Your Dog’s Breed Doesn’t Always Determine How They’ll Behave

12:15 minutes

an illustration of a pitbull from the neck up. its skin is made up multicolored geometric patterns
Credit: Shutterstock

The dog world abounds with stereotypes about the personalities of different breeds. The American Kennel Club describes chihuahuas as “sassy,” and malamutes as “loyal,” while breed-specific legislation in many cities target breeds like pit bulls as stereotypically aggressive. But do these stereotypes say anything true about a dog’s personality and behaviors? 

New research in the journal Science looked at the genomes of thousands of dogs, both purebred and mutt, plus owner reports on personality traits. And their findings were more complicated: Yes, many behaviors have a genetic or heritable component. But breed, it turns out, may be a poor predictor of many things, including aggression or friendliness. 

Guest host Umair Irfan talks to co-author Elinor Karlsson about the complexities of genetics, personality, and breed in our best friends.

Further Reading

Donate To Science Friday

Invest in quality science journalism by making a donation to Science Friday.


Segment Guests

Elinor Karlsson

Elinor Karlsson is a professor of Bioinformatics and Integrative Biology at the University of Massachusetts Chan Medical School, and director of the Vertebrate Genomics Group at the Broad Institute of MIT and Harvard in Cambridge, Massachusetts.

Segment Transcript

UMAIR IRFAN: This is Science Friday. I’m Umair Irfan. Ira Flatow is away this week. And today, we’re going to tackle some of the prejudices and unfair stereotypes about dog breeds. Dog people out there, you know what I mean.

Beagles have a reputation for howling. Poodles are needy. Labs are sweet. Certain breeds are allegedly more aggressive, while bigger dogs are calmer than smaller dogs. And the American Kennel Club website even offers you a chance to pick the perfect breed for your family. You can filter your search by how much boxers bark or how much Spaniels snuggle.

Well, what if I told you that breed is not, in fact, behavioral destiny? New research from a team at the university of Massachusetts looked at thousands of dogs for genetic links to their personality traits. And they found, well, even when there is a genetic link, breed doesn’t have as much to do with how a dog acts as much as people seem to think. So how much is nature? How much is nurture? And how much are we just projecting onto our best friends?

Here to explain more is Dr. Elinor Karlsson, a professor at the University of Massachusetts Chan Medical School and director of Vertebrate Genomics at the Broad Institute. Welcome back to Science Friday.

ELINOR KARLSSON: Thank you very much for having me.

UMAIR IRFAN: I just rattled off a bunch of generalizations we have about different breeds of dogs. But why did you set to disprove them in the first place?

ELINOR KARLSSON: That’s an interesting question. I think partly I was just curious. I’ve been working in dog genomics since I was a graduate student.

But when I entered the field, I didn’t actually know anything about dogs. I never owned a dog. I still haven’t owned a dog.

And I immediately started hearing all of these stereotypes, just like you just explained, about their behavior. But it never totally made sense to me. We know that the dog breeds are only maybe 150 years old. These are the modern breeds with these really defined standards.

And behavior is really complicated. It involves many genes. It involves your environment. And creating and selecting something in 150 years just isn’t something you can do.

UMAIR IRFAN: And from your results, you found that breed doesn’t really shape behavior as much as we thought. Could you unwrap that a little bit?

ELINOR KARLSSON: Yeah, so we took a– so we had 18,000 dogs that people had answered survey questions for. And then we had another, out of those 18,000, 2,000 of them we also had very detailed genetic information for. And we took this data set and asked this question in pretty much as many ways as we could think to ask it, because we knew that this was going to be counterintuitive for a lot of people.

And I think at the end of the day, there’s two ways of thinking about what we found. The first is kind of from the perspective of a dog’s owner. So what we discovered when we surveyed a whole lot of people about their dogs– and our project was open to all dogs. So whether your dog is a purebred or a mixed breed or you have no idea where your dog comes from, we would love to have your dog in our in our project.

And what we discovered was that in every single breed, owners were reporting basically every type of behavior and personality that you can possibly imagine in a dog. So Beagles, we found owners often reported that beagles tended to howl, but there were Beagles that never howled. Labradors aren’t supposed to howl at all, but there are definitely Labradors that howl. And that was basically true of everything that we looked at. And so what that means from the point of view of a dog owner is that if you go and adopt a dog and you get a beagle, that basically you could get anything in that whole spectrum of personalities and behaviors that the owners had told us about.

UMAIR IRFAN: And in particular you looked at mutts, hybrids or dogs that aren’t that don’t belong to any specific breed. What can a mutt tell us that a purebred, like a Saint Bernard or Golden Retriever, cannot?

ELINOR KARLSSON: Yeah, so one of the most fun things I think about this project is that we were able to open it up to all dogs. In the breeds, you’ve got this behavioral stereotype that says all Labs are friendly. But at the same time, all Labs also look like a Lab and are yellow colored and the right size, because that’s what a Lab is supposed to be. And so you can’t actually test behavior separate from what the dog looks like. But in the mutts, that all gets mixed up, because the DNA gets all shuffled up. So you can actually disconnect these things in a way that you just can’t do in the purebred dogs.

UMAIR IRFAN: Interesting. So it seems like there’s a bit of a distinction here between how behavior is influenced by genetics and how behavior is attributed to breeds. But it seems like there are some behaviors that do have a genetic component, right?

ELINOR KARLSSON: Oh yeah, definitely, for sure. And that’s what we found, is that most of the behaviors were shaped by a dog’s genetics. It’s just that that doesn’t have anything to do with their breeds, because people think of breeds as being a uniform thing, but there’s actually a lot of genetic diversity, even within a lot of breeds. And so basically what we’re saying is it’s genetics, but it’s not genetics that’s different between the breeds.

UMAIR IRFAN: Interesting. So then are there any breed stereotypes that do hold up?

ELINOR KARLSSON: Yes, sort of. So this is where scientists get annoying, where we don’t give clear-cut answers on things. So what we found was that some of the things that people would expect, things like the Border Collies test– overall, Border Collies tend to be more biddable, which is whether they want– whether they’ll do what it is that you want them to do, and toy directed, meaning that they like to play with toys.

That’s not to say that every single Border Collie has those characteristics. And, in fact, there’s many, many Border Collies who don’t. But on average, Border Collies do tend to have– do tend to be more biddable and more toy directed. And that definitely lines up with the idea that people have of Border Collies as these dogs that are great to do agility things with, where you’re actually training your dog to do this whole fun obstacle course with you.

UMAIR IRFAN: And there are some real-world stakes to this, too, because dogs also do important jobs for us. And I’m wondering, then, should we be selecting dogs for jobs based on breed or for behavior?

ELINOR KARLSSON: So that’s a tricky question. One of the things we didn’t look at in our paper was the question of working dogs, because it adds a whole new complexity to the story. So what we think is going on in the breeds that we talked about here– so we’re talking about just modern breeds, AKC, those kinds of organizations, where you have a very tightly defined breed standard about how the dog is supposed to look. And there we think that mostly they’re being shaped by selection on what they look like, their aesthetics, rather than selection on anything to do with their behavior.

The working dogs often turn that on its head. And so they don’t actually care anymore what the dogs look like. What they’re breeding for is how the dogs behave and how they perform. And we would expect to see in working dog populations that we are going to see more behavioral differences than what we see in the regular breeds that we’ve looked at so far.

I suspect if you talk to people in the guide dog world, they would tell you that with the Border Collies, the fact that on average the Border Collies tend to be a little bit more biddable and tend to be a bit more toy directed probably suggests that if you wanted to breed a dog that was going to be really good at herding, if you started with a group of Border Collies to try and breed those dogs that you wanted, you would have an easier time than if you started with dogs from a different breed or just dogs overall. But you still would need to have that process of actually selecting on the behavior for many generations. And you can’t just expect them to be what you want right out of the– right out of the gate.

And, conversely, there’s no reason to say that you couldn’t get exactly the dog you want for that job by going down to the shelter and finding it there, because that diversity is in the entire dog population. It’s just that it might be at a bit higher frequency in one of the breeds.

UMAIR IRFAN: Mhm, and I’m thinking specifically about how many dog owners want to avoid aggression, particularly for certain breeds, like Pit Bulls, that have a reputation for it. I mean, is there a way that a Pit Bull owner could have like a genetic test that shows that don’t worry, the genes of my dog show that she’s sweet?

ELINOR KARLSSON: So one thing that did come out of our research is that that is entirely unnecessary. So we asked people all of these questions about their dog’s behavior. I think there were about 110 questions in total. And we took all of those and fed them into something called a factor analysis. And what a factor analysis does is it basically identifies groups of questions that are all asking the same thing and puts them into a single score. And because you’ve got many different questions asking the same sort of thing, you have a lot more confidence that you measured it correctly.

And so we did this and we came out with eight behavioral factors. And one of them was bidability. And another one was what we we’re calling toy-directed motor patterns, which is how much your dog wants to play with toys.

And one of them was something that we ended up calling agonistic threshold. And if you look at the questions that go into agonistic threshold, it’s basically all about whether your dog reacts aggressively to something surprising in their world. And agonistic threshold came out as having almost no difference between the breeds when we were looking at the survey data, and also when we looked at the genetic data doesn’t seem to have hardly any heritability at all. It just doesn’t seem to be a particularly genetic trait. So it seems like whatever that is, it’s something that’s coming from aspects of the dog’s environment or the way people are perceiving the dogs rather than anything that they’re born with in their DNA.

UMAIR IRFAN: So then to reiterate, then, what are the limitations, then, of trying to use genetics to understand our dogs?

ELINOR KARLSSON: Right now, genetics is in its infancy. And so you have stories like this one coming out all the time that tells you all the things that we’re learning from DNA. But at the end of the day, we’re not very good at predicting from genetics what an individual is going to be like, whether it’s a person or a dog.

And we may never be very good at that. It might be that this kind of environment problem persists as being something that dominates that. But even if we get there in dogs, it’s going to be a while. And we’re going to need much bigger sample sizes to get there.

So on the one hand, I think people should be really cautious about over interpreting what they learn about their dog’s genetics. If you want to know what your dog’s behavior is like, you shouldn’t be basing this on what somebody says about what your dog’s breed ancestry is. What you should be basing it on is a dog that you’re actually living with that you know better than anybody else knows.

And so I think for dog owners, that’s the most important message, is that– I’m a genetics person. I love genetics. I think that those genetic tests where you get to go figure out what breed your dog is, I think that’s super fun that people are interested in that. But that doesn’t mean that you’ve actually learned anything new about your dog, if that makes sense.

UMAIR IRFAN: So it makes more sense to pay attention to what your dog actually does than what the genetic test says about them.

ELINOR KARLSSON: Exactly. I will follow up on that just briefly to say that from a scientist standpoint, as I said, these subtle differences in behavior between the breeds are actually really exciting. And they’re going to give us– the fact that we now know that some of the breeds tend to be on average a little bit more biddable, we can actually go in and figure out why that is. How do you actually change DNA to get a dog to be obsessed over retrieving a ball or to be more interested in listening to what it is that you want them to do or to howl?

At the end of the day, as a scientist, I’m really interested in figuring that out. And what we found in the study actually shows that that is something we are going to be able to do and that we’re going to have a lot of power to do. Pretty much the one thing we’ve really managed to get out of this whole study is even more questions than we had when we started.

So we would love for people to go to our website, which is at darwinsark.org. You can sign up your dog. You can fill out all of these surveys for your own dog. We’ll tell you how your dog scores on the eight factors that we talk about in the paper.

And then if you choose to, you can order a DNA kit. And we can also tell you what breeds your dog has ancestry from. It’s an open resource. We’ll share it with other scientists. We don’t sell this data. We don’t own this data. And we’re really just trying to make large-scale dog genetics easier for scientists to explore.

UMAIR IRFAN: Dr. Karlsson, thank you so much for joining us again.

ELINOR KARLSSON: No, thank you so much for having me.

UMAIR IRFAN: Dr. Elinor Karlsson, professor at the University of Massachusetts Chan Medical School and director of Vertebrate Genomics at the Broad Institute.

Copyright © 2022 Science Friday Initiative. All rights reserved. Science Friday transcripts are produced on a tight deadline by 3Play Media. Fidelity to the original aired/published audio or video file might vary, and text might be updated or amended in the future. For the authoritative record of Science Friday’s programming, please visit the original aired/published recording. For terms of use and more information, visit our policies pages at http://www.sciencefriday.com/about/policies/

Meet the Producers and Host

About Christie Taylor

Christie Taylor is a producer for Science Friday. Her day involves diligent research, too many phone calls for an introvert, and asking scientists if they have any audio of that narwhal heartbeat.

About Umair Irfan

Umair Irfan is a staff writer for Vox, based in Washington, DC.

Explore More

Seven Questions About How Your Dog’s Brain Works

You asked your canine cognition curiosities and a neuroscientist answered.

Read More