While everyone crowded around the computer, I sank into a chair, utterly exhausted. I hadn’t realized how intense the last several hours had been. But now the adrenaline that had kept me going drained away, and I crashed. Same for Callie. She had already made her way to Melissa’s pup tent for some quiet time and was sacked out.
But we weren’t done. Now it was McKenzie’s turn.
Callie had forged the way. Based on what we had learned about the noise and the earmuffs, we wouldn’t have to waste any time with McKenzie.
Rebeccah worked her magic with the earmuffs. While Callie wore the small size, McKenzie had to wear mediums. Fully wrapped, McKenzie looked like she was wearing a turban.
Because every dog is a different size and shape, the scanner would again need to go through the shimming and localizer sequence for McKenzie. Melissa and Mark got her settled in the scanner and gave a thumbs-up to start the scan.
McKenzie reacted the same way Callie had. As soon as the buzzing started, she scooted out of the magnet. We did this three, four times, and despite the earmuffs, McKenzie was not having any part of it.
“What do you think we should do?” I asked Mark.
“The problem seems to be the sudden onset,” he said. “The dogs are comfortable in the magnet when it’s quiet, but the scanner starts without warning and scares them.”
I turned to Sinyeob: “Can we start the scan before the dog is in there, and then get her to go in while it’s running?”
He shook his head. “No, the scanner won’t run with nothing inside.”
“Maybe we could mask the transition,” Mark suggested. “I’ll make some noise before the scan starts to distract McKenzie.”
Now on the fifth try, Melissa once again coaxed McKenzie into the MRI. Mark started whooping and hollering at her. Then the shimming began. Maybe it was Mark’s carrying on, or maybe she had finally gotten used to it, but McKenzie stayed put. At least until the klaxon of the localizer began.
She’d been so close.
“Did it complete the shim?”
“Yes,” Sinyeob said, “but she moved before the localizer.” It was almost five o’clock, and we were just about out of time.
“Let’s skip the localizer and go right to the functionals,” I said. “The chin rest should put her head in the same location as Callie’s. We’ll just use the same orientation and field of view that we used on Callie.”
Andrew took up his position at the rear of the scanner and prepared to cue Melissa on peas and hot dogs. Mark started carrying on, making a ruckus to distract McKenzie from the sudden onset of the functional scan. Robert hit the start button.
I fully expected to see McKenzie’s butt start backing out of the magnet. But she didn’t. Mark stopped hollering. Images started appearing on the scanner console. At first, a nose poked into the field of view. Then part of a brain. And a little more. And then it would disappear, only to reemerge a few seconds later.
McKenzie was staying in the scanner. Melissa was putting up hand signals and feeding peas and hot dogs. From outward appearances, McKenzie was doing even better with this part of the scan than Callie had. The images popping up on the screen clearly showed McKenzie’s brain, and they weren’t moving, which meant that she was holding her head perfectly still.
The only problem was that her head was on the edge of the field of view. Even though she wasn’t moving, we were capturing only the front half of her brain. This was a direct consequence of setting the field of view without a localizer image. We’d shot blind and missed by an inch.
I let the functional sequence run for the full five minutes. Even if we wouldn’t be able to use her data this time, it was good training for Melissa and McKenzie. When it was done, I gave them the report.
“The good news is that McKenzie held her head still,” I said. “The bad news is that we got only half her brain.”
“McKenzie’s not too tired,” Melissa said. “We could try again.”
“If you could get her to scoot her head forward an inch,” I said, “that would help.”
Mark, Melissa, I, Sinyeob, Robert, Lisa, and Kristina study the first functional images.
(Bryan Meltz)
Once again, everyone took their positions, and with McKenzie resettled in the magnet, we went through the protocol for what seemed like the hundredth time that day. This time, her head was closer to the center of the field of view. Some images were still clipped, but overall, the run looked very good.
Between Callie and McKenzie, we had exceeded our goal of acquiring a sequence of ten functional images. Even if we had only a partial scan of McKenzie, we had almost one hundred images of both dogs—enough to do a crude analysis of brain activation comparing hand signals for peas and hot dogs.
When we got home, Callie ran right to the kitchen. Even though she had spent the entire afternoon consuming peas and hot dogs, her appetite had not diminished in the slightest. She stood expectantly next to her food bowl. So, where’s my dinner?
Lyra detected the foreign smells of the hospital and sniffed Callie from head to tail. Satisfied that it was still Callie, Lyra wagged her tail and let out a few yippy barks of recognition.
Helen plopped down on the sofa.
“So,” I said, “what did you think?”
“It was pretty cool.”
“Science doesn’t always go the way they teach it at school,” I said. “You never know what will happen.” I paused and continued. “I’m really glad you came today. It was fun having you there.”
Helen nodded her head, and I gave her a hug.
16
A New World
THAT NIGHT, CALLIE CURLED UP in her usual spot on the bed. Exhausted from the day of scanning, she immediately fell into a deep sleep, snoring softly. But I was still too jacked up; for the second night in a row, I didn’t sleep. Images of Callie’s brain danced in my head. But I had no idea what we had actually done. Dog brains were going to be our new world, but we had no map.
It turned out that the dogs’ brains looked nothing like a human’s brain. Apart from the size difference, many of the landmarks that I had become accustomed to seeing in human brains were either absent in the dogs or distorted into unrecognizable shapes. Now that we had dog brain scans, Andrew and I would have to grapple with interpreting all that data.
Dog brains and human brains differ in two important ways: structure and function. Brain structure refers to its shape. You don’t need to be a neuroscientist to see that humans and dogs have brains of different shapes. The brain structure consists of the different parts of the brain and their location relative to one another. This is why neuroscientists refer to prominent parts of the brain as landmarks. Obvious landmarks in all brains include the brainstem and spinal cord, the cerebellum, the ventricles (which make the cerebrospinal fluid), the corpus callosum (which connects the left and right sides of the brain), and a few structures in the basal ganglia—part of MacLean’s reptilian brain.
The dog brain (left) and the human brain (right). (Not to scale.)
(Dog brain image by permission of Thomas Fletcher, University of Minnesota; human brain by Gregory Berns)
But even with these landmarks, the largest part of the brain—the cerebral cortex—is radically different in dogs and humans. Presumably, that is what makes us different from each other.
Imagine comparing a map of the United States to a map of France. What can you deduce about these countries from looking at their maps? There is an obvious difference in size, but that doesn’t say much. Based on the arrangement of roads, the maps would give you a sense of where hubs of activity lie. Many roads lead to Paris, and you might correctly conclude that Paris is a key center in France. You would also notice important port cities like Marseille and Bordeaux and guess that these cities are centers of trade. In comparison, there is no obvious center of activity in the United States, but the road map would give you much of the same kind of information. The Northeast Corridor, from Washington, DC, to New York, immediately stands out, and yo
u would be correct in assuming that this region is a critical center of government and economic activity. Similarly, coastal cities like Boston, Houston, Los Angeles, and San Francisco would stand out as centers of trade.
And so it is with the brains of dogs and humans. Even though we don’t know the exact function of different parts of the canine brain, we can make educated guesses based on what we have learned about other brains. Using landmarks that are common to both brains, we can begin to construct a more precise functional map of the dog brain.
But where to begin?
At first glance, dog brains didn’t look much like human brains at all, so it wasn’t apparent how much of the vast human neuroscience literature we could use. As I lay awake, I pictured the basic divisions of the human brain and tried to imagine how these might look in a dog’s brain. It was very much like looking at a map of a foreign country.
If we think of the brain as a gigantic computer, information goes in, the brain does something with it, and an action is produced, often in the form of movement. In this manner, inputs and outputs form the first great divide in the brain.
Inputs are relatively easy to understand. All information that flows into our brains must come through the five senses: vision, hearing, touch, smell, and taste. From the scientist’s point of view, inputs can be controlled during an experiment. For the experiment we had just accomplished, we controlled the visual channel by the hand signals we gave, and we controlled smell and taste by giving either peas or hot dogs.
Outputs are also easy to understand, especially if we consider movement as the main output of the brain. The earliest fMRI experiments had human subjects lying in the MRI and tapping their fingers for periods of thirty seconds. When the subjects tapped their fingers, activity in the part of the brain that controlled the hand was plainly visible.
The central sulcus is a groove in the human brain that runs almost vertically down the outside of each hemisphere. Everything behind the central sulcus is broadly concerned with inputs and everything in front with outputs. It is a defining landmark that divides the frontal lobe in front of the groove from the parietal lobe behind. The frontal bank of the central sulcus, it’s important to note, contains the neurons that control movement of all the parts of the body. Toward the bottom of this groove, above the ear, we find neurons that control the hand and mouth, and as we move up toward the crown of the head, we find neurons that control the legs. The neurons found along the sulcus control the opposite side of the body. When you move your right hand, a portion of the left central sulcus will become active, and this can be seen easily with fMRI.
In contrast, the neurons behind the central sulcus respond when the corresponding parts of the body are touched. These are the primary sensory neurons. As you move farther toward the back of the head, the functions of the neurons become multimodal, meaning they integrate the inputs from many senses. At the very back of the head, we find the primary visual area, which receives inputs from the eyes.
Another obvious landmark of the human brain is the protuberance along the sides of the brain, just above the ear. This is the temporal lobe. Sitting directly next to the ear, parts of the temporal lobe are concerned with hearing. Other parts of the temporal lobe, along the inner crease next to the rest of the brain, contain structures critical for memory.
With the dog brain, the first thing you notice is that, apart from being smaller, it has a lot fewer folds. The massive amount of folding in the human brain is the solution that evolved to cram more brain into a small space. If you could flatten out the brain, you would find that all the neurons are contained in a thin sheet just a few millimeters thick. It’s like taking a very large sheet of paper and crumpling it up into a ball. Once crumpled, a very large area can be made to fit in a small space, like the skull.
The different amount of folding in the dog brain means that the usual landmarks, like the central sulcus, don’t exist. We can point to only the front and back of the brain and sort of make out the temporal lobe. The next thing you notice is that the dog doesn’t seem to have much of a frontal lobe at all. This is the area that really distinguishes humans from other primates. Humans have the largest frontal lobes of any animal. Because the frontal lobes of the brain are mostly concerned with outputs—in other words, doing things—we think that this part of the brain expanded in humans to accommodate higher-order cognitive functions. Uniquely human functions that reside in the frontal lobe include language and the related ability to think symbolically; the ability to think abstractly about the future and past, which leads to planning; and the ability to mentalize what other people might be thinking.
Although the dog brain looks, at first glance, like a scaled-down version of the human brain, there is one area that is noticeably larger in the dog. The part of the brain concerned with smell, called the olfactory bulb, is huge in the dog brain. When the dog brain is viewed in the dorsal plane at the level of the eyes, the olfactory bulb looks like a rocket ship. There is no human equivalent of this part of the brain. The dog’s olfactory bulb and the parts of the brain surrounding it compose almost a tenth of the total volume. Obviously, smell is important to dogs, but almost nothing is known about how this part of their brain works. That research would have to wait.
We had achieved the first milestone of success in the Dog Project by acquiring a sequence of functional images in both dogs. Over the next few days, we would match up the images with the timing data from the experiment. If everything worked, we would soon have a picture of the dogs’ brains that showed which parts responded to the signals for peas and hot dogs.
Dorsal plane view of the dog brain showing the olfactory bulb (left) and the corresponding view of the human brain (right). The arrows point to the caudate in both brains.
(Dog brain image by permission of Thomas Fletcher, University of Minnesota; human brain by Gregory Berns)
But what would that tell us?
The whole of the Dog Project hinged on the promise of figuring out what dogs think. Even if we succeeded in finding the parts of the brain that responded to different hand signals, that wouldn’t necessarily mean that we knew what the dogs were thinking. To answer this deeper question, we would have to interpret the patterns of activation based on similar patterns in humans. If we saw activity in parts of the dog brain that we could identify, and we knew what those parts did in humans, we could begin to build a functional map of the canine brain. Using the concept of homology, we could infer canine thought processes from their human equivalents.
This was a shaky premise.
In recent years, there has been a bit of a scientific backlash against neuroimaging. Functional MRI has made it easy to dream up poorly controlled experiments and have groups of undergraduates go into the scanner. Many scientists, eager to get a quick publication in a high-profile journal, overinterpreted the patterns of activity they found in the human brain. It became commonplace to point to activity in a particular brain region and interpret that as evidence for a particular emotion or other cognitive function. It was too easy to observe activation of a structure and conclude, for example, that the person was feeling happy or sad or fearful or some other emotional state based on the scientist’s assumptions of what different brain regions did. Eventually, neuroscientists termed this type of reasoning reverse inference, and it became a key factor in rejecting many fMRI papers.
I had always felt that the criticism of reverse inference, usually uttered with the same contempt one would have for a bag of doo-doo, was overblown. I wouldn’t fault scientists for overinterpreting their data. If I doubted their conclusions, I could always look at their results and draw my own inferences. If I didn’t believe their results, I wouldn’t cite them in my papers. Good and valid conclusions stand the test of time, while false ones fade into obscurity and are eventually forgotten.
The Dog Project would not only be relying on reverse inference, it would depend on reverse inference of a dog’s brain as if it were a human’s. Interspecies reverse inference.
I could already imagine what my colleagues would say about this.
Fortunately, Andrew and I had decided to stick with what we knew—the reward system. Our task of deciphering function in the dog brain was going to be a lot easier. Unlike the cortex, with its labyrinthine folds, the reward system belongs to the evolutionarily older reptilian part of the brain. The heart of the reward system is the caudate. Because it is so ancient, all mammals have a caudate, and lucky for us, it looks pretty much the same in dogs and humans.
While neuroscientists can quibble about reverse inference in the cortex, when we did an analysis of reverse inference in the caudate, we found that activity in this region is almost always associated with the expectation of something good. As long as we stuck to the caudate, we would be safe in interpreting activity in this part of the dog’s brain as being a signal of a positive feeling. Everything else we found would have to be interpreted with caution.
Even if we limited ourselves to simple questions of whether the dog had positive feelings based on caudate activation, we could still accomplish a lot with brain imaging. No longer would we be stuck interpreting dogs’ behavior based on tail wagging, which is an imperfect indicator of the emotional state of a dog. Dogs wag their tails when they’re happy, when they’re anxious, or when they’re unsure of what else to do. I still wanted to know if our dogs reciprocated our love for them in any way. And although love is a complicated human emotion, the positive aspects of it have been consistently associated with caudate activation.
The first experiment was a proof of concept. Before we could move on to complicated questions, like love, we first had to demonstrate that we could measure caudate activity in the dog. But that wouldn’t be enough. We would have to show that we could interpret that activity in terms of how much the dogs liked something. Because hot dogs are so much better than peas, especially to a dog, the hand signal for hot dogs should cause more caudate activity than the signal for peas.
How Dogs Love Us: A Neuroscientist and His Adopted Dog Decode the Canine Brain Page 14