Did humans truly domesticate dogs? Canine history is more of a mystery than you think.

A black-and-white Boston terrier named Chevy, as sleek and dapper as a seal in a tuxedo, trots crisply into the soundproof testing room. His jaunty confidence will fade quickly as a team of researchers subjects him to a series of psychological experiments that will daunt, dismay, and ultimately baffle him. Poor Chevy is about to be gaslit for the sake of science.

This spiffy little terrier is volunteer number one on day number one of an ambitious project launched by Harvard University evolutionary neuroscientist Erin Hecht to answer basic questions about what dogs do and why they do it. She plans to collect data on the psychology and behavior of hundreds of them across all breeds over many years: how easily they make friends, how well they behave, how they feel about vacuum cleaners. Four video cameras document Chevy’s reactions to an experimenter’s precisely scripted maneuvers. From a reception room next door, the rest of Hecht’s team watches through a one-way mirror.

After some preliminary scratches and pats, Harvard undergraduate Hanna McCuistion gives Chevy a few treats, then places the next one under a glass jar. He sniffs eagerly at it, then gazes beseechingly at her, cocking his head back and forth, turning up his dials to maximum cute. A classic move, Hecht explains: Faced with a difficult situation, a dog quickly turns to a human for help. After 20 seconds, McCuistion lifts the jar for him, and he gobbles up the snack.

A few more simple tests, then she ushers Chevy into a large wire cage and leaves him alone in the room. He fidgets and softly whimpers. Experimenter two, Stacy Jo, soon enters, but she turns away, facing the wall for a few long moments while Chevy stares fixedly at her back. Without making eye contact or speaking, she approaches his cage and sits precisely 1 foot in front of the door, eyes on his chest. Chevy stands stock-still, ears perked, trembling slightly. Nonscientifically speaking, this dog is completely weirded out. From the other side of the mirror, the scene is both agonizing and hilarious, like the world’s most awkward date. Heroically, Jo keeps a straight face.

The data from these tests—plus DNA samples—will ultimately give Hecht new hints about what changed in dogs after their wild leap into tameness. Biologically, they are almost all wolf; technically, they’re the subspecies Canis lupus familiaris, but they are fundamentally different from their forebears. You can hand-raise a wild animal to be tame, and that individual might be gentle and mild-mannered. But domestication is a different story. For dogs and other animals who live with us, tolerance and trust are engraved in their genes and in their brains.

Hecht’s study is a way to get insight into the broader subject of how neural matter evolves under strong environmental pressures—in this case, the very peculiar circumstances of living with, depending on, and loving another species. “I’m interested in dogs, both for the sake of dogs, and for what we can learn about humans,” she says. “But more generally, dogs are a great way to understand basic processes about how brains evolve.”

She is among a wave of investigators puzzling out exactly how these furballs got to be our face-​licking, tail-wagging, number-one fans. We prefer to think that humans wrote the story of domestication: Some galaxy-brain hunter-​gatherer kidnapped a wolf puppy, then shaped a new species as a prey-sniffing partner, watchdog, and companion. But increasingly, most researchers think that dogs were the original authors of this tale. Long ago, some wolves hitched their destiny to ours, launching an extraordinary love affair that forever entangled both our fates.

Though archaeology can help us pin down the when and where of dog domestication (current thinking is that it happened at least 15,000 years ago in Europe, Asia, or both), bones are mostly silent on the how and the why of this story. By studying other canids like foxes and wolves, and by analyzing dog genes, behavior, and brains—their sweet, friendly, trusting brains—researchers are developing new ideas about how the big bad wolf became the dear little dog. Some argue that their social intelligence is what makes them extraordinary; others point to their devotion, that deep soulful craving for humans.

As the first domesticated species, dogs are also a model for how other mammals—including us—got that way. Scientists see in their genes and minds hints about our own unusually tolerant nature. During much of the human journey from just another primate to world-conquering hominid, our four-legged pals have been right by our side. They are our familiar, our echo, our shadow, and as we now look more closely into their eyes, we can glimpse a new image of ourselves.

One night in 2011, Hecht and her miniature Australian shepherd, Lefty, were on the couch watching TV when a show came on about the legendary Belyaev foxes. Dmitry Belyaev was a Soviet geneticist in the early 1950s, a time when Moscow suppressed genetic research as a product of the imperialist West.

Unable to study his chosen field openly, Belyaev hit upon an ingenious plan. He could experimentally tame foxes raised for their coats. Since animals kept by humans tend to reproduce more frequently, officially he’d be accelerating Soviet fur production. But the project would sneak in some science. His theory was that just by breeding for tameness, what’s now called the “domestication syndrome” would emerge: more-juvenile behavior, and physical changes like white splotches on the belly and face, floppy ears, shorter snouts, and smaller teeth.

The research got going in earnest in 1959 in Siberia. Belyaev’s partners selected animals that were simultaneously less fearful and less aggressive (these traits typically go hand in hand), then crossed them. Just four generations later, in 1963, when collaborator Lyudmila Trut approached a fox cage, one of the kits wagged its tail at her. By 1965, a few juveniles were rolling on their backs and whimpering for attention, just like puppies. The researchers also kept a population of randomly bred control animals, and later, a strain of extremely fearful, combative ones. This landmark study continues to this day.

Hecht already knew this history. But the show sparked a realization: Nobody had analyzed the foxes’ brains. Usually, humans breed goats or sheep or other domesticated animals for many traits, including temperament, size, and coat color, all of which might leave inadvertent marks on the mind. But differences between tame and regular fox noggins could be due only to selection on behavior—what Belyaev and Trut did. They’d stand out like a beacon, illuminating exactly which circuits or new neurochemistry turned a cringing, snarling little vixen into a sweetie. And they’d point the way to a deeper understanding of how evolution can remold a mind.

“On the one hand, there’s the basic question of how brains evolve,” Hecht says. “And the more specific question, which is: What are the neural correlates of domestication? Surprisingly, we don’t know.” At least not yet.

Whatever she found could also provide insight into a few emerging theories. One, articulated in 2005 by anthropologist Brian Hare and psychologist Michael Tomasello, proposes that back in the day, some unusually plucky wolves began hanging around humans to scrounge for scraps, giving rise to a less timid subpopulation. Without fear holding them back, these proto-pooches could repurpose their existing social skills to understand and communicate with us. They self-domesticated. That’s the essence of a dog, Hare and Tomasello argue: reduced fearfulness enabling advanced social cognition, that uncanny ability to read our minds. They called the idea “the domestication hypothesis.”

The proof is that pups just get us, without any teaching. Chimpanzees, for instance, struggle to follow a pointing gesture, but most mutts understand it right away. That thing Chevy did—looking to McCuistion to solve his problem—is another example. He intuitively knew how to ask for help.

In the sulci and peduncles of fox brains, Hecht might see signs of whether this theory or others hit the mark. She emailed Trut, who sent a few dozen specimens from recent generations of the Russian foxes, and used MRI to measure the relative size and shape of various structures in their brains.

Hecht saw changes in the parts of the limbic system and prefrontal cortex involved in emotions and social behavior. These data could support the “domestication hypothesis” but don’t rule out other competing ideas either. This initial finding mostly confirms that the brain regions you’d expect to be different are, in fact, different. So, for a finer-grained picture, Harvard postdoc Christina Rogers Flattery is adding another dimension to the analysis, shaving the fox brains into tissue-thin slices and staining them with a dye that reveals their neurochemistry. She’s looking at the pathways of neurons that make the neurohormone vasopressin and at a serotonin subsystem, both of which are linked to aggression. She’s also investigating cells that make oxytocin, which promotes social bonding. There are many possible neural modifications that could lead to tame behavior, such as the boosting of circuits involved in social bonding, or the tamping down of systems that trigger violent attacks. By weaving together Flattery’s investigation with brain scanning, plus genetic insights from a third collaborator, geneticist Anna Kukekova at the University of Illinois at Urbana-Champaign, the group might identify a Grand Unified Brain Theory of Tameness—​or at least its neural-​circuit diagram.

As Chevy responds to his prompts, he’s representing not just himself, but also his breed. While we all have the sense that pit bulls and Pekingese and Irish wolfhounds have distinct personalities and skills, Hecht hopes to pin down those differences. It’s yet another way to explore how selective pressure—in this case, kennel-​club propagation—shapes a brain. In a recent paper, Hecht analyzed MRI scans from 33 breeds, finding that, for instance, a Weimaraner’s noggin has extra terrain devoted to visual processing, and that of a basset hound is primed to analyze smells.

In that same paper, Hecht also looked at a Boston terrier’s brain, which was loaded up with networks related to social activity. Chevy seems to be no exception. Tests all done and DNA sample collected, he bursts into the waiting room, zipping around to greet each person individually, a tiny whirlwind of bliss and joy.

As the little guy gazes into each human’s eyes, little bursts of oxytocin likely erupt in his brain (and in each of our heads as well), findings from a 2015 study suggest. The hormone promotes bonding, which might be why canines are so good as therapy or emotional-support animals for people who have survived trauma.

This swirl of friend-making ecstasy has inspired a rival origin theory that focuses on feelings rather than cognition: “their hearts, not their smarts,” in the words of Clive Wynne, behavioral scientist at Arizona State University. With collaborators Nicole Dorey and Monique Udell at the University of Florida and Oregon State University, respectively, Wynne proposes that the essence of dog identity has to do with emotional connections—love, to use a word rare in science. “It’s kind of obvious, in a sense,” Wynne says. “They’re amazingly affectionate. It’s just been avoided, in part, because it doesn’t sound serious enough to be a topic of investigation.”

The researchers happened upon this line of inquiry in 2008, when they set out to establish further proof for the “domestication hypothesis.” But their head-​to-​head study of dogs and wolves found quite the opposite. Well-socialized wolves from a research institute in Indiana easily followed human pointing gestures, while some shelter dogs who’d had little contact with people did not. (Later studies showed that coyotes and even some hand-reared bats can do it too.)

Another surprise came from a simple test measuring the amount of time each canid hangs around a familiar person. Dogs stick close; wolves—even friendly hand-raised ones—don’t. Dogs, they reasoned, have a unique drive to bond, even with members of another species. Every pup is born with the capacity, including some 750 million stray “village dogs” worldwide. Incidentally, that ability to form interspecies bonds also explains why livestock breeds can be so vigilant guarding sheep or ducks.

More recently, Princeton University evolutionary biologist Bridgett vonHoldt discovered what might be the root of this affection. In the DNA of dogs, she and her team found a marker of evolutionary pressure on chromosome 6. In humans, equivalent mutations cause Williams-Beuren syndrome, a developmental disorder that leads to indiscriminate friendliness, or hypersociability. “I like to think that, in a very positive, adoring fashion, maybe dogs have the canine version of the syndrome,” she says. Here too the change initially arose in them, rather than through something we humans intentionally did.

Exactly how a few gene changes could transform a canid or a human into everyone’s BFF is unclear, and for unknown reasons, the tendency is stronger in some dogs—cough, Labrador retrievers—than others. In one of Hecht’s tests, known as the “empathy task,” experimenter McCuistion pretends to smash her thumb with a hammer, yelping as if in pain. Some animal subjects leap into the person’s lap, licking the faux wound. Chevy pretty much ignores her.

Nonetheless, studies of different kinds of canines raised under identical conditions hint that neither hypersociability nor social-cognition theories like the “domestication hypothesis” answer every question. Starting a decade ago, teams at the Stockholm University and the University of Veterinary Medicine, Vienna’s Wolf Science Center began raising groups of dogs and wolves in the lab. In their first months, both sets of puppies are with people 24 hours a day; after that, the animals live in packs with extensive human companionship.

These experiments indicate that dogs aren’t just wolves with better social skills. For one thing, hand-raised wolves are quite affable; they happily greet their caretakers and will go for walks on a lead. In 2020, the Stockholm team noted, to their surprise, that a few of their puppies intuitively comprehend “fetch” gestures, just like dogs do.

In fact, research out of the Wolf Science Center has found that in some situations, these wild animals are actually more tolerant than dogs: Given food to share, dogs keep their distance from one another. Wolves bicker and snarl at first, then eat peacefully side by side. In one study, pairs of wolves or dogs must cooperate to retrieve a piece of meat; wolves work together effectively, but dogs were “abysmally bad,” says investigator Sarah Marshall-Pescini. When she tested wolf-human and dog-human cooperation partners, the pattern became clearer. Wolves aren’t afraid to take the lead, while dogs hang back and wait for a human to make the first move.

These unexpected findings led Marshall-Pescini toward yet a third theory of self-domestication: Maybe the shift wasn’t a new social skill or expression of love, but rather a novel conflict-management strategy. Humans probably would’ve killed bold, assertive wolves as a threat. But they might have tolerated deferential, avoidant proto-dogs skulking around the camp, hoping for a handout. (Aggressive varieties are probably a recent phenomenon, the result of dog fanciers in the 18th and 19th centuries who created nearly all modern breeds.)

Her group is looking at village dogs to understand more about canine social structure and how they respond to humans. Compared with our pets, these free-ranging animals are probably far more similar to the early dogs that were their long-ago ancestors—some friendly, some shy, all of them in an uneasy, ambivalent relationship with the hairless apes they rely on to survive.

Lurking around the edges of this research, like some wolf sneaking beyond the campfire, is the idea that we too may have domesticated ourselves. That’s one reason Hecht hopes to find a signature of tameness; if she does, she can look for the same pattern in the brains of house cats as compared with wild ones, and in our gray matter in contrast to apes. Anthropologist Hare’s version of this account of human origins, “survival of the friendliest,” posits that just like dogs, we became more trusting and tolerant of one another in our long-ago past, which in turn allowed us to develop superskills in communication—​language is one obvious example.

The notion of human self-domestication has bumped around at least since Darwin’s time, but today there’s actually evidence, points out primatologist Richard Wrangham of Harvard’s Department of Human Evolutionary Biology. In addition to our unusual (for primates) tolerance of strangers and our long adolescence, we show some of the physical traits associated with domestication syndrome. Compared with our hominid relatives, we have shortened faces and smaller teeth. In 2014, Wrangham and collaborators even proposed a possible biological mechanism in neural crest cells, which help shape many of those body parts during embryonic development. The implication, as implausible as it might seem in these times, is that our species evolved to get along peaceably with one another.

In December 2019, a European group found that the gene BAZ1B, located in the Williams-​Beuren region, influences facial shape by directing such cells. It could explain part of the human self-domestication story, Wrangham says.

Back in Hecht’s lab, a new volunteer named Coda runs through his tests. (Coincidentally, he’s also a Boston terrier.) For one task, McCuistion places a treat on the floor, says, “No! Don’t take it!” and then closes her eyes. Dogs know what eyes closed means, so at this point, most snatch the treat. Not Coda. As his owner points out, he’s always a very good boy. He sneaks a look at it, licks his lips, then stares glumly into space, waiting, deferring, and avoiding conflict, as is his dogly destiny.

Over on the other side of the one-way mirror, the humans are entirely absorbed in this drama. “Goooood boy,” someone says. Even after McCuistion finally gives him permission to eat the snack, he still stands there looking sadly at her. A chorus erupts in the waiting room: “C’mon, Coda, take it!” We can all see his desire, feel his restraint. It’s enough to make you wonder who, exactly, evolved to read whose mind.

To look upon a dog, even through a one-way mirror, is to look upon our own species as well—what it takes to live in harmony, to understand one another, to replace fear and aggression with love and loyalty. Perhaps that is why dogs are so thoroughly delightful. They are a living reminder of a better version of ourselves. His afternoon of psychological prodding over, Coda takes the treat and shakes himself. His owner comes into the room, and he leaps up onto her lap, panting happily, staring deep into her eyes as she looks directly into his.

This story appears in the Spring 2020, Origins issue of Popular Science.