The neuroscientist Gregory Berns, in his new book, “What It’s Like to Be
a Dog: And Other Adventures in Animal Neuroscience
,” considers canine
science from a more methodical perspective. As its title suggests, the
book is a rejoinder to Thomas Nagel’s canonical essay “What Is It
Like to Be a Bat?,” from 1974, in which Nagel argues against material reductionism
in cognitive science. Nagel makes the case that, no matter how
sophisticated our understanding of echolocation, we could never truly
understand a bat’s subjective experience. Berns, who has spent much of
the past five years training dogs to sit patiently in special earmuffs
inside MRI machines and getting them to participate in other
experiments, believes that it is theoretically possible for us to obtain
a partial but meaningful understanding of non-human animal
consciousness—and that, thanks to new neuroimaging techniques, this
understanding is increasingly within reach.

Berns’s argument is simple. Human brain structure is not radically
different from canine brain structure; recent advances have allowed
scientists to definitively link physical activity in the human brain to
“discrete mental states.” It stands to reason that we might apply these
new technologies to understand the cognitive experience of dogs.
“Instead of trying to answer the big question of what it is like to be a
dog, we can be more precise,” he writes. “What is it like for a dog to
experience joy?” And so, in pursuit of the answers, Berns put together a
group of “service-dog washouts,” taught them to stay calm inside MRI
equipment, and mapped their neural activity as the good dogs performed
adjusted versions of classic experiments in children’s cognition—the
marshmallow test, the A-or-B test in object permanence, and so on.

One of the most delightful things about “What It’s Like to Be a Dog” is
the attention Berns pays to each dog’s individual quirks. This comes
from his lifelong dog-owner’s sense that, as he writes, dogs “had
personalities, and they had likes and dislikes, and purposeful behavior
that suggested a higher level of thinking than behaviorist models gave
them credit for.” The awareness of dog personality also governs Berns’s
research. With the “doggie-marshmallow test,” a self-control experiment
that required its subjects to hold off eating a treat for long periods,
Berns’s team quickly figured out that not all dogs would activate the
same cognitive mechanism while performing the same task. Some dogs, like
one named Tug, wanted the treat, and so needed to use self-control to
refrain from eating it. Others, though, such as Kady, a golden retriever
that Berns describes both as “one of the sweetest dogs I had ever met”
and also as “rather vacuous,” showed no sign of wanting the treat. She
would pass the test thanks to either inhibition or a desire to please,
not self-control.

And so, each experiment had to be designed to account for individual
canine decision-making. Some of the research set out to clarify how
exactly that decision-making differed. One experiment asked the dogs to
choose between hot dogs or praise from their owners. Most dogs responded
to hot dogs and praise equally, though not always in the same
sequence—some dogs would seek out a lot of hot dogs and then a lot of
praise; others would alternate evenly. Twenty per cent of dogs chose
praise more often than food, suggesting a surfeit of affection for their
owners. (In another experiment, Berns showed his dogs a series of photos
and video clips while scanning their brains, and found that there was a
discrete region of the brain that lit up when the dogs were processing

Berns’s research is unusual in its emphasis on voluntary participation:
the dogs had to get inside the MRI machine and stay there of their own
(trained) accord. For academic researchers, he wrote, “it was heresy to
allow an animal the choice of submitting or not submitting to a human’s
will. It rejected practices typical of the entire industry of laboratory
animal research, not to mention industrialized farming.” But, as he
reveals at the close of the book, Berns was atoning for an event from
decades ago—a “dog lab” in medical school, where he and his colleagues
monitored the effects of anesthesia on a shelter dog and then gave it a
lethal injection. At the suggestion of a supervisor, Berns severed the
dog’s pulmonary artery to save it ten minutes of slow death. “The lab
didn’t make me a better doctor, and it diminished me as a human being,”
he wrote. “I think now that, by trying to find out what dogs think and
feel, I was trying to make amends.”