Guest Post: We Gaze at Dogs, Dogs Gaze at Us, Is It Love?


15282278241_d6c048c20b_kYesterday, scientists reported that dogs have found an unusual way to steal our hearts. When we stare at our human infants and they stare back, we both experience a rise in the hormone oxytocin, which has been linked to trust and maternal bonding. Now it appears that dogs have hijacked this hormonal response, causing our oxytocin levels to rise when they stare at us, and vice versa. The researchers claim that this behavior may have played a critical role in dog domestication, but is the study all it’s cracked up to be? Judith Lewis Mernit (editor at High Country News and animal watcher) and David Grimm (editor at Science and animal writer) hash it out.

Judith: David, you’ve written a thorough and super-engaging book about our relationship with domestic animals,  Citizen Canine: Our Evolving Relationship with Cats and Dogs, and you often write about the science behind domestication. What do you make of the findings of Miho Nagasawa, et al in their study, Oxytocin-gaze positive loop and the coevolution of human-dog bonds? Do they make a strong case that humans and dogs get an oxytocin rush as they gaze into each other’s eyes? In your book you have an experience of meeting wolves where you’re instructed not to stare at them. But as you also go on to show how dogs have evolved in ways that set them apart from wolves, ways that allow them to coexist amicably, for the most part, with humans. Is the “oxytocin-gaze positive loop” these researchers describe another example of that co-evolution?

David: Hi Judith, I think it’s an intriguing study, though I think it’s heavy on speculation.5676178824_09eca950fc_bDavidDogs were the very first thing humans domesticated (likely between 15,000 and 30,000 years ago), and we have a stronger emotional bond with them than with any other animal on the planet (and I’m speaking as a cat guy). The oxytocin study could help explain why this is the case. But we don’t know if, for example, we engaged in mutual gazing with squirrels for two minutes if we would both experience a rise in oxytocin as well. That experiment would be hard to do, but we need some sort of control like that to really prove this study’s point. (The wolves aren’t a good control because you can’t engage in prolonged mutual gazing with them–they, unlike dogs, view eye contact as a threat). It’s also unclear if this was evolution or co-evolution. Did dogs simply evolve to take advantage of a biological pathway that we use with our human infants? Or did we evolve alongside dogs to allow them to take advantage of this pathway, because it was to our advantage as well to form a close bond with them? (After all, early dogs were still quite feral, and the more we could bond with them, the more we could get them to do useful things like help us hunt.) These latter questions will be much harder to prove one way or the other.

Judith: I believe I have engaged in some mutual gazing with a squirrel or two. Squirrels can really stare a person down, though I suspect they’re tripping more on cortisol than oxytocin. My personal experience, though, tells me that dogs don’t, or at least don’t consistently, like to lock eyes with humans all that much. I’m not a scientist — I’m a journalist who happens to write about wildlife and energy development. But I am a pretty serious dog trainer: I show my two rescue dogs, a pit bull and a cairn terrier, in competition obedience. It’s a strange, old-fashioned sport, and none of my friends ever want to come watch it. But it interests me because it tests all those evolved skills you and other canine researchers have written about. Will the dog retrieve the item I’m pointing to, and not the other two within his view? Can a dog pick an object out of a pile that has my scent on it, and refuse the objects that someone else has recently touched? Will the dog respond to my hand signals 30 feet away?  My relationship with my dogs runs extremely deep because of the training I do. Still, my dogs and I don’t spend a lot of time gazing into each others’ eyes just for the pleasure of it. They’ll look at me and follow my eyes if I hold the key to some problem they want to solve, like how to get past the fence to a rodent den. But eye contact was something I had to diligently shape; there’s even a specific clicker-training exercise for it.

Cats are a different matter. My two cats love it when I stare at them. I’ve actually found that it isn’t enough to pet them — I have to look at them, too, for them to stop yammering for my attention.

David: That’s funny, I’ve found something very similar with my own cats. They seem to like it when we stare at them, and they’re happy to stare back. They even give what we call “eye kisses” when they look at us, slowly blinking their eyes, as if to say, “Hey, I love ya.” I’d love for someone to do the oxytocin study with cats and see if they see the same thing as with dogs.

As for some dogs not liking to engage in mutual gazing, that’s something the researchers found in the Science study as well. Only a fraction of the dogs engaged in this behavior with their humans, which again seems to discount the idea that this was some sort of critical evolutionary transition that helped dogs become domesticated. If it was that important, all dogs should do it.

Judith: That makes sense to me. Other studies have separated dogs according to breed. Working dogs like shepherds have been shown to be better at interpreting human gestures than non-workers like Basenjis and toy poodles, for instance. The Nagasawa paper only makes a distinction between “short-gaze” dogs and “long-gaze” dogs. I suppose I have short-gaze dogs. I know border collies that seem to get big kick out of locking eyes with humans (and, for that matter, squirrels); they might be long-gaze dogs. You can get a border collie to do anything.

David: My understanding is that the researchers didn’t see any breed differences, but they only tested around 30 dogs, so I’m not sure how many breeds they looked at.

Judith: I’m also interested in this term they use, “co-evolution.” I always thought of dog-human co-evolution as a one-way street. At a conference a few years ago, I heard Ádám Miklósi, the Hungarian canine cognition researcher, describe co-evolution as a process in which “one species exerts an important selective challenge on another that causes a specific evolutionary response.” Some researchers think that living with dogs has affected human genetics. Wolves’ and dogs’ smelling ability, for instance, might have freed humans from needing to smell to survive, and caused our noses to recede. But Miklósi was very clear that, while “there is ample evidence that humans exerted selective pressure” on wolves and dogs, so far there’s no evidence that the pressure has been reciprocal.

The authors of this oxytocin-gaze paper clearly believe in reciprocal co-evolution, though. They suggest that it’s not just dogs’ brains that responded to domestication. Our acquired tolerance for dogs means our brains changed, too. I’m wondering whether this is something people are re-thinking.

David: I don’t think we have enough data to say whether we co-evolved with dogs, but I find the idea intriguing. Again, we’ve lived with dogs far longer than with any other animal, theoretically giving us enough time to have adapted to them, as they adapted to us. We believe that early dogs helped us hunt, guarded our campsites, and may have even hauled heavy loads for us. There’s some speculation that this helped our species survive, transition from hunting-gathering to farming, and even out-compete our Neanderthal cousins. That’s some pretty critical stuff, and I wouldn’t be surprised if it had some sort of impact on our genome.

Judith: I’ve read Pat Shipman on that theory about how humans’ wolf/dog alliance drove the Neanderthals to extinction; it’s fascinating stuff. She also talks about gaze, speculating that the visible white in human eyes may have come about because it helped our dog-helpers follow our gaze from a distance. That’s pretty wild, but it’s captivating; it makes me giddy just thinking about it. Which brings me to my next question: Has it recently become more acceptable for academic researchers to study our relationship with domestic dogs? Karen Allen, a researcher at the State University of New York at Buffalo who’s looked closely at dogs’ effect on human stress response, has talked about being ridiculed by other researchers for focusing on the dog-human relationship. “Don’t try it if you don’t have tenure,” she said. And yet there seems to be so many of these studies coming out showing the benefits of our lives with dogs. Has something changed?

David: Yeah, there’s certainly a lot more dog studies than there used to be! For much of the 20th century scientists ignored dogs because they felt that they were somehow “corrupted”. The idea being that domestication had scrambled their brains, and thus they weren’t good subjects for studying behavior, etc. Then Miklósi and Brian Hare (a prominent dog researcher at Duke) independently showed in 1998 that dogs could understand the human pointing gesture, while chimpanzees could not. Those studies sparked a revolution in studying dogs, and now I believe there are close to a dozen labs around the world dedicated to canine cognition. So here’s the big question: why is no one studying cats? Actually, I already know the answer to that one.


David Grimm is a deputy news editor at Science and the author of the new book, Citizen Canine: Our Evolving Relationship with Cats and Dogs. Judith Lewis Mernit is a contributing editor at High Country News who writes about energy and the environment and whose work has appeared in the American Prospect, the Atlantic Monthly, Mother Jones and Sierra Magazine.  Both David and Judith are regular guest posters.

[Ed. note because she can’t help it:  working sheep dogs are a superb example of the human-dog relationship in action]


Gazing dogs:  Terry Kearney, Martin Sharman, both via Flickr



Share Button

3 thoughts on “Guest Post: We Gaze at Dogs, Dogs Gaze at Us, Is It Love?

  1. I suspect that the devil will be found to be in the details – eye contact alone can be a threatening stare or a friendly gaze. My dogs, and probably nearly all others, can tell the difference. Measuring the time that dog and human spent looking at one another is a simplification – an allowable one at this early stage, but it probably obscures some interesting subtleties.

  2. I think dogs like to do stuff. Eat, sleep, poop, run around that sort of stuff. When we stare at them they look back looking for a sign. Are we going outside? Are you going to feed me? How about a belly rub? More than a couple of minutes they give up on us and just go to sleep. My dog rolls over for a belly rub. Maybe that’s how he has trained me?

Comments are closed.

Categorized in: Animals, Behavior, Guest Post

Tags: , , ,