Do Animals Think? What Science Can and Can't Tell Us About Animal Minds

Imagine you’re walking through a park with a friend, and a dog runs up to a tree, plants its front paws on the trunk, and starts barking at the branches. Your friend says, “That dog thinks there’s a cat up there.” You nod. It seems obvious.

But is it? How do we really know what’s going on inside that dog’s head? And what about a honeybee that dances to tell its hive-mates where to find flowers? Or an octopus that learns to unscrew a jar lid to get at food inside? Do they think? Are they conscious? Do they have beliefs?

These questions have been around for a long time. But over the last few decades, scientists and philosophers have started working together to find better ways to answer them—and they’ve discovered that the way we ask the questions can shape the answers we get.

A Strange Starting Point: Darwin’s Hunch

The modern study of animal minds really begins with Charles Darwin. In the 1870s, Darwin argued that the difference between human minds and animal minds is “one of degree and not of kind.” He meant that we’re not fundamentally different from other animals—we just have more of whatever it is that makes thinking possible. A dog dreaming, a bird remembering where it hid food, a monkey using a stone to crack nuts—Darwin took all of this as evidence that animals have emotions, memories, imagination, and even something like reason.

But here’s the thing: Darwin might have been too quick to see animals as like us. When you’re expecting to find similarities, you tend to find them. And that’s where things get complicated.

The Problem of Seeing Too Much (or Too Little)

Scientists who study animal minds have to be careful about two opposite mistakes.

The first mistake is anthropomorphism: seeing human-like mental states in animals when they might not be there. A famous study from the 1940s showed this tendency beautifully. Researchers showed people a short video of simple geometric shapes moving around a screen—a triangle, a circle, a square. Even though these were just shapes, people couldn’t help describing what they saw as a story: “The little triangle is trying to escape from the big square” or “The circle is scared.” We see minds everywhere, even where they probably aren’t.

The second mistake is the opposite: anthropodenial, a term coined by primatologist Frans de Waal. This is the refusal to see animal minds when the evidence suggests they’re there. It’s assuming humans are unique and special, and denying that animals could share our capacities.

So which mistake is worse? Most scientists have historically been more worried about anthropomorphism. They follow something called Morgan’s Canon, a rule from the 1890s that says: never explain an animal’s behavior using a “higher” mental process if you can explain it using a “lower” one. If a dog opens a gate, maybe it’s not because the dog understands how latches work—maybe it just learned that pushing against the latch sometimes makes the gate open, and it wants to get to the street. The simpler explanation wins.

But some philosophers argue that this rule is itself a bias. Why should simpler explanations be more likely to be true? Evolution doesn’t always produce simple solutions. Sometimes the simplest explanation for a behavior is that the animal is actually thinking. And if we always prefer simple explanations, we might end up denying animals capacities they really have.

The War on Simplicity: A Case Study

This fight over simplicity became really intense in a debate about whether chimpanzees understand what other chimpanzees can see.

Here’s the setup: a chimpanzee sees that a competitor chimpanzee is looking in a certain direction. There’s food hidden in one of two locations. If the competitor can see where the food is, the chimpanzee might not bother going for it (since the competitor will take it). But if the competitor can’t see the food—because there’s a barrier, or the competitor is looking the wrong way—the chimpanzee goes for it.

So do chimpanzees understand that others have visual experiences? Do they know what another chimp can and cannot see?

One group of scientists said: “Yes, it’s simpler to explain this by saying the chimp understands seeing. If we had to explain the chimp’s behavior using only learned rules (like ‘don’t go for food when the other chimp’s eyes are pointing at it’), we’d need a different rule for every situation. One mental explanation covers all cases.”

Another group said: “No, it’s simpler to say the chimp just reads behavior. We know the chimp can see the other chimp’s body position and eye direction. That’s all it needs. Adding mental states is an extra layer we don’t need.”

Both sides claimed simplicity. Both couldn’t be right. And this debate showed something important: simplicity isn’t a single thing. There are many ways to be “simpler”—fewer mental processes, fewer evolutionary changes, fewer behavioral rules. And these different kinds of simplicity often pull in opposite directions.

How Should We Decide? Problems with the Science Itself

The chimpanzee debate reveals a deeper issue. Science is supposed to be objective. But the science of animal minds is full of value-laden and theory-laden assumptions—philosophical commitments that scientists bring with them, often without realizing it.

Consider the mirror self-recognition test. A scientist puts a mark on an animal’s forehead while it’s asleep, then shows it a mirror. If the animal touches the mark on its own face, it’s supposed to “pass” the test and be considered self-aware.

Humans pass this test. Chimpanzees pass. Gorillas don’t. Dogs don’t. But does that mean gorillas and dogs lack self-awareness? Maybe gorillas fail because in their social world, staring at another gorilla’s face is a threat—so they avoid looking in the mirror. Dogs fail because they rely more on smell than sight. When scientists gave dogs a smell version of the test (using their own urine), dogs showed that they recognized themselves: they spent less time sniffing their own scent than another dog’s.

The mirror test assumes that vision is the most important sense. But that’s a human-centered assumption. Maybe we should be testing animals using their own ways of experiencing the world.

Here’s another example. Scientists have spent decades trying to teach great apes to use human language. Why? Because we value language. We think it’s the most important capacity. But that choice shapes what we study and how we interpret the results. When apes “fail” to learn language, we count it as evidence that they’re fundamentally different from us. But maybe we’re just asking them to do something that doesn’t matter much to them.

So, Can Animals Think?

Given all these complications, what can we actually say about animal minds?

Some philosophers think the question is misguided. René Descartes, in the 1600s, argued that animals are like machines—they have no thoughts at all. In the 1980s, philosopher Donald Davidson revived this idea with a more sophisticated argument. He said that to have beliefs, you need language. Why? Because having a belief means being able to be surprised when reality doesn’t match what you believed. And being surprised requires understanding that there’s a difference between how things seem and how they really are. To grasp that difference, Davidson argued, you need the concept of belief itself—and you can only get that concept through language, through talking with others about what’s true and false.

If Davidson is right, then animals don’t have beliefs at all. That famous dog barking up the tree doesn’t believe the cat is up there. It just behaves as if the cat were up there.

But most contemporary philosophers disagree with Davidson. For one thing, his argument would also make it hard to attribute beliefs to human infants or people with severe language impairments. And many philosophers think we can know quite a lot about what animals believe by carefully observing their behavior in controlled situations. Maybe the dog doesn’t understand “cat” the way we do—maybe it thinks of the cat as a “chaseable thing” rather than a “small furry mammal.” But that’s still a kind of belief.

Others argue that animals might have non-conceptual forms of thought—ways of representing the world that don’t involve the kinds of concepts we use. A bee might represent the location of flowers using a mental map, without having any concept of “north” or “south.” Does that count as thinking? It depends on what you mean by “thinking.”

What’s at Stake?

This isn’t just an abstract debate. How we answer these questions has real consequences.

If animals are conscious—if they feel pain, pleasure, fear, and joy—then we have moral obligations toward them. Most people already agree that mammals and birds are sentient (capable of feeling). But what about fish? What about octopuses, who have complex nervous systems very different from ours? What about bees? Philosopher Michael Tye argues that crabs and bees are conscious. Others disagree. The evidence is messy, and our theories of consciousness don’t give us clear guidance.

If some animals have more sophisticated capacities—like the ability to plan for the future, to form friendships, or to understand fairness—then our obligations might be even stronger. Maybe it’s not enough to keep animals free from pain. Maybe we also owe them freedom, companionship, or the opportunity to live meaningful lives.

Some philosophers and activists have argued that great apes, dolphins, and elephants should be granted legal personhood—the status that protects individuals from being treated as property. This argument depends on evidence about their cognitive capacities: self-awareness, empathy, social bonds, the ability to suffer.

The Bigger Picture

The study of animal minds forces us to confront something uncomfortable: our own biases. We want to know whether animals think, but the tools we use to find out are shaped by what we value, what we assume, and what we’re afraid of finding.

Maybe the most honest answer is: we don’t really know. The science is young. The philosophical questions are deep. But we’re learning something important along the way—not just about animals, but about ourselves. Every time we design an experiment that fails because we assumed animals sense the world like we do, we’re reminded that there are other ways of being in the world. Every time we catch ourselves saying an animal “failed” a test when it might have just been solving the problem differently, we’re reminded that our way isn’t the only way.

The next time you see a dog barking up a tree, or a bird caching seeds for winter, or a cat watching a mouse with patient stillness, you can ask yourself: what’s happening inside that head? The question is harder than it looks. But maybe that’s what makes it worth asking.

Key Terms

Term	What it does in this debate
Anthropomorphism	The mistake of assuming animals have human-like mental states when they might not
Anthropodenial	The mistake of denying that animals share mental capacities with humans when they actually do
Morgan’s Canon	A rule that says scientists should prefer simpler explanations for animal behavior
Theory-laden	When observations are shaped by the theoretical assumptions a scientist holds
Value-laden	When scientific choices are shaped by what scientists care about or think is important
Self-awareness	The capacity to recognize oneself as an individual, often tested with mirrors or smell
Moral status	The standing that determines whether something is owed moral consideration
Sentience	The capacity to feel pleasure and pain, often used as the basis for moral consideration

Key People

Charles Darwin — The naturalist who argued that human minds and animal minds differ only in degree, not in kind
C. Lloyd Morgan — A psychologist who worried about reading too much into animal behavior, and proposed Morgan’s Canon
Donald Davidson — A philosopher who argued that animals cannot have beliefs because they lack language
Frans de Waal — A primatologist who introduced the term “anthropodenial” and argued that scientists are too afraid of seeing minds in animals

Things to Think About

If we can’t ever be certain that an animal is conscious, how much evidence is enough to justify treating them as conscious? When does it become more risky to assume they aren’t conscious than to assume they are?
The mirror self-recognition test assumes that seeing yourself matters. But what if you’re an animal that cares more about smell or hearing than sight? Can we really say animals who “fail” the mirror test lack self-awareness?
Both scientists and philosophers claim their explanations are “simpler.” But what counts as simple depends on what you focus on. Is it simpler to say a chimpanzee understands minds (one explanation) or that it has learned hundreds of behavioral rules (many explanations)? Which kind of simplicity should we prefer?
If we discover that bees or fish are conscious, should that change how we treat them? What if they’re conscious but can’t feel pain in the same way we do? Does that matter?

Where This Shows Up

Animal rights debates — The question of whether animals should have legal rights depends partly on what science tells us about their minds
Pet ownership — How we treat our pets reflects assumptions about what they feel and understand
Conservation — Decisions about whether to protect habitats for insects and fish often assume these animals don’t have rich inner lives
Artificial intelligence — Scientists who build “thinking” machines face versions of the same questions: what counts as thinking, and how would we know if something else is conscious?