Alexandra Horowitz

Alexandra Horowitz is the author of the bestselling Inside of a Dog: What Dogs See, Smell, and Know and Being a Dog: Following the Dog into a World of Smell. She earned her PhD in Cognitive Science at the University of California, and teaches psychology, animal behaviour and canine cognition at Barnard College, Columbia University. Alexandra is the Head of the Horowitz Dog Cognition Lab, which undertakes research into the behaviour and cognition of the domestic dog.

Alexandra recently published the results of an experiment in which dogs were presented with an ‘olfactory mirror’ test. The traditional mirror test proposes to determine whether a non-human animal is capable of visual self-recognition. A mark is placed on the animal unawares, in a part of the body that they cannot see, and the animal is presented with a mirror. A handful of animals, including apes, dolphins, orcas and magpies ‘pass the test’ by investigating their altered reflection in the mirror. Alexandra’s study designed a scent-based equivalent, presenting dogs with canisters charged with an odorous stimulant. The study demonstrated that dogs successfully distinguish the olfactory ‘image’ of themselves.

In this Q&A, the Charter for Animal Compassion asks Alexandra whether the study showed that dogs possess a sense of self, whether other animals might pass such a test, and how we can train our noses to become more attuned to the world of the dog


Charter: Why was this study needed?

AH: I should say that I don’t know if it was ‘needed’, but I sorely wanted to investigate the question. It’s very difficult to prove that non-human animals have meta-cognitive abilities — things like theory of mind, self-awareness — but I’m fascinated by these abilities. Who hasn’t wondered what animals are thinking — and especially, if they think about themselves, reflect on their past, and plan their future?

There are comparative cognition experimental designs to attempt to answer these questions. One problem with these designs is that they were developed first in studies with humans; and the ‘task’ isn’t always clear, or even perceptible, to non-human subjects. I’m interested in remedying that with my own work. In this study, I tried to create an analogy of the mirror self-recognition test (that’s been used with many animals from chimps to dolphins and elephants) that is relevant for dogs. As dogs’ primary sense is olfaction, it makes sense to ask our experimental question — do dogs recognize ‘themselves’? — using smell, not vision. That at least needed to be done.

Charter: Did the study show that dogs have a sense of self?

AH: The idea with the mirror self-recognition test is that if you pass, that is an indication that you have a sense of self. I’m not sure that my own ‘olfactory mirror’ test with dogs — or any of the mirror tests — actually can do that. But that isn’t to say that ‘self’ isn’t being invoked in these studies. In all of them, the animals who pass have shown selective examination of a reflection of themselves which isn’t what they might expect. That behavior looks like ‘I’m checking this out because that’s not what I thought I looked/smelled like.’ There’s an ‘I’ in there, even if non-humans aren’t using language.

Charter: Do you think there are other animals that would pass an olfactory self-recognition test?

AH: That’s fascinating to consider. One African elephant passed the mirror self-recognition test, but elephants are also known for having very acute olfaction, so perhaps they would pass my test more readily. And insofar as many animals use chemosensory means to communicate, leave information about themselves, and get information about others, maybe many would. It makes me think that we have to expand our concept of ‘self’ to include animals with perhaps very high performance at this kind of task but a very different cognitive structure than humans have.

Charter: What role does empathy play in our understanding the inner lives of dogs?

AH: Insofar as we can be highly empathetic to other species, I think it’s a useful means to begin compassion for other animals. That said, I’m not sure that it always leads to complete ‘understanding’: we then have to do more work to try to understand the animal on its OWN terms, not simply as analogies to us.

Charter: Is it possible for us humans to develop our olfactory aptitudes (i.e. become better sniffers) and thereby to develop a deeper understanding of the world of a dog?

AH: Very much so! I tried to do as much over the last several years (and I wrote about this project in my last book, Being a dog: Following the dog into a world of smell). Humans don’t bring terribly much attention to smells, though we have plenty good noses. Smells usually just ‘come upon’ us; we feel afflicted by odors. But if we learn to smell intentionally, this changes. Nearly everything has a smell — sometimes one must move to sniff it closely; sometimes it is just a matter of noticing what is already entering your nose. In either event, what the exercise of practicing smelling did for me was to celebrate the dog’s olfactory acuity even more. I love my dogs’ smells; I let them smell me; I watch with wonder as they smell in the breeze, in my bag, follow a trail in the grass, find a lost glove. I got a glimpse into their world.


Visit Alexandra’s website:

Visit the Dog Cognition Lab:

Read Alexandra’s research: ‘Smelling themselves: Dogs investigate their own odours longer when modified in an “olfactory mirror” test’, Behavioural Processes, Volume 143, October 2017 (Link)