Read our COVID-19 research and news.

Oszkár Dániel Gáti/Getty Images

Humans haven’t just changed what dogs look like—we’ve altered the very structure of their brains

In the thousands of years we’ve lived with dogs, we’ve transformed them from fearsome wolves to fluffy, tail-wagging Frisbee catchers that range in size from tiny pomeranians to towering great Danes. Now, a new study of dogs’ brain scans suggests our impact on our canine pals has been even more profound: We’ve changed the very structure of their brains.

“This is really exciting new work,” says Daniel Horschler, a comparative psychologist at the University of Arizona in Tucson who has studied the evolution of dog brains but who was not involved with the current work. “Dogs haven’t really been studied in this way before.”

To conduct the research, Erin Hecht, a Harvard University neuroscientist (and the caretaker of two incredibly hyper Australian shepherds), and her colleagues assembled a library of MRI brain scans from 62 purebred dogs from 33 different breeds. As soon as she saw the images lined up next to each other, “You could just see the results staring at you,” she says. The dogs, which included bichon frises, Labrador retrievers, and more, had a variety of head shapes and sizes. But neither of those things alone could explain the variation in the layout of the dogs’ brains.

Hecht and her team identified six networks of brain regions that tended to be bigger or smaller from dog to dog, and that varied in tandem with each other. The pattern led Hecht to think these regions were probably working together in different behaviors. She wondered whether the varying layouts might be due to behavioral differences between breeds. Beagles can sniff out cancerous tumors in humans and let doctors know, for example, and a border collie can herd hundreds of sheep (or even turkeys) into an enclosure with remarkable speed and agility.

Her team looked at how the six networks differed between dogs based on the traits they were bred for, as defined by the American Kennel Club.

Each of the six brain networks correlated with at least one behavioral trait, the researchers report today in the Journal of Neuroscience. Boxers and dobermans—sometimes used as police dogs—showed significant differences from other breeds in the network that was linked to sight and smell, for example. Dogs bred for sport fighting showed changes in the network that represented fear, stress, and anxiety responses.

Hecht was particularly interested in the differences between dogs bred for sight hunting and those that hunt by scent. Dogs that specialized in scent hunting showed differences not in the early regions of the brain that detect smells, but instead in the more sophisticated areas that help the dogs understand and communicate that information, which made sense to Hecht. “I’ve heard trainers that are working with scent hounds say you don’t have to train a dog to be able to smell something,” she says. “You just have to train them to report it.”

One drawback to her study, Hecht says, is that all the dogs examined were pet dogs, not working dogs. “It’s kind of amazing that we can see these differences in their brains even though they’re not actively performing the behaviors.”

She also says her findings may have other implications. The fact that we’re altering the species around us so much that it affects their brain structure is “deeply profound,” Hecht says. “I think it is a call to be responsible about how we’re doing that and how we’re treating the animals that we’ve done it to.”