Mind Matters Natural and Artificial Intelligence News and Analysis
Colorful bluestreak cleaner wrasse on black background.
A bluestreak cleaner wrasse, Labroides dimidiatus, on a black background. This small colorful fish can be found on coral reefs Africa, Red Sea and Polynesia
Licensed from Adobe Stock

Did a fish just show self-awareness?

What if the whole question is founded on a mistake about the nature of the mirror test?
arroba Email

The bluestreak cleaner wrasse has passed the famous mirror test for self-recognition (originally intended for primate apes and monkeys).

According to a recent paper (open access), three out of four fish tested by researchers from Osaka City University in Japan were able to learn to identify the object in a mirror as their own images. But what does that mean?

When chimpanzees, dolphins, elephants, and magpies passed the test1, researchers theorized that these animals, recognized as intelligent, were demonstrating a concept of “self.” Now they are not so sure. Is the cleaner wrasse, which grooms other fish for parasites, really self-aware? Are fish much smarter than we think?:

A lot of people think fish are vacant animals with three-second memories,” study author Alex Jordan, a principal investigator at the Max Planck Institute Department of Collective Behaviour in Germany, told Gizmodo. “But if you educate yourself on what these animals can do, it shouldn’t be surprising that they can do something more complex. Ryan F. Mandelbaum, “A Fish Just Passed a Mirror Test for Self-Awareness, but What Does That Mean?” at Gizmodo

The mirror test was devised by Gordon Gallup of the University at Albany, New York, and first used on chimpanzees in 1970:

At first, they behaved as if they’d seen another individual by vocalizing and threatening. But after getting used to the mirror, they began using it to groom parts of their body they couldn’t otherwise see and to pick food from their teeth. To add evidence that the chimps knew they were looking at themselves, the scientists removed the mirrors, then put red dye on the chimps’ faces under anesthesia. They observed the chimp not touch their faces until they put the mirror back next to the cage, after which the chimps tended to the marks. Ryan F. Mandelbaum, “A Fish Just Passed a Mirror Test for Self-Awareness, but What Does That Mean?” at Gizmodo

Because monkeys couldn’t learn the trick, Gallup concluded that chimps had an “advanced form of intellect” and that the mirror test was “the first demonstration of a self-concept in a subhuman form.”

But most researchers are reluctant to make that assumption about the similarly dye-tested fish. True, fish can do many complex things, like create an intricately patterned geometric nest.

But so can insects and it doesn’t necessarily demonstrate “self-awareness.” As Jordan acknowledged to Mandelbaum at Gizmodo, the bluestreak cleaner wrasse (Labroides dimidiatus) are probably not self-aware, “They’re probably just using the mirror as a mechanism to see their body in the same way they can turn their head to see their body.”

What if the whole question is founded on a mistake about the nature of the mirror test? Let’s look at what we know:

First, all animals are probably “smarter” than we sometimes suppose. An amoeba is, in some senses, smarter than your computer and so is a fruit fly. The human brain exceeds the most powerful computers in efficiency.

Second, there does not seem to be a “ladder of intelligence.” It was once thought that mammals, especially those physiologically close to us, must be smarter than reptiles and reptiles in turn must be smarter than invertebrates because of their position on the Darwinian Tree of Life. Actually, crows can be as smart as apes, though they have very different brains. Intelligence doesn’t seem to reside in the details of the brain mechanism. Even lizards can be smart though their low metabolism reduces the need for problem-solving.

The octopus has been described as a “second genesis” of intelligence because it demonstrates that some invertebrates can be as smart as some vertebrates. Yet many life forms classed with the octopus on the Tree of Life (Mollusca) appear to lack anything we would consider intelligence. And plants can communicate in as sophisticated a way as animals though any resemblance between plant physiology and an animal nervous system is an analogy. Intelligence does not seem to be distributed in accordance with what we believe about evolution.

Third, the question is sometimes confused with the Hard Problem of (human) consciousness. We sometimes read that in the past people thought that animals lacked consciousness. For example, at The Atlantic, Ross Anderson writes, “This idea that animals are conscious was long unpopular in the West, but it has lately found favor among scientists who study animal cognition.”

No. If anything, the major problem has been the contrary; many people have assumed that animals think like people. Sometimes we infer complex thought processes of which the animal in question is most likely incapable. Unfortunately, we often fail to grasp a much simpler explanation for an animal’s behavior, unintentionally causing suffering.

Andersen sees however that human consciousness is a conundrum:

Even in a secular age, consciousness retains a mystical sheen. It is alternatively described as the last frontier of science, and as a kind of immaterial magic beyond science’s reckoning. David Chalmers, one of the world’s most respected philosophers on the subject, once told me that consciousness could be a fundamental feature of the universe, like space-time or energy. He said it might be tied to the diaphanous, indeterminate workings of the quantum world, or something nonphysical. Ross Andersen, “A Journey Into the Animal Mind” at The Atlantic

But to go onto claim, as he does, that “all across the planet, animals large and small are constantly generating vivid experiences that bear some relationship to our own” involves the risk of confusion about the nature of human consciousness.

A dog experiences needs, gratification, a sense of threat, jealousy, etc., as a self. In that sense, he is certainly conscious. But human consciousness involves deeper and more complex things, starting with the fact that we ask about his consciousness but he doesn’t ask about ours. One difference between humans and dogs is, as Michael Egnor notes, the human ability to process abstractions, including “consciousness.” That is a hierarchy; the higher includes the lower but the lower does not include the higher.

Lastly, there is a fundamental problem with the mirror test. It was originally developed to test chimpanzees, whose responses to a mirror invite a direct, physical comparison with those of humans:

So, what does this mean? Well, in part, it brings into question the MSR and mark tests themselves. Both become less and less justified the further one moves from apes and humans, with certainty decreasing as “taxonomic distance increases between the test species and the primate taxa for which the test was initially designed”. Stephen Fleischfresser, “Can fish be self-aware? The answer is far from easy” at Cosmos

Emory University primatologist Frans de Waal agrees that making the use of a mirror the sole test for self-awareness may be arbitrary:

An “olfactory mirror” test was tried with dogs that involved altering the scent of their urine. The interpretation of such tests may not be so clear, but they attempt to size up the cognition of animals that are less visual. The original mirror test “is not really an ideal test for an animal like an elephant or a dog who is so much oriented to smell. It’s sort of unfair for these animals to do these tests,” says de Waal. “There’s many more [tests] that could be developed, and they will be developed because this field is sort of taking off at the moment.” Carolyn Wilke, “The Mirror Test Peers Into the Workings of Animal Minds” at The Scientist

“It’s a question of whether you can become the object of your own attention,” Gallup told The Scientist recently. But he is surely blurring the central problem: Who is the “you” he is referring to? Self-awareness, as humans understand it, is a specifically human concept that includes many abstractions like name, identity, status, relationships, history, and so forth. A life form could surely learn to use a mirror to groom itself without experiencing anything on the spectrum of what a human means by self-awareness.

Using a mirror may not even be an equivalent experience between different types of animals. How would we know? What it is like to be a bat (Nagel) or a crab scuttling the silent seas (Eliot)? We don’t know because our ability to process abstractions does not transform itself into the ability to apprehend and experience utterly different life forms from the inside.

Overall, this has been a curious outcome for the mirror test. Those who felt reassured by close kinship with chimpanzees reacted quite differently when they were offered close kinship with fish, as a science fact. But we now have greater clarity about what sort of light a test of this type can shed on animal minds.


1 Horses, squid, parrots, dogs, and sea lions did not pass the test and the elephants’ and dolphins’ result were not yet replicated as of this writing.

See also: How is human language different from animal signals? What do we need from language that we cannot get from signals alone? (Michael Egnor)

Why only human beings speak: Language is a tool for abstract thinking—a necessary tool for abstraction—and humans are the only animals who think abstractly. (Michael Egnor)


Animal minds: In search of the minimal self

Denyse O'Leary

Denyse O'Leary is a freelance journalist based in Victoria, Canada. Specializing in faith and science issues, she has published two books on the topic: Faith@Science and By Design or by Chance? She has written for publications such as The Toronto Star, The Globe & Mail, and Canadian Living. She is co-author, with neuroscientist Mario Beauregard, of The Spiritual Brain: A Neuroscientist's Case for the Existence of the Soul. She received her degree in honors English language and literature.

Did a fish just show self-awareness?