Skip to content

Can Dogs Recognise Themselves?












Scientists have been testing animal intelligence for decades. Many have tested a variety of species’ sense of self by investigating whether or not they are able to recognise their own reflection in a mirror.

The mirror test was born when psychologist Gordon Gallup Jr. anaesthetised four chimpanzees and applied red dye to their eyebrows. When the chimps woke up and saw themselves in a mirror they began touching their eyebrows and so demonstrated that they could recognise their own reflection.

But how do you test an animal’s sense of self when they rely more on smell than sight to perceive the world?

The Efficacy of the Mirror Test

Many intelligent species fail the mirror test. Because these species perceive the world differently to the way we do, their failure is not indicative of a lack of self-awareness but rather evidence that the mirror test in inappropriate for them.

Dogs, for instance, have perfectly good eyesight but it is their sense of smell that they mainly use to interpret their world. It wouldn’t make any sense to present them with a mirror so how do you judge whether or not dogs possess self-awareness? Well, it would make sense to use a test involving scent and that is exactly what researcher have done.

Testing Jethro

Marc Bekoff of the University of Colorado tested his own dog Jethro. Over the course of five winters he would scoop up snow which Jethro and other dogs had urinated on and move it to different locations. He discovered that Jethro spent less time sniffing his own urine than that of other dogs. He could clearly recognise his own scent.

Adulterated Urine Test

Inspired by this experiment, Alexandra Horowitz of Barnard College designed a bigger study. She recruited 36 dogs and presented them with two canisters. She put a tiny drop of each dog’s own urine in one and in the other, she placed a drop of their urine mixed with another scent. This was the equivalent of marking a chimp’s head with red dye. The dogs spent more time sniffing the adulterated urine and so confirmed Bekoff’s findings.

Horowitz then presented the dogs with a choice between their adulterated urine and the other smell on its own. They sniffed both equally. This result suggested that dogs did not recognise their own scent and were merely interested in unfamiliar aromas. But Horowitz suspected that she might have chosen an alternative scent which was too powerful – cancerous tissue.

A New Experiment

She then repeated the tests with a less powerful scent, anise oil, mixed with the urine. The dogs then sniffed the adulterated urine for longer than their own urine alone and the adulterated urine for longer than anise oil alone. The new test showed that dogs are more interested in a new scent when it is mixed with their own. Just as with the chimps and the red dye on their head, something new may not be intrinsically interesting but becomes fascinating when it is connected to the animal.

Asking the Wrong Question

Asking a dog to recognise its own reflection is like asking a human to complete an exam in a foreign language that they do not speak. Dogs do have a sense of self but perhaps not in the way that we understand the concept of self-awareness. Dogs can recognise their own scent and probably possess limited understanding of how they are distinct from other people and animals. Right now, that is all we know.

We often try to interpret the intelligence of animals by comparing their ability to our own skills. In many cases animals have different skills rather than lesser skills. In investigating whether or not an animal can recognise its own reflection in a mirror, we are probably asking the wrong question.



We are very sorry, but the browser you are visting us with is outdated and not complient with our website security.

Please upgrade your browser to a modern secure version to view our website.