When we adopted a tiny black and white pup in February, we couldn’t even guess what breed it was. She was one of more than 100 dogs rescued from a hoarding situation, and she arrived in foster care with no siblings or mother. We immediately ordered a dog dna test and I learned that our little Gabriela was 81% terrier – a mix of chihuahua, toy fox terrier and dachshund. She also had a bit of a beagle and a Boston terrier.
Gabriela’s personality began to shine when she realized she was in her forever home. She is now making little woo-woo sounds to get our attention. Is it, we wonder, because she’s a four percent beagle? And she loves chasing squirrels from our yard and then holding her paw up so our two other dogs are alerted to the prey. Is it because she’s mostly terrier?
Scientists now have a better understanding of how breed types influence behavioral traits. In a new study, a team of researchers from the Broad Institute of MIT and Harvard combined thousands of dog DNA tests with surveys of pet parents to determine which behaviors were strongly associated with breeds. The results showed that we should rethink race stereotypes. Although researchers can attach certain traits to specific breeds, dogs are much more individual than previously thought.
From OCD to DNA
The Broad Institute team started studying dog traits not because they were interested in woo-woo sounds or paw pointing behavior.
“We study compulsive disorders, which dogs suffer from,” explains Elinor Karlsson, director of vertebrate genomics at the Broad Institute. “We were trying to understand the genetics of compulsive behaviors in dogs, and that might tell us something about OCD in humans.”
Karlsson Foundation Darwin’s Ark, a community science project in which people could register their dogs and share genetic data. The project was able to collect data from a diverse population, including purebred dogs with papers and shelter dogs with no known pedigree.
The team interviewed 18,385 pet parents and then sequenced the DNA of 2,155 dogs. They analyzed the data and looked for patterns of behavior. They divided behavioral traits into two main categories. The first – “more hereditary” – involved behavior considered breed-oriented, such as “biddability,” or a dog’s willingness to respond to commands.
The second, “less hereditary” category was for traits they did not associate with breed type, such as fear of thunderstorms. The results did not live up to the expectations of Karlsson or his team.
Races and behaviors
The study found that certain behaviors are indeed associated with race type. But these behaviors accounted for only nine percent of behavioral variation. Retrieval was the most inherited behavior trait.
For the submissive trait, the team found that breeds like Belgian Malinois, Vizsla, and Border Collie were most likely to respond to human direction in training scenarios. In contrast, the Basset Hound, Alaskan Malamute, and Shiba Inu were most likely considered “independent.”
The results meant that for some breeds, retrieving, submissive, or howling was a predictable, but not absolute, trait.
“Owners told us beagles were more likely to howl. We totally saw it,” says Karlsson. “Dogs that had beagle, bloodhound and husky ancestry were more likely to howl.”
The team concluded that breed type was only a moderate predictor that a dog might perform a certain behavior. A border collie puppy, for example, could indeed be intelligent and easy to train. But he may also not be eager to please, which can make it harder to learn his commands.
Rankings that claim certain breeds are “most stubborn” or “smartest” or “best for families” are based on generalizations, not data. And the study results demonstrate that most dog behaviors are not consistently observed among a breed type to support such generalizations.
“Having a dog will never be like shopping in a catalog,” says Karlsson. “If you get a beagle, he’s more likely to howl than a Labrador. But that doesn’t mean you can’t get a Labrador that doesn’t howl.
With certain traits, genetics influence the behavior of the dog. However, the study found consistent patterns, not absolutes. Dog breeds like golden retrievers, huskies, Labrador retrievers and pit bulls were more likely to be social compared to a randomly selected dog in the sample. But the study also found wide variation between races. So while Golden Retrievers are more likely to be social, it’s possible that some are introverted.
Behaviors such as chasing, pointing and howling are more genetically coded, and Karlsson says his team hopes to one day identify behavioral traits at the chromosomal level. Such nuanced tests would one day determine if little Gabriela’s woo-woos are indeed the result of her four percent beagle ancestry.
“I would like to go through the chromosomes to see […] behavior and see where it comes from,” says Karlsson. “If we looked at that part of her DNA, would we see that she had ancestry from a howling dog? We don’t know how to do that.
Until then, Karlsson says all pet owners need to see their dog as an individual and give up trying to explain the behavior through breed stereotypes. The woo-woos, for example, could be related to the beagle. Or they could just be gabi-ism.
Read more: Our dogs are manipulating us, according to science