that puts them in a very exclusive club of animals that seek out mind-altering substances. Some species of lemur have been observed biting millipedes and rubbing the secretions on their fur, and various birds practice “anting,” placing ants on their feathers, though these behaviors are generally interpreted as pest control rather than intoxication. The dolphin case, if confirmed, would be distinct because the animals appear to seek out the experience repeatedly and share it socially.
What unites all these examples is that the human label “playing with food” bundles together behaviors with fundamentally different causes. A cat prolonging a hunt, a shrike aging its catch, a bear eating only salmon brains, and a dolphin mouthing a pufferfish are all doing something different, for different reasons, with different neural machinery. The common thread is that food, for many animals, is not just something to swallow. It’s something to assess, manipulate, process, learn from, and occasionally, it seems, enjoy.