Healthy Brain, Happy Life (9 page)

I spent hundreds of hours over six years sitting alone in a darkened room staring at brain tissue under a powerful microscope trying to discern a clear pattern. Some days, after hours and hours of looking in the microscope at the cells that made up these brain regions, the images started dancing in front of my eyes like beautiful abstract pieces of art. It was hard, detailed work. Often, to fill the silence, I listened to classical music. Saturday mornings were my favorite microscope days. Sitting in the lab all alone in the dark with my slice of brain tissue I listened to a radio program called
Adventures in Good Music with Karl Haas
—a wonderful show from which I learned about everything from the mysteries of how Stradivarius made his famous violins to subtleties of the violin passages in Mendelssohn’s symphonies.
Adventures in Good Music
was followed by the Metropolitan Opera’s Saturday matinee, which broadcasts operas in their entirety. I should have gotten an additional PhD in classical music appreciation with all the hours I spent listening to these programs during graduate school. I had no human company during the time I spent in that dark room, but at least I had the music.

What did all this work tell me? It turns out that the cortical areas in the medial temporal lobe I studied, called the perirhinal and parahippocampal cortex, provide massive input into the hippocampus via a structure called the entorhinal cortex. In addition, my studies showed that these cortical areas are a major brain interface, or “gateway,” receiving input from a wide range of brain areas involved in all kinds of sensory functions and other higher-level brain areas important for things like reward, attention, and cognition. Far from being simple visual areas, as researchers had previously thought, these regions are where high-level information converges in the brain. While I used relatively old-fashioned research approaches, my work revealed new information about why these brain areas might be so important for memory. Their
connections
were the key.

But just characterizing the connections of this region could not tell us exactly what its functions are. I went on to show that damage limited to these mystery cortical areas in animals causes devastating memory impairment that is similar in severity to the magnitude of impairment seen in H.M. This was another shocking finding. All the attention on memory research so far had been focused on the hippocampus and the amygdala. These new studies showed that neuroscience had been missing a key player in the game all along—the cortical areas that surround the hippocampus and amygdala. It was also clear that just because we implicated selective cortical areas important for memory, that did not mean that Lashley was vindicated. He had proposed that memory emerges from a complex interaction from widespread cortical areas across the brain and that no single area can underlie memory function. My findings showed that, in fact, you can identify specific and highly interconnected areas critical to the ability to form new long-term memories: specifically, the hippocampus and the cortical areas that immediately surround these structures. While Lashley was wrong about the localization of brain areas important for the formation of new memory, his ideas about the importance of large networks of brain areas did foreshadow findings that long-term memories can be stored in the same widespread cortical networks that process the incoming information in the first place.

My graduate studies helped identify two new brain areas and showed exactly how important they are for long-term memory function. In addition, the studies also pointed to another brain area, sitting between the perirhinal and parahippocampal cortices and the hippocampus, called the entorhinal cortex. Research shows that this area also plays a big part in the system of brain areas critical for declarative memory. Indeed, the recent Nobel Prize in Science or Medicine was awarded to two colleagues from Norway who characterized a major function of the entorhinal cortex in the processing of spatial information.

The U.C. San Diego research team and I hypothesized that patient H.M.’s severe memory impairment had to have been due to the damage both of the hippocampus and of
these surrounding cortical areas.
And sure enough, as soon I completed researching and writing my thesis, a brain scan was taken of H.M., which allowed researchers to visualize for the first time the true extent of his brain damage. This historic MRI scan (this is a technique that allows brain structure, including differentiating white matter, or axons, from gray matter, or cell bodies, to be visualized) confirmed that H.M. did sustain damage: not only to the hippocampus and amygdala but also to the surrounding cortical areas. This scan validated all the work that I had done for my dissertation, for which I earned my PhD and was awarded the prestigious Lindsley Prize, given by the Society for Neuroscience to the best doctoral dissertation in the field of behavioral neuroscience.

While I never met patient H.M., I thought so much about his brain and about what he could and could not remember that I felt like I
knew
him. I’ll never forget the morning of December 4, 2008, when I opened up the
New
York
Times
to see his obituary on the front page. My first shock was to learn his full name for the first time in the twenty years I had been studying him. Henry Molaison. This was very likely the best-kept secret in all of neuroscience, revealed only at the time of his death. It was like learning something precious and very personal about a friend the day that he died. I happened to be teaching a big lecture course that day on the topic of memory. I shared the news with the class and even got a little emotional as I told them. They must have all thought I was a bit strange, but I couldn’t help it. Henry Molaison, patient H.M., had given up so much in his life for our understanding of memory. Since the day of his surgery, he could never remember another Christmas or birthday celebration or vacation—he couldn’t have a deep relationship with another person or make any plans for his future. He lost something precious the day of his surgery, but his misfortune enriched our knowledge of the brain and memory in a profound way. I will always honor his sacrifice.

MRI

MRI
stands for “magnetic resonance imaging,” and it is a powerful and common imaging tool that uses strong magnetic fields and radio waves to form images of the body, including the brain. This general imaging approach, also called structural imaging, is widely used to see the gross structure of the brain and the boundary between the so-called gray matter (cell bodies) and white matter (axonal pathways) of the brain.

MOVING ON: STUDYING MEMORY AT NIH AND STARTING MY OWN LAB

I had spent six years at U.C. San Diego mastering neuroanatomy and behavioral approaches to examine the connections of key brain areas in the medial temporal lobe as well as the effects of damage to those areas. While these are important areas of study, they still don’t let you look firsthand at what’s happening in the brain during the formation of new memories. That’s what I wanted to do next. I wanted to learn new approaches by which I could examine the patterns of electrical activity in brain cells as animals performed different memory tasks. I wanted to look directly at the cells and the activity in the hippocampus that was occurring as animals learned something new. I secured a position as a postdoctoral fellow in the lab of Robert Desimone at the National Institutes of Health to do just that.

Desimone’s lab was in the larger laboratory of neuropsychology run by Mort Mishkin, the same neuroscientist who had published key findings on the effects of hippocampus and amygdala lesions in animals and whom I had first heard about while in France. I spent the next four and a half years at NIH learning how to record the activity of individual and small groups of living brain cells as animals performed various memory tasks. This general approach is called behavioral neurophysiology, and it’s powerful because we can examine how patterns of electrical activity in the brain relate to actual behavior. It is also powerful because it gives us a direct window on understanding exactly how particular brain cells respond to a given behavioral task. This contrasts with the studies of what happens with brain damage, like in the case of H.M. While transformative for our understanding of memory, lesion studies are, by their nature, indirect. We are studying the lack of function that used to be there before the damage. By contrast with behavioral neurophysiology, we can start to understand how the normal brain typically responds during a memory task.

It’s important to note that there are no pain receptors in the brain, so the microelectrodes we used for the recordings didn’t cause any discomfort, but they did allow us to record the brief electrical bursts of activity (called action potentials or spikes) that occur as an animal is learning or remembering something new. I basically trained animals to play video games focused on learning and memory and then recorded the activity of individual cells to figure out how the brain signals different aspects of the task and what happens to the pattern of brain activity when the brain remembers or forgets. I focused on one of the cortical brain areas in the medial temporal lobe, the entorhinal cortex, and characterized the patterns of neural activity in this area as animals performed a memory task. This was one of the only studies like this done in the entorhinal cortex. But I knew that there was much more left to understand relative to the physiological response properties of other key areas of the medial temporal lobe. That’s what I wanted to focus on in my own lab.

Those four years at NIH were intense and very valuable because they taught me the ins and outs of this powerful approach of behavioral neurophysiology, which I brought with me when I started my own neuroscience research lab in 1998. This is where things really got interesting in my career. I had at this point been studying memory for ten years. I was thrilled beyond belief to now be able to build my own research program focused on my scientific obsession—understanding what happens in the hippocampus when a new memory is first formed. My desire to learn this was inspired directly by the original description of patient H.M. He could appreciate the things around him in the present moment, but unlike us, he could not make that information stick in his brain longer than he could focus his attention on it. We knew that the ability to retain it depends on the hippocampus and surrounding cortical areas, but we had no idea what these cells do when a new memory is formed.
That
was the question that I wanted to investigate in my lab.

So as head of my own lab, the first decision I needed to make was what kind of information I was going to have the animals learn. It had to be something relatively simple, so they could do it easily, and be a task we knew was impaired by damage to the hippocampus and surrounding structures. I settled on something that required animals to associate particular visual cues (such as a picture of a dog or a house or a building) with a particular rewarded target to the north, south, east, or west on the computer monitor. We knew this form of learning, called associative learning, was a subcategory of declarative memory (in other words, it could be consciously learned and brought to mind), and there was good evidence that damage to the hippocampus and/or its surrounding brain structures caused significant impairment in learning these picture–target associations.

I set about teaching animals to learn multiple new associations each day, and when they could do the task very well, I introduced a thin electrode into their brain to record activity as they were in the process of learning.

Finally! I was going to be able to peer into the brain and see what happens in the hippocampus as we learn something new.

One reason people had not done this kind of experiment before is because it’s difficult to get animals to learn new associations. It turns out that the task that I chose was a good one; animals could learn multiple new associations in a given session. This was exactly what we needed to start looking at how new associations are signaled in the hippocampus.