Welcome to Your Brain (29 page)

declarative memory. This form of memory requires the temporal lobes at the sides of the

brain, the hippocampus, and parts of the thalamus, a football-shaped region at the core of

the brain.

Other types of memory rely on different brain regions. For instance, the intensity of

memory for a terrifying experience like an encounter with an angry bear depends on the

amygdala. Learning some types of movement coordination, such as how to make a smooth

tennis stroke, requires the cerebellum. A skill such as driving a car uses a number of brain

regions but does not require the temporal lobe system, where Leonard’s brain damage is

located. People with damage to these areas remain capable of learning new skills, like

drawing upside down, though they typically have no memory of having practiced the skill

before.

The study of the streets of London is a major undertaking that culminates in a daunting examination

known as “The Knowledge.” Would-be drivers roam London on motor scooters armed with a phone

book-sized map, running the maze of streets over and over until they can mentally locate each street

and figure out how to get there from any other place. This process culminates in certification exams

that require months to take. The average time to be licensed to drive a taxi in London is two years.

Neuroscientists at University College London examined the brains of taxi drivers to see if this

intensive study had any effect. The scientists used magnetic resonance imaging to map out the

structure of the brains of fifty male drivers and fifty males who did not drive taxis. Only one part of

the brain was different in drivers and nondrivers: the hippocampus, a structure that is shaped like a

partially unfurled scroll. This difference was small but measurable. The posterior hippocampus of

drivers was on average 7 percent larger than in nondrivers, and the anterior hippocampus was 15

percent smaller. Compared with these numbers, the variation within each group is large enough that it

would not be possible to tell which group someone belonged to simply by examining the

hippocampus. But on average, compared with nondrivers, drivers had a larger back end of the

hippocampus and a smaller front end. The more years of experience a driver had, the larger this

disproportion tended to be. This difference was not observed in bus drivers, who also drove every

day but repeatedly followed the same route. Could it be that acquiring and using The Knowledge

makes the hippocampus grow?

What might cause these differences? Active neurons secrete growth factors known as neuro-

trophins that can cause dendrites and axons to extend their existing branches and even generate new

ones. As we mentioned before, neurotrophin secretion is a key event in early development. Similarly,

extensive use of neural tissue may lead to growth later in life. New neurons are also born in adults at

a low rate, which is higher in the hippocampus than in other brain regions. We don’t know for certain

how the expansion in size and number of neurons would affect function, but a plausible guess is that it

would expand as well.

This leads us to one of the core questions in neuroscience: what is it that changes in the brain

when we learn something? The difficulty is that few of these changes are likely to be visible when we

look at gross structure. Instead, new information is likely to be stored as changes in the strength of

connections between neurons, and as changes in which connections are made. These changes don’t

necessarily alter the size of a brain structure any more than the size of a piece of paper changes when

you write on it. So measurement of the sizes of brain structures is a fairly crude and indirect way to

assess their capabilities.

The original reason that these researchers decided to look at the hippocampus was that it is

known to be involved in spatial navigation in humans and in other animals. As rats run around in a

maze, neurons of the hippocampus fire only when the rat is in a particular location. Because the rat

hippocampus contains millions of neurons, each place in the maze is then associated with hundreds or

thousands of neurons that fire when the rat is there, but not before or after. Taken together, all the

neurons of the hippocampus, firing and not firing, hold a map composed of place cells, in which

subsets of firing neurons signify where the rat is.

The same phenomenon has been found in humans during a video game that is very similar to what

London taxi drivers do every day. Recording from individual neurons in humans is normally not

advisable because it requires opening the skull, but it has been done in people with severe epilepsy.

In these patients, electrodes are often implanted to identify places in the brain where seizures begin,

so that those parts can be removed without damaging neighboring regions important for normal

function. Researchers took this as an opportunity to spy on the activity of neurons as patients played a

taxi-driving video game. The game involved driving to various destinations in a simulated town, like

a very boring version of Grand Theft Auto, without the gangs, crime, or sex.

Very much like in rats, the hippocampi of human virtual cabbies had place cells. For instance,

some cells fired when the player was in front of the drugstore, but did not fire when he was at the

grocery store. The specific response of cells to various imaginary locations began after subjects had

played the game just a few times. How does this happen so quickly? One possibility is that something

like a blank map is already in place in your head, waiting to be linked up to experiences of actual

places. This may be the first step in learning to navigate a new locale—like a taxi driver in training

going around on a scooter with a map.

In addition to being involved in forming memories of places, the hippocampus also is important

for declarative memory (the recall of facts and events). For example, if you remember the taxi trip

through London earlier in this chapter (and we hope you do), you are using declarative memory.

Canadian psychologist Brenda Milner was the first to appreciate the importance of the hippocampus

and structures near it for this form of memory. In the 1950s, Milner examined a patient, HM, who had

undergone radical surgery to treat severe epileptic seizures. Like the patients who played the taxi-

driving game, HM’s seizures began in the hippocampus or nearby in the temporal lobes of the

cerebral cortex. However, at that time it was not standard practice to record activity before surgery.

Doctors only knew that seizures often began in the temporal lobes and the hippocampus. So they

surgically removed these structures in their entirety.

After the surgery, HM’s seizures were indeed less frequent. He was also able to have

conversations, solve logic puzzles, and carry out the activities of daily living. But he had an odd

deficit as well. He suffered a profound loss in his ability to remember an event, even a few minutes

after it happened. Milner tested him many times over the following months. He did well on the same

tasks and even improved with repetition. Yet he could not form new memories of events or people.

For instance, each day he greeted Milner as if meeting her for the first time.

Milner and other neuroscientists eventually reached the conclusion that temporal structures are

essential for forming declarative memory. The problems experienced by HM have now been seen in

many patients after strokes have damaged their temporal brain structures, including the hippocampus.

Myth: Recovered memory

Memories are not played back like a tape or a file recalled from a computer’s hard

drive. Instead, they seem to be stored in shorthand, broken into chunks in which the

uninteresting bits are discarded, leaving only the details that your brain considers

important. As we discussed in
Chapter 1,
your brain also invents details to create a more

coherent story. This has occasionally caused memorable tragedies.

In a wave of scandalous cases in the 1980s and 1990s, social workers and therapists

identified “repressed memories” of childhood abuse. The stories were uncovered after

interviewers repeatedly asked leading questions and then rewarded the most interesting

answers with attention. In Manhattan Beach, California, a lawsuit claimed that hundreds of

children had been sexually abused at the McMartin Preschool, some in nonexistent

networks of underground tunnels. These unbelievable tales led to lengthy court cases and

the wrongful imprisonment for five years of Ray Buckey, a counselor at the school.

Filling-in of memories is a well-documented phenomenon. In one study, researchers

asked people where they were when they learned that the space shuttle
Challenger
had

exploded. People gave different answers several years later than they did immediately after

the explosion, providing more evidence that people sometimes invent plausible

explanations when they don’t recall what happened.

Researchers have stimulated false memories in the laboratory as well. For instance, if

you are shown a list of words with a similar connotation—
ice cream
,
honey
,
lollipop
,

sweet
,
candy
,
chocolate
—and later asked if the word
sugar
was on the list, there is a good

chance that you will say yes with confidence. This is an example of filling in, in which a

reasonable inference is made that an event might have happened, even though it did not.

The fragility of memory plays into another common myth, which dates to the teachings

of Sigmund Freud. He speculated, without hard evidence, that traumatic events could be

repressed and thereby made unavailable to the conscious mind. The concept has become so

entrenched that it is believed today, even by many mental health workers. However, almost

no scientific evidence exists for repression. The weakness of the evidence is detailed in

psychologist Daniel Schacter’s
Searching for Memory
. Severely traumatic experiences are

forgotten only if the trauma leads to unconsciousness or brain damage, or if the experience

happens to a person too young to be able to form long-term memories, a process that begins

around the age of three or four. Most memory researchers agree that recovery of a lost

traumatic memory is very rare.

Practical tip: Can’t get it out of my head

Anne Waldman is stuck. She and her son are hard at work on a collection of songs

based on her poetry, entitled
The Eye of the Falcon
. As she polishes the songs, she finds

that she can’t get a certain phrase out of her head. It’s driving her crazy. Why is this little

phrase so persistent?

Think of the phrase bothering her as an example of a sequence. Sequence recall has a

special and useful place in our memories. We are constantly called upon to remember

sequences, from the movements involved in signing your name or making coffee in the

morning to the names of the exits that come before the turnoff you take to drive home every

day. The ability to recall these sequences makes many aspects of everyday life possible.

As you think about a snippet of song or speech, your brain may repeat a sequence that

strengthens the connections associated with that phrase. This in turn increases the

likelihood that you will recall that phrase, which then leads to more reinforcement. This

cycle of repeated recall may be necessary for the normal strengthening and cementing of

memories.

In Anne’s case, though, the repetition helped form a positive feedback loop and a

vicious cycle. At first she recalled the phrase on purpose, but after a time it arose unbidden.

In her case, the bothersome phrase is one she was actively working on, and one with

substantial emotional impact. Emotions can highlight the effect of experience and make

events more likely to be consolidated in memory.

How can one break this unending cycle of recall and reinforcement? One way is to

introduce other sequences that interfere with the reinforcement of the memory. Thinking of

another song may allow a competing memory to crowd out the first one. Anne attempted to

overwrite her repetitive memory by listening to a Poulenc opera with Jean Cocteau. That’s

the best therapy we can suggest: find another infectious song—and hope that the cure

doesn’t become more annoying than the original problem.

Since both place memory and episodic recall require the hippocampus, scientists speculate that

these two forms of memory may share some common principle. One idea is that they both rely on

placing events relative to one another in context. In spatial memory, the relationship is physical, in

space; in episodic memory, the relationships are more general, in time or even by logical connection.

What physical property of the hippocampus allows it to make these logical connections?

About a hundred years ago, the psychologist William James suggested that our experiences trigger

sequences of activity in the brain. Under the right conditions, these sequences can then lead to changes

that increase the likelihood that they will occur again, even after the experience is past. If the activity