Healthy Brain, Happy Life (8 page)

This showed that Lashley’s theory was wrong: There is a particular part of the brain that is specifically involved in allowing us to create new memories. What part of the brain was it? This is where Scoville and Milner were appropriately cautious. H.M.’s operation damaged both the hippocampus and the amygdala on both sides of the brain. This general region of the brain that houses the hippocampus and amygdala is often referred to as the medial temporal lobe. That is, it’s the part of the brain’s temporal lobe located toward the middle of the brain (
medial
in anatomical terminology). But in examining the nine other psychiatric patients who had varying amounts of medial temporal lobe damage, Scoville and Milner noticed that the more the hippocampus was damaged on both sides, the more severe the memory deficit. This led them to suggest that the large extent of hippocampal damage on both sides was likely underlying the severe memory deficit in H.M.; however, they could not rule out the possibility that concurrent damage of the amygdala plus the hippocampus was at the root of the memory loss.

THE FASCINATING STORY OF PATIENT H.M.

Patient H.M. is one of the most fascinating and most extensively studied neurological patients in the learning and memory literature. After Brenda Milner’s work with him, her then graduate student and now professor emerita at MIT Suzanne Corkin studied H.M. for a total of forty-seven years, until his death in 2008. If you want to know more about patient H.M. and his story, I recommend Corkin’s wonderful book,
Permanent Present Tense:
The Unforgettable Life of the Amnesic Patient H.M.
Listen to me interview Suzanne about H.M. on the podcast
Transistor
by PRX.

But this was not all Milner noticed. Once she characterized the severity of H.M.’s everyday memory loss, she got to work figuring out if there was anything at all he could learn and remember normally. She and others later showed that H.M. had an inability to form any new memories for facts (termed semantic memory) or events (termed episodic memory), typically referred to together as declarative memory—the kind of memory that can be consciously brought to mind. Next Milner revealed that H.M. did have normal memory for some things. Namely, she showed that he still had the ability to learn new motor or perceptual skills at the same rate as people who had not undergone surgery. Milner had him do tests in which he had to learn how to trace a figure accurately while looking in a mirror. H.M. improved steadily day by day but, strikingly, had no memory of ever having done the task before. Similarly, he was able to learn perceptual tasks in which he was given a vague outline of a picture and, after a variable amount of time looking at the incomplete figure, gradually picked out the image. He learned to identify those objects at the same rate as nonpatients as well. This was another revelation in the memory field. This finding suggested that different brain areas outside of the hippocampal region were necessary for these forms of motor and perceptual memory.

So the partnership of Scoville and Milner revolutionized the way we understand memory. Their studies led to our understanding that the medial temporal lobe, which includes the hippocampus, is essential for our ability to form new memories for facts and events. The researchers also showed that memories are not stored in the hippocampus because H.M. retained normal memories of his childhood and demonstrated that different forms of memory, including perceptual memory and motor memory, depend on different brain areas outside the medial temporal lobe.

But one additional contribution of that original Scoville and Milner paper cannot be overlooked. The report served as a grave warning to the neurosurgical community that the bilateral removal of the hippocampus should never be done again. H.M. lost his ability to form any new memories and spent the rest of his life depending on his family to care for him. The operation took away his ability to retain anything new about what happened to him and what was going on in the world. It was a terrible price to pay for the reduction of his epileptic seizures, and Scoville and Milner made sure that the entire neurosurgical community understood.

DIFFERENT KINDS OF MEMORY

The kind of memory that H.M. lost with his brain damage is called declarative memory, which refers to those forms of memory that can be consciously recalled. In addition, there are two major categories of declarative memory that depend on the structures of the medial temporal lobe:

•  
Episodic memory
, or memory for the events of our lives, which are those memories of our favorite Christmas celebrations or summer vacations; such “episodes” make up our unique personal histories.

•  
Semantic memory
, which includes all the factual information we learn throughout our lives, such as the name of the states, the multiplication table, and phone numbers.

We now know there are many forms of memory that do not depend on the medial temporal lobe, for example:

•  
Skills/habits:
These are the motor-based memories that allow us to learn to play tennis, hit a baseball, drive, or put our keys in our front door automatically. They depend on a set of brain structures called the striatum.

•  
Priming:
This describes the phenomenon that exposure to one stimulus can affect the response to another stimulus. For example, if you give someone an incomplete sketch of an object that she can’t identify but then show her a more complete sketch of the image, on the next round, she will be able to recognize the object even if less information is provided. Many different brain areas participate in priming.

•  
Working memory:
This form of memory has been called the mental scratchpad and helps us keep relevant information in mind where it can be manipulated. For example, you are using your working memory during a talk with your financial adviser, who is describing the different mortgage rates for you as you try to decide which one is best for your situation. The ability to keep the figures in mind and manipulate them to come to a decision is an example of working memory. That H.M. could keep topics in mind enough to have normal conversations showed that his working memory was intact.

FINDING MY PLACE IN THE MEMORY MYSTERY

The groundbreaking report of Scoville and Milner in 1957 cracked the study of memory wide open and started an avalanche of new questions for neuroscientists to explore. Two questions at the top of the list were, first, figuring out which exact structures in the medial temporal lobe were critical for declarative memory: Was it just the hippocampus or the hippocampus and the amygdala? And, second, how do you visualize the specific change that occurs in the normal brain when a new declarative memory is formed? I didn’t know it when I first began graduate school, but I was going to tackle the first question as my graduate thesis and the second question when I was an assistant professor at NYU.

By the time I entered U.C. San Diego in 1987, we knew a lot more about the important contribution of the hippocampus to memory, but the raging debate at the time focused on whether it was damage to the hippocampus alone that was underlying H.M.’s deficit, as Scoville and Milner hypothesized, or if it was the combined damage to the hippocampus and the amygdala, another possibility that could not be ruled out. A benchmark finding in animals in 1978 by Mort Mishkin appeared to provide evidence that it was the combined damage to both the hippocampus and the amygdala that led to the most severe memory deficits. Yet, in 1987 when I entered graduate school, Squire and Zola-Morgan at U.C. San Diego were finding evidence that the amygdala might not be involved after all. They had shown in animals that damaging the hippocampus on both sides caused a clear memory deficit, but they found no deficit after damaging just the amygdala on both sides of the brain. Then they did what turned out to be a key experiment. They added very precise damage to the amygdala in animals that had both their hippocampi removed. The researchers saw the addition of the selective amygdala damage did not in fact make the memory deficit worse, as predicted. The question was, If the additional memory impairment was not due to damage to the amygdala, then damage to what brain structure was it due to? A clue to this mystery came from a careful examination of the anatomy of the brain lesions. Neuroanatomist David Amaral was looking at the extent of damage in the brains of these animals in thin sections of tissue and noticed something obvious only to a neuroanatomist: There was a lot more damage than to just the hippocampus and amygdala. Namely, a lot of the cortex surrounding the brain areas of these animals was also damaged, in varying degrees. It was likely that the same damage would be present in patient H.M., given the surgical approach used to make his brain lesion. Maybe the nondescript cortical areas surrounding the hippocampus and amygdala that nobody had ever considered very important, and had previously thought to be part of our visual system, were the key to the mystery.

This is where I entered the picture. Amaral ran a neuroanatomy lab at the Salk Institute in San Diego right across the street from U.C. San Diego, and he was a leading expert on the anatomical organization of the medial temporal lobe. It seemed clear to me that we needed a more careful understanding of the basic structure of this part of the brain, so when they asked me if I wanted to take on that challenge, I jumped at the opportunity. I literally felt like a neuroscientist version of David Livingstone, entering one of the deepest, darkest parts of the brain—somewhere few others had gone before.

I had thought for sure that all parts of the brain had been carefully examined and mapped in 1987 when I entered graduate school, but I soon found out that the areas I was focusing on had fallen through the cracks. I was one of the first to study them carefully. I used some of the same basic techniques that had been used by neuroanatomists since the early 1900s. I examined thin slices of the brain from key temporal lobe areas and stained them with a chemical to show the size and organization of the cell bodies of the neurons and glia that made up the tissue (this technique is called a Nissl stain). I looked at some slices to see if I could identify features that would allow me to differentiate one area from the next. In other studies, I tracked where these areas received inputs from and where they projected to.