\ Interview with Professor Alexandra Horowitz | Columbia University Scientists and Engineers for a Better Society

Interview with Professor Alexandra Horowitz



Present is SEBS’s VP Events Christa and OCMs Laura and Christine L.





Christa: This is an interview with Professor Alexandra Horowitz and she will give us some background information on what she does, what she teaches and all her work.


Horowitz: Hi my name is Alexandra Horowitz and I teach in the psychology department at Barnard. I teach also in biology, animal behavior, and right now I am teaching introduction to psychology but I also teach a canine cognition class, which is an upper level seminar. And that reflects my professional interest in dog cognition, which is studying the minds of dogs and studying their behavior, making extrapolations from what they do to what they know, which is a field of animal cognition which comes out of my graduate research in cognitive science.


Laura: Could you tell us how you became interested in the field of animal cognition, specifically olfactory discrimination in dogs?


Horowitz: Sure, it’s actually a circuitous path to being an animal cognition researcher. I was in graduate school as a cognitive scientist, and I was interested in animal minds, but also human minds. But I especially interested in what humans could do but animals could not, and how we could test whether or not nonhuman animals have those abilities, especially cognitive abilities. I was particularly interested in a cognitive phenomenon called theory of mind, which is the knowing that others have knowledge and feeling sand perspective that is different than our own. And we all learn that as developing children, but it is a question for science whether non human animals have it. How do you test that? Usually we ask people if they know something that we know. We can’t ask animals and expect them to answer. So we look for behaviors that might show it. I won’t go into the long story, but I was interested in play behavior. I was looking for an animal that I could watch that did a lot of playing and that would show me about their cognition. Well, primates play but they don’t let you watch them or they don’t play all the time, meanwhile, I used to take my dog out three times a day and they play all the time. So as a graduate student, I transitioned to just studying dogs which was unusual at the time especially cognitively, and I got interested in how do we know what their experience is like, what their olfaction is like and how they see the world through smell.


Christine: What breed of dog do you use, and what determines your choice?


Horowitz: There is a difference in olfactory, cognitive, a lot of different ways of perception in breeds of dogs. But, I don’t particularly study specific breeds, because there is already so little research in dog cognition, so I wanted to study the platonic dog. It could be a mixed breed, any breed, because if I got all of one breed, it wouldn’t be any good for me because I would have a bias on my results. Once we have more foundation on the research, we will start doing specific breed research. In this field, we do have people that look at specific breeds, for olfaction, or sensitivity to human action, or their work drive, but for now, I just look at any breed. A dog is a dog.


Christa: How do you control for different breeds and their responses?


Horowitz: Dogs, regardless of breeds, are still representatives of their species. So in that end, you should be able to wash out any peculiarities of the breed, and even any temperament or personality differences. Assuming my samples are big enough, we should be able to assume that any dog represents a species.


Laura: How do you go about finding a large sample size for your experiments?


Horowitz: Here, I’m much luckier than most animal cognition researchers, I don’t keep any dogs, in fact, dogs aren’t even allowed here on campus, people keep dogs for me. People own dogs, live with dogs and are terrific volunteers of their dogs for experiments, because fundamentally, people are very interested in what’s happening in this animal that they live in, that’s always in their house. We have a natural interest in dogs which is why they are so ubiquitous, and the result of that is that the owner will bring them in for the experiment. So we will recruit from vet hospitals and put up flyers, use social media, and use the space like a dog day care space, or a park setting or owner’s home to do experimental trials. But we get a lot of volunteers, it is really wonderful.


Christa: What is the most important finding you have made in this field?


Horowitz: I think I do 2 strains of research in which I like a lot. One is about anthropomorphisms in dog, in other words, things that we think dogs know about, or things that are human like. One I tested was if dogs experienced guilt. I am interested in the fact that owners are always saying, my dog feels this, thinks this, wants this, etc. With guilt, it was pretty clear that owners think that their dogs feel guilt. As such, they punish their dogs if they feel they did something wrong, and the indicator is a specific look. And it varies a little from dog to dog, but it could their head is down, ears back, tag is wagging fast and low, sometimes they will look away. It looks guilty and is called a guilty look for a reason. So the empirical question was, does that look indicate that the dog knows he has done something wrong? So I did a really simple test to see what prompted that look. So the finding was that the look appeared most often not when a dog had done something wrong, but when they were confronted by an owner who thought they did something wrong and scolded them. So the dogs are putting it on as a submissive behavior, like don’t punish me. So we still don’t know if they feel guilt, but it is a good result because it was a way of challenging the types of things that we know automatically about dogs. So I thought that was a really interesting finding. And this, more recently the thing that we’ve done that I like a lot is about olfactory discrimination, which was mentioned. I started to study the pet dog’s olfactory world, not just trained work dogs but pets. Their olfactory world because there’s no doubt that they don’t see the visual world and they experience it in terms of the smell world. That’s got to be so different, in terms of what it’s like as a human going through the world. So, some of the types of studies people have done in the field seeing if dogs can distinguish number, like two treats from four treats. That’s interesting, and they can, but I was interested in whether they smell the difference between two and four because presumably that’s the more important sensory modality. So, and there it looked like actually a lot of pet dogs. And in that result it looks like, although pet dogs notice a difference, they often act, they’re much more likely to act following their owner’s commands about which one they should go to to choose which of them has more. If for instance they smell that there’s a plate that has two treats and one that has four treats and the owner stands in front of the one that has two, and its covered and says something like “yummy, treats” they’ll go where the owner point has pointed instead of following their nose. And I think that’s a pretty interesting result, that their so powerfully attentive to us that it even overcomes their natural way of perceiving the world.


Christine: Interesting. So how do you test whether it’s two or four treats?


Horowitz: It’s usually a forced choice. You allow them to smell both, in that case there would be plates that are covered with the different amount of quantity in there and they’re allowed to smell them one at a time or randomized. And then we measure two things, one how long they spent sniffing each one and also then we put the plates down a few feet away from them and they were allowed to make a choice about which one they could get.


Christine: Generally, do they go for the one with more?


Horowitz: If it’s visible, they’ll go for the more. I mean they always go for more, dogs. You can count on dogs appetite in a lot of experimental settings it’s based on, like, getting what the treat is. Almost all the experiments are about getting the treat in some way. If its visual, they’ll get more. So we thought they would also choose more if they could smell that it’s more. But they didn’t, they chose randomly. They smelled the one that was more much longer, so I think they either forget by the time they’re making the choice or the owner steps in and which one to go to, they just follow the owner and forget what their nose told them.


Christa: So I’m assuming that during the visual setting, it was the same amount of time that lapsed between them looking at it and their choice.


Horowitz: Yes.


Christa: So, in theory, they shouldn’t have forgotten about it, right?


Horowitz: That’s true. But they keep getting more information about the visual, they keep getting the visual information continually, whereas the olfactory information is this fundamentally different percept where it changes over time. That’s why it could have been a difference. You know, if the dog is uncertain or forgets what he smelled, is it one or three on the left or right, as he approaches the plates he might be able to remind himself. But with the smell, it’s not like that because the wind is not going in that direction or it’s harder to… but that’s part of what’s interesting to me about what it might be like to be a dog. They don’t just smell the plates, like you see the plates. And that’s interesting. What does it mean about the dogs experience when it comes into a room or when he sees you or when he doesn’t know and he barks at you instead of greeting you, but then suddenly realizes it’s you. That has everything to do with smell not traveling the same way light waves do and I’m really interested in that difference.


Christa: Of course. Was there ever a study conducted where you let them see the plates and then covered them up and wanted them to make that choice?


Horowitz: No, no not yet. It would be an interesting variation.


Christa: Just to see if they remember it visually, but not by smell.


Horowitz: Yup, yes that would be a nice change.


Christa: Okay, well I guess we been focusing on dogs a lot, but how does olfactory discrimination differ in dogs compared to other mammals?


Horowitz: Well actually, yeah a lot of mammals have quite good olfactory acuity. So, although especially some of the scenting hounds have probably much more epithelial tissue devoted to olfactory receptors than other mammals, cats and many other domestic mammals have pretty good olfactory acuity. I don’t know that it’s been tested between, for instance, between cows and dogs, but I’d assume dogs have more. But cows have better noses than we do, I’m almost sure. So, you know, humans are really anosmic compared to other species. You can just look at an animal and see if the nose is important to them. Not only by the size of the nose, that doesn’t tell you everything, but things like is it close to the ground. There’s a theory that one of the reasons that we don’t rely on smelling as much, because we have plenty good noses we just don’t smell that much, and a lot of our genes, a huge number of our genes, something like a thousand genes, are all about coding olfactory receptor cells, which is incredible, right? Much more than the amount that we would expect given how much we use olfaction. But probably sometime when we became bipedal people suggest that we stop relying on smell nearly as much. And so animals whose nose is near the ground, guess what, those are the smelling animals. But there are many many animals that have much better senses of smell than we do.


Christa: Have you done any studies on how interconnected vocalizations and if humans can actually mimic those in a comprehensive way?


Horowitz: I have looked at how dogs use vocalizations with each other in play, which is difficult to do because the sound of vocalizations for dogs is what people who study frequency analysis would call “noisy”. And that’s a technical term. It’s hard to find the fundamental frequency of barks and of a lot of other vocalizations that they use. So, I have only studied it between dogs, I haven’t studied person-dog comprehension or person-dog mimicry. Although some people have, they’ve done research on whether people can discriminate types of barks, for instance, or understand the context based on the sound. And even people who are not dog owners are good at that. But I’ve never done that kind of study.


Christa: How was the study between dogs, and what were your most significant findings?


Horowitz: Oh we never could find, we could never well-characterize what the distinct vocalizations were because the sounds that they make weren’t reliable from dog to dog, each one had so much individual variation. So we don’t know, we don’t know yet.


Christa:  So it’s still a mystery.


Horowitz: Yes, that’s right.


Christa: Do you think you could walk us, not from a research basis but from back when you were an undergrad, through your experiences and how you chose this path? I know you mentioned how you got into this as a grad student, but could you provide us with more background regarding your perspective as an undergrad?


Horowitz: Well I was not a scientist as an undergrad. I was a philosophy major. I loved philosophy and I loved the analytic approach to questions that philosophy has. It is one of the early sciences essentially and it has a rigorous approach and something about that appealed to me but I didn’t go into science. I left school and went and worked in an entirely different field. I worked as a lexicographer who is a dictionary definer and I worked in the New Yorker as a fact checker. I was in particular interested in the sciences because I had felt that I had not pursued a method that would lead to an answer to those philosophical questions. Philosophy is a great structure but there is no satisfying ending. There is no data. I went to Columbia as a post bac. I did a cognitive science mini degree – neuroscience classes, computer science classes and psychology classes. I worked with Terrace briefly who is in the Columbia department still and applied to cognitive science programs, which is not predominant. It really is all these fields that are interested in cognition. I went to UCSD for my graduate degree and I was a little focused on animal cognition but I thought I would study memory. So this whole time, I didn’t know that I was going to study animal cognition. So if there is anything that is to be taken form my experience is that at age 18, you do not have to know ahead of time what career you are going to be. You can and you might know that. For me, I was doing analytic philosophy, not dog behavior. But I bring a lot of my philosophical approach to my design of experiments and my analysis of behavior. Science and philosophy has a lot in common. As it turns out it have made me more appealing as a graduate student candidate because I have a varied background whereas other applicants have a lot of science background. Do not forget about having humanities classes because it changes the character as an applicant and you will have a more creative approach to your science.


Christa: Were you at all nervous deciding what to do as an undergraduate?


Horowitz: Probably I should have been more nervous than I was. I was a philosophy major and the career prospects are limited. Yes I was nervous, but I had a feeling that if I was doing something that I was driven to do, then I could make it happen. I have been able to. The reason I worked a dictionary company is because I love reading dictionaries. After a few years, I thought that I had enough of writing dictionaries. You read all sorts of magazines and books to try and get examples of words used in context and we would read the New Yorker, so I wrote to the New Yorker and said that I wanted to work there. I got a job there. When I wanted to get my degree in cognitive science, I knew I didn’t have enough background in science so I did a post bac year. I worked in somebody’s lab to do the stuff you need to do and that was sufficient. Did I get lucky? Maybe. But I was very interested in those moves along the ways and followed my interests. I pushed it and didn’t let it happen to me. It was something about having confidence in my own interests and following them.


Christa: So you went to grad school in UCSD. How did you end up here as a professor?


Horowitz: Well as a matter of fact. That is also surprising. I had lived in New York before going to UCSD and I wanted to come back to NY so I just came back and started applying to adjunct positions and immediately started working at CUNY. I looked up researchers who had done interesting work in animal cognition (Ristau) and she taught at Barnard and I wanted to meet her. So I looked her up and made contact with her and she introduced me to someone in the department and they asked me to propose a course for them and I did. And then I started teaching a year after graduating at Barnard. I started teaching more and more classes until I became a full time professor at Barnard. I started writing too and I’m working on a third book now. I don’t spend all my time as a professor because I spend some time as a writer and that’s good because I’m following my interests.



One Response so far.

  1. Simone says:

    Halloween look is so great and I enjoyed lot in my first visit. That first photgoraph of black home is really dangerous as well as now a days, Halloween theme is very famous for art work.