Episode 134: Dog Cognition & Social Learning

Dr. Angie Johnston joins us today to talk about how dogs learn (and how that compares to human learning). Dr. Johnston is an Assistant Professor of Psychology at Boston College where she is the director of the Canine Cognition Center and Social Learning Laboratory.

Learn about the similarities & differences between human and canine cognition, what dingoes can teach us about the domestication of dogs, and more!

Education is at the core of our mission here at Good Dog, and we’re always finding new ways to provide the latest and greatest in canine health and research to our community of Good Breeders. We’re excited to offer members of our community free and exclusive access to Good Dog Courses, a series of online educational courses that include in-depth videos, checklists, breeding tools, and more. This is completely free for members of our community, and you can access them by visiting here.

If you aren’t yet a member of our community and would like free access to all of this educational content and more, we invite you to apply to join at gooddog.com/join

Transcript

Mikel Delgado [0:29] Hello everybody, and welcome to this week’s episode of the Good Dog Pod. I’m excited to have Dr. Angie Johnston here with us today. She received her PhD in psychology from Yale University and is now an assistant professor of psychology at Boston College. She is also the Director of the Canine Cognition Center and Social Learning Laboratory. Dr. Johnson’s research investigates the evolutionary origins of human teaching and learning. She previously evaluated how children assess the information they learn from others, but in more recent work, she’s investigated which aspects of human learning are unique, and which are shared with our canine companions. She has explored some of the striking similarities dogs have with humans in their capacity to learn from others. We have invited Dr. Johnston here onto the Good Dog Pod today to talk about some of her recent research. So, Dr. Johnston, welcome to the Good Dog Pod.

Dr. Angie Johnston [1:19] Thanks for having me!

MD [1:20] Let’s start by setting the background. We could talk a bit about the idea of natural pedagogy—a fancy word for methods of teaching. But, specifically in humans—that’s what you’ve dedicated a lot of your work to—but partly because I think people are very interested in what dogs can do or “is your dog as smart as a 3-year-old child?” without first understanding why we care. Maybe you can tie in your early work with how children learn from others to explain how adults actually facilitate learning in children.

AJ [1:50] This theory of natural pedagogy, which is (like you said) just a fancy way of how children learn from other people, suggests that there are expectations that even young infants come to the table and you’re teaching them with. One of these expectations is when we use that high-pitched baby speech, like “Baby! You’re such a cute baby!” and you use the infant’s name or make eye contact, these are actually cues that tell the infant, “Hey, I’m about to teach you something.” This prepares them to learn from you. 

MD [2:25] It seems like we naturally want to talk to our companion animals with that voice. I know I do that all the time. 

AJ [2:32] There’s research that suggests it’s the same tone of voice that we naturally use between infants and companion animals like dogs. So this is a major reason why we like to study dogs with these questions, because we treat dogs much the same way we treat our own infants. 

MD [2:50] I mean, I think we’ve all seen that in our daily lives. Now, you were studying children. What made you decide to switch to dogs? Did you fully switch? Are you still doing kids, too? Is this comparative work?

AJ [3:00] It’s fully comparative work now. When I was in undergrad, I was in a lab that studied how children decide who to trust for new information. Some of the main findings we had are that when you teach preschool children new information, they sometimes are likely to go with a nice person who doesn’t know what they’re talking about over an expert who may be a little bit grumpy.

MD [3:24] Good to know. Be nice!

AJ [3:25] Be nice! Yeah. That was some of our original findings that we found. I went to grad school expecting to do much of the same work at Yale. As a side project, when I was a first-year student, I went with a postdoc to her dog daycare and worked with the dogs at her dog daycare. We just fell in love with it and decided to start a lab on campus at Yale that studies dogs. It was around my third year—halfway through my PhD—that I had a moment when I realized I’m in love with dog research. I’ve got to switch! So I switched, but I do still study research with children. It’s just always comparing to dogs. 

MD [4:10] Speaking of comparisons to dogs, there’s a research method called the violation of expectation, which has primarily been used with pre-verbal human infants as a way of getting information from them when they can’t tell you. If you’re doing research with adult humans, you often ask them questions and get answers, but with kids, they can’t talk yet and we can’t do that. Can you explain this method in a little more detail and how you adapted it for use in dogs? 

AJ [4:33] Yes. Basically with infants, this is a way that we can see what they’re thinking without asking them questions, like you mentioned. Perhaps the easiest way to think of it is like a magic show. If I were, for you (even as an adult), to show you a sleight of hand where I made something appear in a completely different location, you might stare and think, “What’s going on here?” and try to figure it out. Infants do the same thing when they’re surprised. They look for a longer period of time. We’re able to do magic shows where we have an object change identity or change location, and infants will react to be surprised by these changes. We use the same idea with dogs, where we show them magic shows, and we measure how long they’re looking at the outcome. Longer looking times suggest (in theory) that they’re surprised. 

MD [5:25] It’s great that you can describe your research as “putting on a magic show for dogs.” I can’t think of anything better than that! My background is in animal cognition, so I’ve probably seen about 1,000 studies about how dogs respond to the human pointing gesture and ostensive cues, which is another fancy word meaning “cues that draw your attention to something.” When you point to something, we certainly know (as humans), “I should look at that, because someone is pointing it out to me.” You tested how dogs respond to—this is a magic show, I think, based on what you just said—when a toy they’ve been presented with either changed location or type and whether it mattered that a human had pointed out the toy to the dog first. Can you explain—I think you can probably do a much better job than I just did trying to explain what happened during this study—and what you found? What was going on with the human and the dog in the study?

AJ [6:14] I think it’s helpful to start really quickly with the infants. What happens with infants is if you show them that an object either (like you said) magically appears in a new location or magically transforms into a new object, they tend to look longer on average if they’re just watching the show to changes of location. They seem to be more surprised if something has suddenly changed location. But this changes if you use those ostensive communicative cues that we talked about: the high-pitch speech and the eye contact and pointing. In that case, if you use those cues and point at the object, then they become more surprised when it transforms into a new object. People think this is because when you’re teaching a child something, you’re probably teaching them about how this tool works or what the name of this object is or things that are about the object—not “look at this location!” You’re usually trying to point out something that’s a little bit richer. So that’s what infants are expecting when you use these communicative cues. We tried this across two studies with dogs, using the same sort of magic show. The dogs did not seem to care about our ostensive cues in this case. They seem to be mostly surprised if it transformed into a different object in general. It didn’t matter if we were using ostensive cues or not. They were very focused on, “Hey, that’s a different toy!”

MD [7:40] That was my next question. What kind of objects were the dog-relevant objects, like dog biscuits or toys?

AJ [7:46] We thought about using food, but sometimes with magic shows, the food gets them too excited and they really want to run up and get the food, as any dog owner might know. The objects we used were just different toys, so we had a couple of different pairs of toys that were really different from each other. One was a plush star-shaped toy and one was a rope-tug toy. We tried to have different textures and things like that that they would notice.

MD [8:13] There’s been a lot of research that is looking at canine cognition and the human–dog relationship. Often they start with using the wolf as a comparison species. Your research has looked at the dingo. Why did you choose to compare dogs to dingos? 

AJ [8:30] Dingos are really cool because they give us a snapshot into an intermediate point of domestication. Essentially, what we know happened is that  tens of thousands of years ago, there was this animal that was very wolf-like. The last common ancestor of dogs, wolves, and dingos. But then, for several thousands of years, it went under a process we call “self-domestication,” where the animals that were more tame were more willing to come into human camps and eat the human trash, and those animals were more likely to procreate, and then their pups would be more likely to be tamed. This is a taming process that was not at all directed by humans—just a natural selection process. And then humans took those tame animals a couple thousand years ago and started breeding them for different purposes. Domestication is actually a two-part process. Dingos show us the intermediate point, because they went with seafarers to Australia from Asia and went back into the wild and haven’t been artificially selected by humans. But they did go through the self-domestication process at least partly. Wolves are showing us what a canid is like that hasn’t undergone any of these domestication processes, and dingos show us the self-domestication—

MD [9:50] Without the human imposition. 

AJ [9:53] So they haven’t been bred by humans. 

MD [9:55] That’s really cool that you incorporated them into your research. What’s the importance of eye contact in the dog–human relationship? You talked about that as an ostensive cue that’s used with children. Obviously dogs are sensitive to some of those same ostensive cues. Also, since you’ve done this comparative work, maybe you can talk a little bit about how dingos and wolves compare to dogs when we’re talking about eye contact.

AJ [11:58] Eye contact is a foundational behavior in the human–dog relationship. There is some research that suggests with dogs, when dogs and humans make eye contact, a hormone oxytocin is released in both species, which is known as the “love hormone.”

MD [12:14] This is why we feel all squishy inside when we’re staring at our dog. 

AJ [12:20] That gaze really does elicit that, so that’s an important cornerstone of dog–human bonding. In the project that I did with the dingos, we compared eye contact in dingos to a previously published study that looked at dogs and wolves. Basically, what that previous research with dogs and wolves showed was that dogs range a lot in their eye contact that they make with their owner. This is a simple task where the owner is just in the room for 5 minutes with the dog. They just see how much the dog naturally looks at the human. That’s it. Dogs range from about 6 seconds to about 2 minutes. Half the wolves never made any eye contact. The other half made like 1 second of eye contact. Huge differences there. These are wolves that are very familiar with the humans.

MD [13:12] Right, you’re not going out in the wild and finding wolves and trying to stare at them. 

AJ [13:16] Yeah, that probably wouldn’t work out too well. No, these are essentially pet wolves. The dingos we worked with to see where they would fall, and we found that the dingos—we tested 23 dingos, and all but 2 of them did make eye contact. Much more than wolves. But the duration of eye contact was much less than dogs. It ranged from something like 1 second to 20 seconds. 

MD [13:40] That’s a pretty noticeable difference. Interesting. 

AJ [13:44] What I think may have happened is that self-domestication process, that taming process, may have shaped these animals’ initial motivations to make any eye contact, but then it was artificial selection where we started selecting dogs that were making that ooey-gooey eye contact.

MD [14:00] Perhaps in the hope that we would give them a treat.

AJ [14:01] Perhaps. 

MD [14:04] Very cool. How would you summarize what your research has told us about the dog–human relationship so far? 

AJ [14:12] I think that a lot of the relationship that we’ve looked at has been on this domestication scale. I think, first of all, it is important that domestication is shaping this relationship. I think there’s been some controversy about that. There’s been some back and forth about how much is this a genetic difference in dogs, versus dogs just having more experience with humans. I think our research is really starting to show that this has been shaped by domestication. That’s one piece of it. I think the other piece is that we’re still needing to figure out exactly where are the similarities and where are the differences. Because it is surprising. The study we talked about earlier, with the disappearing objects—the dogs are different from humans. I think that what we’ve started to carve out is exactly how and where are these similarities between dogs and infants, and where are we starting to see that infants are showing unique behaviors that might be the sort of behaviors that can support our unique human culture and other aspects of human cognition that are uniquely our own? 

MD [15:19] Just thinking: why do you think there is that difference between human infants and dogs in that task, where human infants are paying attention to one feature and dogs are paying attention to a different feature?

AJ [15:29] I think that it comes down to human infants needing to be participants in our culture. So they grow up building on the shoulders of giants. They’re learning this information, and then they’re adapting and innovating on top of it. Dogs are just like… We just feed them, and they don’t need to do anything. They may not need to have some of these learning mechanisms that humans do. But when it comes to being motivated to learn from us and interested in our cues, that’s something that helps them work with us. There’s another level beyond that that they’re just not going. 

MD [16:08] It makes sense that, in some ways, when they’re dependent on us, we don’t ask them to get as much information as they might need if they were not completely dependent on us for food and toys and everything else. 

AJ [16:20] I think that’s right.

MD [16:20] So, you’re the head of this program at Boston College, the Canine Cognition Center and Social Learning Laboratory. Tell us a little bit about what you’re doing in your lab, what direction your research is taking right now, and what you think is the next big question in canine cognition.

AJ [16:38] The questions that we’re tackling right now are we’re still trying to piece apart the similarities and differences in human learning and canine learning. So we have several studies that are designed to look at whether dogs, for instance, will copy a human’s actions exactly, even when they’re not necessary to solve the problem. That’s something that human children do. It’s called overimitation, and there are other aspects of learning that dovetail with that in interesting ways, where children will stop exploring if you show them how to solve one particular part of a puzzle, because they think that’s the only part of the puzzle that matters. All of these different aspects of learning are things that guide children towards copying the teacher’s actions exactly.

MD [17:27] How do you test that in dogs? 

AJ [17:28] It’s very challenging. Our most exciting projects that we have right now are studies where we’re directly comparing dogs and children on the same puzzles. They have to be very simple puzzles, as you might imagine: things that involve just knocking things over, touching a button that you don’t have to press—just touch the button. Simple, physical tasks for our canine friends. 

MD [17:54] Very nice. How do you find participants? Can our listeners sign up to be in one of your studies? 

AJ [17:58] Absolutely. It’s really fun. I actually bring my own dog in to help pilot test some of our studies. It’s his favorite part of the semester, when he gets to come in and participate. Dogs of any age, breed, temperament. The only requirements we have are dogs can’t be aggressive towards humans. And they need to have their rabies vaccinations up to date. That’s it! Any other dogs can come in, in the Boston area. It’s just a lot of fun for the dogs. They work their way through their degree program. We started having some graduates. Starting their master’s programs.

MD [18:33] So if you want your dog to be a master of canine cognition, where can they sign up?

AJ [18:38] They can sign up at our website, which is bc.edu/doglab.  

MD [18:43] We’ll drop the link in the show notes as well. I guess the question I still have is: where is canine cognition going next? This field is relatively new compared to some other aspects of animal behavior research, but there’s been a lot of research on certain topics like the pointing gesture. Where do you think the field is going next? Or where do you think there are questions that we still need to answer? What are your priorities as a researcher?

AJ [19:09] I think there’s a couple of different domains. One that’s a topic-based question is eye contact. We still don’t fully understand what’s underlying the motivation to make eye contact. In our lab, we’re starting up some work that’s looking at why are dogs making eye contact. Are they trying to communicate something with us? Are they just trying to affiliate? Is it different in different contexts? We’re trying to build up our understanding of that, because it’s such a major behavior that shows up. It’s one of their only behaviors they can use to communicate with us, when you think about it, that’s human-like. They have a lot of body language features that are very different from us, so that’s a topic-based area. But besides that, we’re a very young field. The field really only started in 2000. I think we’ve laid the groundwork for a lot of “What do dogs do?” in general, but I think a lot of the questions coming up are trying to look at breed differences. We need more help to understand breed differences genetically. The geneticists are starting to figure this out for us. But I think breed differences. I think individual differences. Why are some dogs so likely to stare at their owner for long periods of time versus other ones that do not? Also thinking about service and working breeds. What does this tell us about those different groups? Are there things we can be doing to help train them more efficiently or ways that will make them have a closer bond with their handler? What are some of the applications? I think that’s where a lot of the field has a lot more work to be done. 

MD [20:41] Excellent. Well, I always like to end my conversations with guests on a fun note. Do you have any pets of your own? You already told us you have a dog, but can you tell us a little bit about them?

AJ [20:51] Yes! I have 3 dogs and 2 chickens. The dog that likes to come into the lab, his name is Vader. He’s a Pitbull Lab mix who’s about 10 years old, a rescue. He’s great. He loves doing puzzles. He loves being around new people. And then I have a 12-year-old Greyhound mix named Scout who is very sweet. And then a 5-year-old Hound mix named Finley, who is actually kind of dingo-like in some of her behavior. She takes a little bit longer to get comfortable with people, but then gets really close with them once she is comfortable and is pretty clever.

MD [21:35] Nice. And what are your chickens’ names?

AJ [21:35] Yara and Margaret. 

MD [21:38] Perfect. Thank you so much for being here with us today! We’re going to put the link to your lab in our show notes, so if people want to check out your research some more or participate in one of your studies, they know where to go. Dr. Johnston, thank you so much for being here with us today. 

AJ [21:53] Thank you for having me!