If you spend enough time strolling along sidewalks or into public parks, eventually you’ll see a dog that bears an eerie resemblance to its owner. The experience is common enough for art to imitate it: recall the famous montage in the movie 101 Dalmatians with those uncanny human-canine couplets. And if for some reason these encounters have escaped you, just take a look at the following pairs of photos showing people and their twin-like pets:
The similarity is–well, pick whatever description you’re most comfortable with, but it’s certainly evident. And it’s evidence-backed, too. In several studies over the past decade, behavioral scientists have found that some people look so much like their pets that outside observers can match them based on pictures alone. The above image, for instance, comes from a 2005 study in which test participants identified owner-pet pairs at a success rate far greater than what you’d get with random guessing. The effect has held in the United States, South America, and Japan, suggesting it just might be universal.
So the resemblance truly exists, according to science. The question then becomes why. Humans do occasionally keep their young ones on leashes, but they don’t actually give birth to pets–or the Internet would surely know about it–so it’s safe to say the similarities aren’t genetic. It’s possible that people and pets somehow grow to look like one another over time, though how exactly that would occur is a bit of a mystery–short of a person telling a barber to give them the Bichon Frise.
Far more likely is that some people, either intentionally or subconsciously, choose a dog that resembles them, says social psychologist Nicholas Christenfeld of the University of California-San Diego. “I’ve certainly heard stories of people coming to resemble their pets,” he tells Co.Design. “It’s not really clear what the mechanism for that would be. I guess you could both exercise together–both catch Frisbees in your mouth, or something. But really coming to look like your dog would pretty much have to be you changing your appearance to resemble the dog, rather than the other way around. So it’s not entirely crazy. But picking a dog that looks like you seems more plausible.”
About a decade ago, Christenfeld and then-graduate student Michael Roy put this question to the test by taking photographs of 45 dogs-owner pairs at local dog parks. They also noted the length of the dog-owner relationship and identified the dogs as either purebreds or non-purebreds. If dogs and people converge in looks over time, length of ownership should be an indicator of resemblance. If people pick dogs that look like them, then being purebred should be an indicator, since its final appearance is much more predictable than that of a mutt.
For the study, Christenfeld and Roy showed test participants three pictures–one person, and two dogs–and asked them to pair up owner and pet. The researchers found no evidence that these outside judges could match owners and non-purebred pets, and they also found no evidence linking matches to length of ownership. But test participants did successfully match 16 purebreds with their owners out of a possible 25 picture pairs, significantly more than chance.
“The results suggest that when people pick a pet, they seek one that, at some level, resembles them, and when they get a purebred, they get what they want,” Christenfeld and Roy concluded in a 2004 issue of Psychological Science. (For the record, the statistical analysis of this study was subsequently challenged but then defended by the researchers, who also replicated the initial findings in a follow-up experiment. Christenfeld says the “quite subtle statistical issue” has been resolved.)
More recent research, conducted by Japanese psychologist Sadahiko Nakajima of Kwansei Gakuin University, suggests it’s the eyes that hold the key to pet-person resemblance. When test participants saw photo pairs of purebred dogs and people, they successfully identified actual owner-pet matches based on the resemblance–even when the mouth regions of the photos were covered with black bars. But when the eye regions were covered, test participants failed to make true pet-owner matches, Nakajima reported in a paper published in late 2013.
What exactly compels people to select a dog that does or will look like them, wittingly or not, is tougher to pinpoint and no doubt varies. In some cases, people might experience what psychologists call the “mere-exposure effect,” a tendency to prefer familiar sights. Hence the woman who sees her long hair in the mirror every day picks out the Afghan Hound. Other cases may come down to simple narcissism.
Christenfeld suspects evolution might have something to do with it. The impulse to care for a child is enormously adaptive in the eyes of natural selection, and one way a child might trigger this caretaking desire–aside from the whole emerging from the womb thing–is by looking like a parent. So when a person sees a “little, helpless, non-verbal creature that looks like them” in the form of a pet, Christenfeld says, some of those same basic nurturing instincts could spring into action.
“The feeling people have about their children is often very similar to the feeling people have about their pets,” he says. “Lots of couples will use pets as a sort of practice trial for kids. And when kids go away to college, they’re often replaced by pets. They come back for spring break and it’s, ‘Sorry, you have to sleep in the garage. Fluffy has your room now.'”