Dogs: When They Smell Their Pee They Know It's "Me"

An interview with dog expert Alexandra Horowitz about self-recognition in dogs

Posted Aug 18, 2017

What a dog's nose tell a dog's brain

I'm always interested in innovative research on dogs. Thus, I was pleased to see a new study by Barnard College dog expert, Alexandra Horowitz, called "Smelling themselves: Dogs investigate their own odours longer when modified in an 'olfactory mirror' test." In an email to me about this novel research, Dr. Horowitz wrote, "I did a mirror self-recognition study inspired by and continuing in the vein of your yellow snow experiment lo those years ago." She's referring to what has come to be called the "yellow snow" study I did with my dog, Jethro, almost 20 years ago (for more details please see "Hidden tales of yellow snow: What a dog's nose knows - Making sense of scents"). 

Dr. Horowitz's study is available online. To set the scene for what she did and some of her conclusions we read: "... this experiment presents dogs with various canisters for approach and investigation. Each holds an odorous stimulus: in the critical test, either an “olfactory mirror” of the subject − the dog's own urine − or one in which the odour stimulus is modified. By looking at subjects' investigation times of each canister, it is shown that dogs distinguish between the olfactory “image” of themselves when modified: investigating their own odour for longer when it had an additional odour accompanying it than when it did not. Such behaviour implies a recognition of the odour as being of or from 'themselves.'"

free images es.freeimages
Source: free images es.freeimages

An interview from the dog's perspective about how they make sense of scents

I really like Dr. Horowitz's new and novel study, so I asked her if she could take time to answer a few questions about it. Gladly, she agreed. Our exchange went as follows.

Why did you do this study?

I've always been interested in the abilities that we attribute to humans that some assume distinguish us from non-human animals -- for instance, theory of mind and self-awareness. I've looked at theory of mind (in play behavior -- following your play research!), and now I wanted to look at self-awareness. The "mirror self-recognition test" is the best experimental test of whether an animal or human animal has a sense of self: if you see yourself in the mirror and there's a mark on your body, do you use the mirror to touch/remove/investigate the mark. Children pass the test by age two or so; chimps pass it, dolphins, and so too do various other animals. Dogs don't pass the test. They look at mirrors, they can use mirrors to see someone behind them, but they don't seem to care about that mark.

Without knowing a priori whether dogs have self-awareness or not, I still thought that their failure at this test didn't show that they don't. Dogs are olfactory creatures, not visual; they need a smell mirror. So that's the kind of thing I tried to design.

[Note: Dr. Michael Fox and I tried replicate the standard mirror test many years ago, and we also found that dogs and wolves didn't pay any attention to the mark. We tried to publish the paper, but it was rejected because, we were told, "negative data" weren't worth publishing. We tried to spread the word, pre-internet, but others subsequently tried to do a similar study and also failed. Had our data been published researchers likely would have tried other novel approaches, but that's another story. For more on the importance of publishing "negative results" please see "Why science needs to publish negative results."]

How does it follow up on previous work in this area?

In this area you were also an inspiration! Your study looking at Jethro's sniffing behavior of the "yellowed snow" around the area where you lived -- some yellowed by Jethro, some by others, and some moved by you -- was really asking the same kind of question: does Jethro know that this is his pee, his smell. In other words, him. I just brought it into a more standardized setting, with a larger subject pool, and using a more traditional "marking" of the dog's smell. [For further discussion of Dr. Horowitz's study please see Ed Yong's "Can Dogs Smell Their ‘Reflections’?"

How do your methods differ from previous studies?

Apart from the above, I also compared the dogs' time spent sniffing various odors, presented in pairs -- they included their own odor (a very small bit of their own pee), their odor with another odor added ("marked"), the "mark" by itself, and another dog's odor.

What are your major results and did they surprise you?

My main result was that the dogs spent longer sniffing their own odor, marked, than their odor by itself. That is analogous to what the mirror self-recognition studies look for: more examinatory behavior of the mark on themselves than of their image alone.

This was my hypothesis, so it wasn't entirely a surprise -- but it was entirely a delight that the data bore it out.

What does they tell us about dog behavior and how should your findings be used by people who share their lives with dogs?

Most owners, I suspect, have no question about whether their dog has a kind of sense of self -- whether he or she thinks about himself or herself as being an agent with a past, present, and future -- distinct from other dogs. They assume that dogs do. My study supports that assertion. 

Perhaps more importantly, to me, is the idea that when we ask questions of dogs -- do they have a sense of self; do they understand what I'm saying; do they have memories -- the kinds of approaches we use to ask may not be right. Looking at dogs' behavior in a mirror isn't relevant for this question. Looking at their reaction to their "smell reflection", as it were, is. And that's because dogs are more olfactory than visual.

What are some of your current and future projects?

Our lab is working on designing more studies around how the dog's sense of smell organizes their experience of their environment (if your readers live in NYC with dogs and want to participate, they can sign up at!). And I'm working on a book about dog-human interaction which I'm really excited about. In other words: dogs, dogs, dogs.

Many thanks Alexandra. In her email to me telling me about her new study, the subject line was "Jethro's example lives on," and I'm so pleased it does. The more we know about the behavior of our dog companions, the more we can appreciate who they are and what they need and want. It's a win-win for all.

There's still a lot to learn about the cognitive capacities of dogs, and it's going to take novel research projects like this to tap into what's happening in their brains and noses.  Please stay tuned for more on the cognitive and emotional lives of dogs and other animals. 

Marc Bekoff’s latest books are Jasper’s Story: Saving Moon Bears (with Jill Robinson); Ignoring Nature No More: The Case for Compassionate Conservation; Why Dogs Hump and Bees Get Depressed: The Fascinating Science of Animal Intelligence, Emotions, Friendship, and Conservation; Rewilding Our Hearts: Building Pathways of Compassion and Coexistence; The Jane Effect: Celebrating Jane Goodall (edited with Dale Peterson); and The Animals’ Agenda: Freedom, Compassion, and Coexistence in the Human Age (with Jessica Pierce). Canine Confidential: Why Dogs Do What They Do will be published in early 2018. Learn more at