Episode 29: Alexandra Horowitz, PhD

The following interview was conducted in-class, during the Spring 2024 session of Hidden Figures: Brain Science through Diversity, taught by Dr. Adema Ribic at the University of Virginia. What follows is an edited transcript of the interview, transcribed by Bora Lynnette Ya Diul, Mary Elizabeth Schwab, Armand Muhametaj, Ryan Renee Lynch, Branson Kenneth Fowler, Sydney Marina Holton, Aidan M Kelley, Mia Bess Fabrizio, Imahkus G Manns, Ryan Bright Song, Aishani Patnaik, Shagun Ghotra, Evan Winston Thompson and Paige Cecelia O'Neil, who also drafted Dr. Horowitz’s biography. The final editing was by Dr. Adema Ribic.

Dr. Alexandra Horowitz is a Professor of Psychology at Barnard College. Her expertise are canine cognition and animal behavior. Dr. Horowitz is known for her New York Times bestseller ‘Inside of a Dog: What Dogs See, Smell, and Know’, among other significant works that have changed how we understand our canine companions. With a rich educational background, including an M.S. and Ph.D. in Cognitive Science and a B.A. in Philosophy, she explores the sensory experiences of dogs, challenging and enlightens our anthropomorphic views. Dr. Horowitz's teaching and research not only advance academic fields of animal behavior and cognition, but also deeply impact dog owners' relationships with their pets. She engaged wider audiences through her podcast ‘Off Leash’, where she delves into the human-animal bond. Dr. Horowitz continues to inspire many with her commitment to understanding and improving the lives of dogs and their owners.

Did you have any experiences in your early adolescence that sparked your interest in cognitive science and philosophy or did they strike you later in life?

It's always interesting to puzzle out whether there was any sign of what was going to happen in one's early life. I had a sensitivity toward animals and an interest in non-human animals, but I did not think that was going to manifest in a career in science or studying and writing about animals. It wasn’t a field I knew existed or even really did exist in the case of dogs, specifically. I think I was clueless.

Have you been interested in animal cognition your entire career? Who were some people who inspired you? 

When I decided to go to grad school, I was working at The New Yorker as a fact checker. I distinctly remember thinking, would I want to be the person doing the science or writing about it? I thought to myself that I wanted both. I had some interest in animal cognition at the time, but I didn't know that that was what it was. During my post-bac at Columbia University, I took a class in animal cognition, taught by Herb Terrace, who's still an Emeritus Professor of Psychology in the Psychology Department at Columbia, and Diana Reiss, who studied dolphins. Herb studied macaques at the time and I was in his lab briefly studying memories in macaque monkeys. Diana taught a class about early animal cognition research. This was in ‘96 and there were a handful of studies, mostly with big-brain animals like cetaceans and primates. I think that planted the seed. I was sure I wanted to do cognitive science and I didn't know how animal cognition was going to fit into that in graduate school. It took a few indirect turns before I found my way in grad school.

How did you know you wanted to do research with dogs and what made you fascinated about dog cognition specifically?

I did not want to do research with dogs. I just wanted to do research with non-humans. Most of the animal cognition research at the time was with primates, cetaceans, and some people studied behavior in pigeons or rats. When I was in the program before, I realized I didn’t want to do experimental work. I wanted to do work based on natural observation because the concept of thinking about what is going on in the minds of non-humans was emerging. I wound up studying play behavior. We all learn to pretend play, role play, and play with other people. Huge numbers of non-human animals also play, but they don't always play right in front of you, allowing you to get data from their play behavior. I had a dog named Pumpernickel and after many, many months of trying to research play and thinking about how I could do this, I realized that I had a playing species right in front of me and that I should just be studying dogs. Then, I needed to find dog biologists who could show me how to observe dog behavior, which I did with Marc Bekoff, who was at UC Boulder at the time. He studied all sorts of canids, including domestic dogs. It was one of his theses about the intentional nature of communication in dog play in a book he wrote with a philosopher named Colin Allen, which convinced me to study communication in dog play and see if I can observe if it's intentional and done with an interest in the attention of their playmate.

What is important about studying dogs in particular compared to other animals? 

I am very interested in seeing dogs just to sort of know about their life from their point of view and understand another species, which seems (to me) profound and interesting in its own way. The notion that there are these gazillion parallel lives of other sentient animals is one that we don't really appreciate and think about. I think if one has to bring it back to people, it also helps us empathetically understand people who are in different cultures, have different backgrounds, different ages-whatever it is-to be able to imagine others' points of view. I also think that as a domestic animal, they turn out to be an interesting species because we have changed their evolutionary development and that's unusual, and we've changed it specifically to be pleasing to us.

Is there anything from your early life that directed you toward animals specifically?

I grew up in a suburban rural area, where we would let our dog spend the night outside. He would just come back in the morning and that was pretty common. This idea that we should be in control of dogs all the time was not existing back then. That was when I started to think about what his experience was like apart from us. My current dogs don't have any experience apart from me. They're always with me. Even if they're off-leash, they're near me. In addition, we would go to school and pass this field where prairie dogs lived. Later on, prairie dogs were discovered to have elaborate communication and they live in a big social community. My attention was drawn by things like that. Can I say that it led to the career I have? I don't know, but certainly those were the salient moments in my development.

In your work, you've emphasized the importance of education for dog owners. Could you elaborate on how owners' increased awareness of their dogs' needs and behaviors can improve the dogs’ lives and also the relationships between the dogs and the dog owners?

I definitely think dog cognition research or any dog behavioral research can lead to an improved understanding of the dogs we are living with. That is inevitably going to have an improvement because I think we do have a very strange way of living with dogs now. One element of that is that they're considered property and it's really easy to get one. Most people in the US consider dogs to be members of the family. So there's this weird dissonance between the fact that we get them so easily, but then they're going to be with us for a really long time. We don't have to have spent a lot of time thinking about living with dogs before we get to be living with them. I think the result for most people is they don't spend a lot of time thinking about whether they're up for it and what the dog needs. Barking, for instance, could be considered just a misbehavior, but it's communication. By the way, they probably evolved to bark to communicate with humans because wolves don't bark. Barking is in the range of sounds that we can perceive and make, so it's probably communication with us. But we mostly think of it as misbehavior. To understand why the dog is barking is to get a different perception of barking, how to fix it (if you don’t like it) or what to do about it, and what they might be telling you, which is super interesting. It will also be really informative to learn that some barks are play barks, some are just “I'm alone” barks, and some are “There’s somebody at the door. I need to tell you about that” barks. So getting more information about what the dog is actually doing from their point of view changes people's understanding of their behavior in their homes. That almost inevitably seems to improve their relationship with dogs.

In your dog-human play study, your results indicated that not all types of play are created equal, meaning some forms of play produce more benefits for the dog and the owner. Is there a best way to play with our dogs to enhance our dogs’ qualities of life (and maybe ours)? Did anything about this surprise you?

We have different results. Different things happen to both us and dogs when we play with them differently. I think the best play one could do is one that suits the dog. And the dog will, if you let them, tell you which type of play they like to do. Some dogs are rambunctious players or they really like to run in parallel with you or they like to play with dogs more than people or they fetch. Having an expectation that all dogs are going to  play a certain way can lead to some dissonance in the dog–human relationship. I was surprised, actually, about people who played retrieving games with their dogs. [They] do a lot less movement than their dog. The dog is moving a lot, but they’re basically leaning over to pick up the ball. While the dog often seemed to be having a great time and showed a lot of positive affect, the people did not look like they had a lot of positive affect. They had very neutral affect, whereas people who played with their dogs very physically, where they were moving a lot, almost always looked delighted. I think that part of the co-participation of playing with your dog is moving with the dog. I have a dog who loves the ball, but I don’t just stand in one place and toss a ball for her. I’ll move around with her and create other elements of the game, and that’s as satisfying for me as it is for her.

Can you share some insights from your studies on dogs' understanding of their own size and the ability to navigate certain spaces and how this can help us better understand dog cognition?

Our study was really about their understanding of space but might also be considered about their understanding of self. We called it Can Dogs Limbo because we used a doorway aperture with a plexiglass that could be moved to lower positions. We had the dog on one side of the barrier and their owner on the other side of the barrier, so the dog kept having to try to get through this aperture to their owner. What was interesting to see is that dogs of all sizes stopped trying to get through at about the same size opening relative to their body. They also used the same kind of order of adjustments to modify their body to get through, like ducking their head and then putting their elbows down and then lowering their back and so forth. So there was a consistency across dogs and they were also able to gauge, either from looking or from attempting, whether they could fit through a space or not. There's human research that's sort of analogous to this that shows when our size changes drastically, such as when a woman is pregnant, we're very poor at estimating what size space we could move through. Our study showed that dogs know how big they are relative to an opening. I think the more meta-cognitive part is that they seem to be thinking about themselves in this way, so people would often ask me a question like, does my dog realize they're a small dog? Or do they realize they're so big? And in play, it does look like they do a lot of behaviors to self-handicap so that they can play with a small dog who's smaller and not as strong. This is another way of getting at this idea of having a sense of self, which is a really big question in cognitive science.

How does our understanding of dogs’ sense of smell influence the way we should interact with our dogs? Are there any practices you would recommend for owners to respect or enrich this sense? 

Dogs seem to exist predominantly in the olfactory world. Olfaction is maybe their primary sense, the way vision is our primary sense. They have perfectly good sight, but they're seeing the world through smell and smells are very meaningful to them. There are all sorts of ramifications for living with dogs. One is that we tend to mediate their smelling behavior, like a lot of people if they take their dog out for a walk, don't want them to smell things. For dogs, [smelling] is looking for them. I recommend letting them sniff and sniff each other. Sniffing another dog closely in the rump is just a way of getting information, it's like a handshake and should be allowed as long as the dogs both have consent and they will tell you if they do not consent. Also, I think it's useful for people to take their dogs on dedicated smell walks if that's how they see the world and what they appreciate about the world. A walk for a dog is pursuing the things that interest them, which is olfactory to a large extent.

You have mentioned that you seek to understand what a dog perceives and knows by using puzzles, objects, and types of games; could you describe what one of these tasks would look like and what would be measured?

One of my favorite studies was a study of what I called olfactory self-recognition, which was modeled on something called the Mirror Mark test. [This] is just a test where people ask non-humans whether they recognize themselves in the mirror and they try to see if they do by artificially putting a mark on a part of their body. For instance, chimpanzees can have the mark on their head and then the researchers can observe if the chimp then uses the mirror to guide their hand to the mark to investigate it, just like us looking in the mirror. Dogs don't seem to care about what's in the mirror when they see themselves in the mirror - they might think it's another dog and they might act socially toward it. They might even be able to use it to see what's around them, they [might] notice something's off about their appearance, but they don't seem to care about that. I thought the one reason they might not, besides [the fact that] they're not grooming animals, is that it's not an olfactory mirror. What I did was to present dogs with their own smell, their own odor, and then do a number of different comparisons. One of the comparisons was how they behave when smelling their odor vs their odor with a different smell added to it, just like a mark added on the forehead of the chimpanzee. We videotaped this and we then code the video later to measure exactly how long they sniff the canisters with odors. We found that dogs spent more time sniffing their own odor when it had another odor added to it, similar to how animals perform in the Mirror Mark test.

Many people ascribe human traits to their dog's, such as guilt or jealousy. Based on your work, how should we interpret these behaviors and what are they really telling us?

Anthropomorphism, using human words to describe non-human behavior,  is really natural and normal for humans to do and we've been doing it for thousands of years. With dogs, I think it's worth investigating whether our anthropomorphism, which we so readily come up with to explain the dog's behavior, is in fact accurate. Guilt is one I wanted to probe to see if that look a dog gives in all those dog shaming videos is a result of internalizing the feeling of guilt or the result of something else prompting their behavior. What I found was that when dogs had done something wrong, they didn't show the guilty look more often overall than when they hadn't. However, when their owner thought they had done something wrong and moved to begin to scold them, that was when they showed the guilty look, and this look is more aptly described as a submissive behavior or appeasement behavior designed to avoid punishment. Because the look doesn't happen more often when they've done something wrong doesn't mean that they don't feel guilt. Guilt is an internal subjective feeling that a scientist can't at this point in our life, not even with neuroscience, detect without verbal report. I think the important thing with anthropomorphisms is to empirically query them. 

What made you want to start writing books on canine cognition and are there any in particular that you recommend for us to read? 

When I started studying dogs, it changed my relationship with my own dog who I was living with at the time. That dog, Pumpernickel, went through grad school with me, and I just saw her differently. I was amazed because I had lived with her for over a decade and I hadn’t seen her so much. This area of science is very exciting. Every dog owner has questions about what their dog is thinking and doing with all these behaviors. [I thought] I should write a book that would help translate this new science. That’s when I wrote Inside of a Dog, which I’m revising now. Hopefully, it’ll be finished by the end of the year and there’ll be a second edition out next year. Reading it again, I still think it holds up. 

Do cognitive abilities vary among therapy dogs, do they have the capacity to comprehend complex human emotions, and have you found a significant difference between different breeds of dogs?

I don't study different breeds per se. Breeds are a very recent phenomenon in human history. They've only existed for the last 150 or so years. Many of my colleagues have done work on emotion, and they're mostly interested in whether dogs can perceive and distinguish facial expressions. [Dogs] are really good at distinguishing the facial expressions of humans, and they are interested in certain emotional expressions, like crying behavior. However, it is very unusual for a species to really understand the varieties of emotional expression of another species. For instance, we're really bad at understanding dog emotional expressions and we are looking at dogs all the time, so I think it's pretty neat that they appreciate ours.

In some of your past articles, you've talked about wanting to shift the focus of research to center on animals as participants recognizing the importance of their well-being but also for improving the quality of scientific research. Can you elaborate a little bit more about how we can ensure that research on animal behavior is more ethical?

Dog cognition research is done in the visual modality. In other words, we're presenting them with visual stimuli and asking them to make decisions which will give us what they know and understand. They're often able to do that at high levels. If we really wanted to understand dogs, we would be exploring their olfactory acuity and abilities and how they see the world through smell. Some people are starting to do that, and I think that's a great trend. Interaction, or what's called HAI research (Human Animal Interaction research), is about using animals to see if human health can be improved. Does it improve children's reading levels? Does it improve people's emotional state? Does it improve the life of people who have PTSD? Very few projects ask about how the dog is doing in those settings, and I think that deserves more examination. There are people who are researching the welfare of dogs in animal human interaction studies, and I think that's all for the positive.

What projects are you looking forward to working on next? Are there any new questions or topics that seem particularly promising to generate a greater understanding of dog cognition?

On some level, I’m just excited about whatever questions occur to us as we work as a lab to probe dog behaviors that we are familiar with but have never been investigated. For instance, right now we are trying to understand the function of shaking behavior in social interactions. It seems to be used specifically and intentionally in social interactions. That’s pretty neat, but nobody’s studying it. I also would really be excited to do more olfactory work where I have a little bit better equipment to explore exactly what it is that a dog is smelling about a stimulus.

Your career spans across multiple fields. How has the breadth of your studies influenced your work with dog cognition? What were some of the challenges in navigating both fields simultaneously?

I always feel like a dilettante and I think that's a challenge. But as it turns out, everyone in academia always feels like they're an imposter and all writers feel like that too, so that's apparently how you define yourself. My writing has definitely informed my research and my research has also informed my writing. I'm always thinking about how the research can be translated into changing societal attitudes about non-humans and in my writing, I'm very interested in representing the integrity of science and making sure that we keep advancing. I think they reflect back on each other pretty well. Overall, I feel pretty lucky.

In your book, The Year of the Puppy, you detail the first year of a dog's life; could you explain more about how you got interested in that idea and provide some of the things that you find really interesting about that first year of life?

The reason I was interested in doing that, apart from the fact that it would be fun to meet a lot of puppies for my profession, is that I adopted dogs who were several months or years into their lives, and anybody who does that wonders about their early life and how it might influence who they are today. I thought this was an opportunity to anecdotally begin to get a picture of that by following a dog from the beginning to whom we then come to live with. It is a really profound time in the dog's life, that early development, and it can absolutely have long-term ramifications if it doesn't go well. It can also have long-term ramifications if it does go well, but one of the things I want to mention that I found was that the dog we now live with, who was that puppy of that year, still has a lot of behaviors where if I met her later on and didn't know her as a puppy, I would say, “I wonder what happened that now she feels like she has to bark at every small dog she sees.” Well, you know what? Nothing happened. That's just who she is. That's just her personality. And I think that was an interesting revelation to me.

Is it important to you to make sure that your research is accessible and understandable to all age groups in order to provide better understandings of dog cognition and to potentially inspire younger audiences to be interested in this type of research?

I think we all as children are interested in non-humans. As young children, our picture books are full of non-humans. Our life is full of non-humans, and then at some point we stop caring about them. To generalize broadly, we change attitudes about non-humans, and I think that anybody younger than me is going to be a force for change in our society. I want them to have all the facts that we have about dogs and to continue that kind of empathetic interest and curiosity about non-humans. So the books that we’ve revised, we revised to target the young readers are aimed for precisely that reason.

What are the challenges that come with research concerning animals, whether that involve animal welfare or getting approval for such research? How have you been able to overcome these challenges?

The struggle with dealing with non-human animals is that we are very convinced by verbal responses and they’re our gold standard. If I ask you if you know something and you say you do or can show me that you know it, then I believe you and that's proof. We don't have any verbal animals. They're all communicating, but not in a language that we're great at reading. A challenge is finding ways to let them understand our task. Designing a task which allows their behavior to tell us something about what they know is challenging. We deal with owned dogs and we're not keeping dogs, so it's much easier to get approval because we don't have to provide housing for the animals. We are only responsible for their well-being and welfare in the half hour or hour or whatever that they're with us. Frankly, it's kind of interesting because who knows what their welfare is like for the rest of their life, but we're assuming at least it's not the college's responsibility. We do have to go through that and the approval can be challenging because it's an unusual species, but the Animal Care and Use Committee now knows me.

In your work, do you often find people trying to take insights from your research and use it to learn more about humans? With that idea in mind, have you ever felt that you had to justify work on dogs as important as its own field regardless of whether it could provide insight into understanding humans or human cognition?

I don’t think my work has been used that much to reflect back on human cognition. Perception is certainly part of cognition, but I think it’s reflected back in only that capacity. It’s making people pause and think about how we study smell in humans, which until recently was not that much. In terms of justifying my research, I am really lucky because I study owned animals, so the result of that is I don’t need to keep animals and it’s not expensive to run a dog cognition lab. In fact, a lot of my observations are outside in public such as at a dog run or something like that, and that doesn’t cost anything. My students get academic credit for working with me, or I have small grants and gifts that enable me to pay people or pay for the dog toy that we give all subjects when they end a study with us. We also get donations and things like that, so I don’t need a lot of money to run the research that I do. And thus, I’ve never really needed to justify my research, and I don’t think I could justify my work as improving human health. In terms of how it’s thought of by non-dog researchers, opinions are all over the place. Some people just think it’s not an important or interesting science and other people think it's a great evolution of human cognition and reflection on human cognition to study all sorts of non-humans. The opinions run the gamut.

What are some of your hobbies and do you feel that any have provided you with further insights or benefits to your work or research?

I wish I had a great hobby I could describe to you. The thing about being a researcher and writer is that I never leave my work at the office. I'm always doing that work. I have a family with 2 dogs and a cat, and I spend a lot of time with them. Outside of that, I read a lot of fiction and nonfiction because I think curiosity improves your thinking in all domains, it doesn't matter if it's relevant. I don't read a lot of dog cognition, but I love reading fiction and contemporary literature. I'm also a compulsive exerciser and that is very good for the brain. I have to say that's just reflective time, but I don't have any other especially interesting hobbies.

People often compare dogs' cognitive abilities to toddler specific ages. Do you agree that these comparisons are accurate?

People always say that dogs are like two-year-old or three-year-old children. I find that to be so funny, but maybe it tells us more about humans than it does about dogs or children because we love to draw these types of analogies. Dogs and humans are on a completely different trajectory. I don't think there's a great comparison to be had there, although I'm happy to just point out the ways that they're similar or different. In The Year of the Puppy, I enjoyed pointing out how developed a baby would be at the age of the puppy I was writing about. In 9 weeks, puppies are like an adolescent, while in 9 weeks, a baby is just starting to be able to hold up their head on their own. That’s all they can do. I could say puppies are brilliant and babies are dumb, but I will say they're just on different trajectories.

There’s a dog on TikTok named Bunny who presses buttons with words associated with them to express desires, emotions, and needs. What are your thoughts on this?

I have heard of the button dogs. They press these buttons that were developed for speech therapy for developing language learners, and there are many people who use these buttons with their dogs. I think that the buttons show that dogs can learn to make associations, and that’s about all that we know so far. There are people doing research on it, but it's premature to say that the dogs are telling us anything else. The dogs use words which we assume are in their vocabulary, and our tendency as humans is to think that they're telling us something. That's not seeing what dogs are actually doing. They're doing a lot of communicative behavior with us that I'd prefer we were paying attention to.

What is the biggest piece of advice that you would give to those who are interested in animal cognition or research with animals in general?

Find somebody who's doing research that you're interested in and tag along because the only way you can learn anything about animals is by spending a lot of time with them and looking closely at animal behavior. Try to forget what you think you know already and build it up again from your observations. It's brilliant because it's easy– it just takes time. You have the equipment, which includes curiosity, an open mind, and the animal in front of you. Any long-term observation about an animal will allow you to see something you haven't seen before. I am still seeing new things with dogs, and I've now been looking closely at them for over 20 years.

Is there a lingering question you continue to have in relation to dog behavior that you look forward to studying in the future?

It's funny because I've been asked this before and I feel like my answer always shifts, but it always has to do with olfaction. I want to know what the dog knows via smell– what they know about me, what they know about their environment, what they know about who's been by, when it's happened, how they're reading the weather through smell, and what they think of themselves via smell. I want to know what that world is like, and I don't know how, besides being a dog myself, I can do that.

Is there research on canine interactions with differently aged humans and are there any differences in the ways dogs interact with babies/children versus adults?

I have not studied this but I have lived around and observed it. From my non-scientific perspective, they are very perceptive about the capacities of babies and toddlers, which I think is interesting. Dogs do see babies and toddlers as different from adults, and, although you should never leave a dog alone with a baby, dogs are usually great with them. Similarly, when they see another dog who has a limited capacity, is older, or has a disability, dogs change their behavior so they can interact with that dog. They also see that babies and toddlers have a lot of resources, like stuffed toys, and they realize how they’re part of the gaze of the adults who used to gaze at them. Because of this, there can be a little bit of a competitive urge, but the solution to that is giving them a duplicate toy of what you buy for your baby.

In your research, do you find dogs are able to understand verbal commands or hand gestures better?

I haven't done that research, but there is somebody who shows that dogs are pretty reliable with hand gestures. They can be more powerful. That’s probably because we are more consistent with hand gestures than we are with verbal commands. Dogs are always trying to figure us out, so the more you can be consistent in communicating with them, the better they learn.

Do dogs develop a preference for distinct people based on some kind of metric, like olfaction? What makes a dog like one human more than another?

I don’t know the answer to that, and there’s a lot in dog cognition we don’t know. My sense would be that dogs are very attuned to a couple of things: what somebody provides in terms of affection, food, options for going out, other resources, and reactions. People react differently to dogs, and they are very sensitive to that. Somebody who is scared of a dog holds themselves differently than somebody who is not, and dogs can notice that. Humans can notice that if we look at it. In fact, there was a recent study that showed that people who were fearful smell different and dogs can presumably smell that too. If you’re somebody who’s scared of dogs, that doesn’t mean they’re going to take advantage of your fear. It just means they can tell that you're not having a great experience. I think that they're drawn toward people who are interested in them and not aggressive toward them. They’re also drawn toward people who can provide things.

Through your research, what do you think is the most important trait that a dog owner should embody or the best piece of advice that you would give to a dog owner?

Let them sniff. This is their olfactory world. Not letting them sniff is like going to the Grand Canyon and having your head turned away from looking at the giant gorge. Let them sniff. That’s how they experience the world and their welfare will improve.