How Dogs Love Us: A Neuroscientist and His Adopted Dog Decode the Canine Brain (25 page)

Callie drifted off to sleep. Pretty soon I could feel her legs twitching as she started to dream. I pondered the possibilities of using fMRI to see what was happening in her brain while she dreamed.

My reverie was snapped by the
thwack-thwack-thwack
of her tail on the sofa. She was still dreaming.

Maybe she was dreaming of taking down one of those coyotes. Or maybe catching a tasty rodent in the yard. Or maybe it was just the dogness of being there, in my lap.

And if that wasn’t love, then I would surely accept it as a reasonable facsimile.

21

What’s That Smell?

T
HE RESULTS OF THE HOT DOG
experiment made me wonder what the dogs thought of us humans. It seemed like there was more going on than just love of hot dogs. Each trip to the MRI, Callie got more and more excited. By the final scan session, she was making a beeline to the portable steps up to the patient table. She would shimmy into the scanner bore even before we had her chin rest in place. Her look said,
I’m ready, let’s go!
She liked interacting with all the people, and, as everyone agreed, she liked showing off. She had become a diva.

Callie had also gotten used to McKenzie. If I had to characterize their relationship, I would call it one of mutual nonthreatening coexistence. We used the lab as our staging area before walking everyone across the campus quad to the MRI facility. Callie and McKenzie would greet each other in the lab with a perfunctory butt sniff and tail wag. That would usually be it, as both dogs preferred checking out the humans in the room. Once at the scanner, though, Callie would start getting more excited. If it was McKenzie’s turn to go into the MRI, Callie would climb up on the patient table and try to get into the MRI before McKenzie. Callie would have to be carried off the table and kept in the control room while Melissa and McKenzie got situated.

I found this behavior fascinating. It seemed clear to me that the dogs treated each other differently from the way they treated us humans. This is despite the popular notion that we humans are a “pack” to the dogs—a sort of extended doglike family.

This gave me an idea for another fMRI experiment.

How do dogs categorize humans? Either dogs have separate categories for dogs and humans, or they lump us together as either pack or not pack.

To my eyes, Callie and Lyra behaved like pack mates. They ate together. They slept together. And they played together. It was no different from what we humans in the house did with them. And while we viewed them as family members, it would be nice to know if they viewed us that way too. Because Callie and Lyra were unrelated, and they were obviously unrelated to us humans, the notion of a pack would have to be what anthropologists call
fictive kin
.

Humans are particularly good at treating genetically unrelated friends as if they were family, especially if they go through an intense experience together. This is why soldiers call each other “brother.” If people do this, maybe dogs did too. If dogs viewed their humans as part of their pack—a sort of extended family—then dogs and humans should result in similar activation in the dogs’ brains.

So what might distinguish humans and dogs—at least in the dog’s mind? Apart from appearance, the most obvious is smell. After a dog sees another dog, it will make a visual assessment of body language, like how the tail is held, and decide whether to approach. If it does approach, then they will sniff each other. It is similar when dogs see humans. After a visual assessment, a dog will usually approach and scent the person.

A dog’s sense of smell is about one hundred thousand times as sensitive as that of a human. They also have an additional structure, a fluid-filled tube called the
vomeronasal organ
(VNO), thought to be specialized for detecting scents from other dogs and therefore to function in some capacity for social signaling.

With that powerful of a sense, you can be sure that a large portion of the dog’s brain is devoted to processing smells. Even so, I was still shocked when we got the first images of Callie’s and McKenzie’s brains. Where we would normally see a big frontal lobe in humans, the dogs had almost nothing. Instead, extending toward their snout, was a massive phallic protuberance—the olfactory bulb. A rocket in the socket. Humans have nothing like that. And it accounted for about 10 percent of the dog’s entire brain.

We usually think of smell as one of the five senses and a generally passive process. Odorants drift into our noses; receptors detect them and send signals to our brains. However, more so than vision or hearing, smell is an active process involving many groups of muscles. Animals can control the rate at which odors enter the nose by the way they sniff. Sniffing involves muscle movements in the face and the nose. It requires movements of the diaphragm to control the rate of air movement. And there is likely some control over the fine hairs inside the nose. This means that for smell in particular, we would also expect to see the involvement of parts of the brain that control movement.

If the scent of a dog activated the brain in the same pattern as the scent of a human, then that would tell us that dogs lumped us in the same category as them. If, on the other hand, dog and human scents caused different patterns of activation, then we would know that dogs have different categories for us and them.

Like the hot dog experiment, the dogs wouldn’t have to do anything except hold their heads still, and they were already pros at that. We would hold up cotton swabs in front of the dogs and let the scent drift into their noses. Later, we could analyze the fMRI data to see which parts of their brains reacted to different scents.

This experiment presented certain logistical complications. Where would we get the scents and how would we get them? These questions became the topic of heated debate in the lab, especially as it became apparent that everyone would have to give something for the cause.

“So let me get this straight,” Andrew said. “We are going to present scents from dogs and humans to Callie and McKenzie.”

“That’s right,” I said.

“What kind of scents?”

“Well,” I said, “we all know what dogs do when they greet each other.”

Andrew didn’t like where this was going. “You’re suggesting a butt wiping?”

“I don’t think there is any other way.”

Lisa chimed in and offered an alternative: “Dogs sweat from their paws. You could get scents from there.”

“But we would also pick up all sorts of smells from where the dogs walked,” I said. “Besides, dogs go right to the butt. As far as they’re concerned, that is where the good stuff is.”

“Are we talking about a butt wiping or something more substantial?” asked Andrew.

It was a good question. A swabbing of the perianal area would probably do the trick, but there was good evidence that urine would be a more powerful signal. Dogs can differentiate their own urine marks from those of other dogs, suggesting that dog urine contains unique pheromones that are the equivalent of doggie fingerprints.

“I think we need urine,” I said.

“What about the humans?” Andrew asked.

“If you’re doing it for the dogs,” Lisa said, “I think you should do it for the humans.”

Andrew and I looked at her aghast.

“What?” she said. “The first thing Sheriff does is stick his nose in someone’s crotch.”

Although Lisa had a point, there were some boundaries that we couldn’t cross. Besides, what she was suggesting could be construed as biohazardous waste by the university lawyers.

“How about a good sweat sample from the humans?” I offered. “As long as they don’t wear deodorant, we could have people do a workout and wipe their armpits with a gauze pad.”

There was reluctant agreement with this plan, but this immediately raised the issue of who the canine and human “donors” would be.

The pack versus not-pack question became one of familiarity. To Callie’s nose, all the scents in the house were familiar to her: me, Kat, Helen, Maddy, Lyra, and even Callie’s own scent. This was her dog-human pack. Melissa and I would be at the scanner already, and our scents would pervade the scanner, setting a backdrop against which other smells could be measured. Ideally, we needed scents from other people in our households to serve as the “familiar human.” It would have to be Kat’s sweat for Callie and Melissa’s husband’s for McKenzie.

We would also need a comparison for not-pack, or unfamiliar, scents. We would need scents from strange dogs and strange humans. We drew a chart on the lab wall and listed all the humans in the lab, along with their dogs, and cross-referenced them against whether they had met Callie or McKenzie. Andrew’s American Eskimo, Mochi, had never been to the lab. She quickly emerged as the leading contender for “strange dog.” Plus, she urinated whenever she got excited, and Andrew would have no problem getting a urine sample from her. Since Callie and McKenzie had met all the humans in the lab, we still needed some “strange humans.” Considerable discussion ensued about the logistics of getting fresh sweat on scan day, as
well as the possibility that the dogs had already been exposed to the smells of spouses, girlfriends, and boyfriends inadvertently as scents carried on lab members or, as the cops say, “on their person.”

In the end, I convinced a neighbor to donate her sweat to be the “strange female” as a control for Kat’s sweat. Kat’s kickboxing coach agreed to donate his sweat to be the “strange male” control for Melissa’s husband.

Timing was critical. Everything depended on getting samples as fresh as possible. For the dogs, that meant morning pee, which, we reasoned, would be the most concentrated of the day. For the humans, they needed to get a good sweat going, which could be mopped up from their armpits. Each of the human donors had been instructed to not shower or wear deodorant for the twenty-four hours prior to sample collection. Everyone was provided with sterile gauze pads, gloves, and a specimen bag to place the sample in.

As usual, the scanner was booked for one p.m. We would need all the samples at the lab by noon so that Andrew, who had volunteered for pee-pad duty, could prepare them for the experiment. He would have to cut the pads into strips with sterile scissors and carefully attach each sample to a six-inch-long cotton swab. Each swab would be numerically coded. This way, neither Melissa nor I would know the identity of the samples and inadvertently cue the dogs during the experiment. Only Andrew would know the code.

That morning, Kat and I took Callie and Lyra for their walk. I trailed the dogs, looking like a crime scene investigator, wearing purple surgical gloves and toting specimen bags. Callie loved to pee on her walks. As soon as she caught the scent of what I presumed was another dog, she would squat and dribble out some urine. She had a peculiar way of doing it, though. Her bottom never quite made contact with the ground. Instead, she sort of hovered and continued
walking, giving the appearance of a duck waddling. Callie never left pee spots. She left pee
trails.

Her urination habits made it easy to collect a pee sample. As Callie tracked a scent, she intensified her sniffing of a location on our neighbor’s lawn. I knew she was about to pee and had my pee pad ready. As soon as she squatted, I thrust the pad beneath her lady parts and was rewarded with a warm yellow stain. Callie looked over her shoulder at me.
Hey! What are you doing back there?

Lyra was more difficult. The deep fur, stained and matted around her butt, appeared less clean than Callie’s. Plus, Lyra peed more conventionally for a female dog: back straight, butt in contact with the ground. The best I could do was wipe her bottom right after she peed. It was enough.

Poor Andrew. We had to lock him in a closet while he cut up all the samples. We couldn’t have the dogs getting whiffs of all those great smells before the experiment. After an hour of cutting up urine and sweat pads, Andrew emerged.

“Are you all right?” I asked.

He waved me off. “I just need to get some air.”

By now, the excitement of parading the dogs across the quad to the hospital had worn off, and only lab members who actually had a job to do on the Dog Project accompanied us. I still got a thrill out of the walk, though.

Like a well-oiled machine, everyone took up their positions at the scanner. Melissa and McKenzie were there, of course, and they relaxed in the control room until it was their turn. Andrew set up a test-tube rack on a plastic worktable at the rear of the magnet. He inserted the cotton swabs, business end down, in each tube.

For each dog, Andrew had prepared seven swabs: the four combinations of strange and familiar humans and dogs, plus an intermediate category of “acquaintances.” Callie and McKenzie were acquaintances. They knew each other, but there was no reason to expect that they viewed each other as part of their pack. We would present their scents to each other for this category. In this fashion, we would have a continuum of familiarity from stranger to acquaintance to household member. Using lab members’ sweat, we created the corresponding human acquaintance category. Finally, for a baseline, we used the dogs’ own urine as a “self” category.

With Callie in the scanner, the shimming and localizer sequence took less than a minute. She knew the routine. For the functional scans, we modified the hot dog experiment. Instead of holding up a hand signal for ten or fifteen seconds, Andrew would hand me a swab, and I would hold it in front of Callie’s nose for a few seconds. She would continue holding still to allow enough time for the hemodynamic response to peak, and then I would reward her with a hot dog treat. It was the same as before except we would insert a smell during the middle of the repetition.

To get her used to a cotton swab being shoved in front of her face, Callie and I had practiced this at home. The first few times she backed away, but she soon realized nothing bad was going to happen and just sniffed.

With the functional scans running, she performed flawlessly. Each swab was presented eight times in random order. It took two functional runs lasting six minutes each, and then we were done. Another four hundred scans in the can.