Alexandra Horowitz always wanted a dog. But when she and her college boyfriend visited a shelter, she had no idea that the shaggy black puppy they brought home would end up inspiring her career. Even six years later, as a grad student in cognitive science at the University of California at San Diego, she still didn’t guess that the dog greeting her at the door might have more to say than the bonobos and white rhinoceroses she was observing for her degree.
But as Horowitz steeped herself in her studies, the way she looked at her dog, Pumpernickel, changed. Where she once simply saw a pooch at play, she began spotting glimmers of sophisticated behavior. Before long, she was bringing a video camera to the dog park and watching the footage in slow motion.
“It took a real change of perspective to consider studying dogs,” Horowitz says. They seemed so familiar—happy creatures that chased balls and inhaled kibble—what else was there to say? As Horowitz has since discovered, a lot. A decade and a half later, Horowitz heads up one of the premier canine-studies labs in the country, helping scientists and pet owners alike uncover new truths about man’s best friend.
Why do dogs wag their tails? Sniff one another’s butts? Save children from drowning? The answers aren’t what you think.
Until she starts casually dropping terms like “efficacy” and “cognitive understanding” into conversation, Horowitz might not strike you as a scientist. For starters, she never wears a lab coat. “I’m usually covered in a fine mist of dog hair,” she says, makeup-free and dressed in jeans.
And at the Horowitz Dog Cognition Lab at New York’s Barnard College, jeans do seem the natural choice. Her lab doesn’t have a set location. Instead, she tows a camera to dog parks—or pet owners' homes or anywhere dogs are—where she catches dogs playing, fetching, fighting, and mating on video. Horowitz’s mobile office has the benefit of being cost-effective, but its real asset is scientific: Observing dogs in the field is her lab’s signature strength. Rather than drag the animals into unfamiliar settings, Horowitz watches dogs being dogs on their own turf.
“My goal is to get inside the mind of the dog,” Horowitz explains. “They can’t tell us how to treat them; we just decide. I think that decision should be informed by what their experiences are actually like.”
Still, not everyone finds merit in Horowitz’s work. When she first switched her focus to dogs, her decision was met with head-scratching and even outright derision. For her colleagues, “there were no data in dogs.” So she set out to find her own way. While she didn’t know it at the time, Horowitz was one of a handful of scientists around the world blazing trails into the dark continent of the canine mind. As years passed, studies on canine cognition began cropping up in scientific journals. Next came labs devoted to dog behavior. "Just a few years ago, [studying canines] felt like a novelty," Horowitz says. "Now it feels mainstream." In 2009, Horowitz opened her lab at Barnard. Today, she has a full-time researcher and 10 undergrads assisting. And what they’re learning, as they piece their data together in conference rooms and coffee shops, is that for centuries now humans have gotten dogs all wrong.
It begins with the "guilty" look. You know the one. At some point, every dog owner has come home to new couch cushions or shoes or drapes that have been reduced to confetti. Lurking over the demolition work is a pup bearing his most pathetic face: head slung low, ears pinned back, eyes big and wet, emanating guilt. The common assumption is that the dog is genuinely remorseful. But is it true?
To find out, Horowitz ran an experiment where she asked dog owners to place a treat in front of their dogs, instruct them not to eat it, then leave the room. If, in the owner’s absence, dogs ate the treat, their owners scolded them when they returned a few seconds later. But in some of the trials, Horowitz had owners scold their dogs even when they hadn’t eaten the treat. In others, certain treat-eating dogs got off scot-free.
By videotaping the interactions and scoring how guilty the dogs looked, Horowitz discovered something interesting: Even the innocent dogs looked guilty when scolded. Meanwhile, treat-eating dogs who weren’t scolded didn’t look remorseful at all. A dog’s “guilty face,” in other words, didn’t spring from guilt; it appears to be a reaction to the owner’s finger-wagging, a ploy to get off easy. “Dogs may feel guilt,” Horowitz clarifies. “Science hasn’t figured out how to test that yet, but this behavior isn’t evidence of it.”
Of course, Horowitz the scientist and Horowitz the pet owner aren’t always on the same page. As she told the pet enthusiast site Dogtime.com, “I sometimes tell people to try to forget everything they know about the dog, and pretend it is an alien animal arrived in their home: What is this alien doing?”
Talking about her own dog, Horowitz recognizes that he looks proud when he’s run off with a ball or sheepish when he gets carried away roughhousing. “Those looks are real,” she says. “But I remain agnostic about whether they map to emotional experiences that are just like mine.”
Horowitz also worries about the greater dangers of anthropomorphism, such as relying on your pet’s secret identity as SuperDog. While newspapers are quick to print stories of doggy heroes—devoted animals rescuing people from burning buildings or raging rivers—in reality, their motives may be less pure. In one study by other canine researchers, dogs witnessed a staged crisis—their owners pinned under a bookcase made of lightweight particleboard, though the dogs didn’t know that. But the dogs didn’t leap to the rescue. Instead of dusting off their Lassie routines, the majority of pooches ignored their owners’ cries for help.
The conclusion? Dogs can be trained to rescue people or may even do so on their own, but that doesn’t mean that they know what they’re doing. Instead, their behavior may be something simpler, like a desire to be near their owner or to bark when distressed—still a useful skill in attracting help but not the heroism we attribute to them.
While feel-good stories about dogs rescuing people will no doubt continue to surface, Horowitz believes the “evidence” ignores a basic statistical fact: “What of all the cases when a dog didn’t save the drowning child or the lost hiker? Newspaper headlines never crow ‘Lost woman dies after dog fails to find and drag her to safety,’ ” says Horowitz.
Does it matter whether a dog’s feelings are real or imagined? If an experiment proves that dogs don’t love us in the same way we love them, do we even want to know?
It may require some mental adjustments, but Leslie Irvine, associate professor of sociology at the University of Colorado at Boulder, believes so. “By unpacking how dogs experience the world, it can help people interact with them in ways they can understand,” says Irvine. And that can make for “a more compassionate relationship.” In the past, for instance, trainers used to think it was best to yank dogs around by choke collars and rub their noses in their feces if they relieved themselves indoors. But studies on reinforcement have curbed these practices by proving that they don’t work.
Horowitz herself makes a strong effort to debrief the owners of the dogs she works with, and many come away with a new understanding of the animals living under their roof. Just ask Jo Anne Basinger, who’d enlisted her two dogs for experiments ranging from which scents dog dislike (lavender, in particular) to whether dogs can sense fairness in people.
“One thing that I’ve realized is that the things dogs do that annoy me are important to them—like excessive sniffing,” Basinger says. In fact, Horowitz’s research suggests that sniffing is not just important; it’s the crux of how dogs perceive the world. Humans see first, dogs smell first. Even their sense of time in some way comes down to their noses, as older smells fade and hints of smells to come arrive on the wind. And of course, the canine habit of sniffing new friends from behind makes more sense once owners learn that canine anal glands emit a cocktail of chemicals as unique as a human voice, which may indicate a dog’s age, their interest in mating, and what they ate for dinner. Sniffing, in other words, is the doggy version of small talk.
Despite the inroads she’s made into the canine mind, Horowitz savors the mystique. “If I woke up and my dog said to me, ‘Alexandra, I’m going to tell you the whole business right now,’ I would hesitate,” she says. “I appreciate a dog’s quietness. There is something about the dog’s mystery that I treasure.”