The human brain is responsible for the painting of Van Gogh, the creation of democracy, the design of the atomic bomb, psychosis, and the memory of one's first vacation and of the way that hot dog tasted. How does this organ encompass such diversity?

The brain is not a neatly organized system. It is often compared to an overgrown jungle of 100 billion nerve cells, or neurons, which begin as round cell bodies that grow processes called axons and dendrites. Each nerve cell has one axon and as many as 100,000 dendrites. Dendrites are the main way by which neurons get information (learn); and axons are the main way by which neurons pass on information to (teach) other neurons. The neuron and its thousands of neighbors send out roots and branches-the axons and dendrites-in all directions, which intertwine to form an interconnected tangle with 100 trillion constantly changing connections. There are more possible ways to connect the brain's neurons than there are atoms in the universe. The connections guide our bodies and behaviors, even as every thought and action we take physically modifies their patterns.

This description of the developing brain was heresy until recently. For decades scientists maintained that once its physical connections were completed during childhood, the brain was hard-wired. The tiny neurons and their interconnections were fixed; any neuron or link could die, but none could grow stronger, reorganize, or regenerate. Today, these axioms have been amended and enhanced. Thanks to sharp imaging technology and brilliant clinical research, we now have proof that development is a continuous, unending process. Axons and dendrites, and their connections, can be modified up to a point, strengthened, and perhaps even regrown. Temple Grandin's achievement demonstrates that the brain has great plasticity. But what was actually going on, physically, inside her head? We get a strong clue from Michael Merzenich at the University of California, San Francisco.

Merzenich implanted electrodes in the brains of six adult squirrel monkeys, in the region that coordinates the movement of their fingers. Using computer imaging, he created a map of the neurons that fired when the monkeys manipulated objects with their hands. He then placed four food cups of decreasing diameter outside each of their cages. He put a single banana-flavored food pellet in the widest cup. The monkeys would reach through the bars and work their fingers into the cups until each was able to grasp its pellet and eat it. They practiced dozens of times for several days.

Once they had mastered the widest cup, Merzenich put the food pellets in the next smaller cup. After several days of repetition, the pellets were moved to the third cup, then the fourth. By the end of the experiment the monkeys were extremely skilled with their fingers.

After only one day, the computer images showed that the area of the brain that became active when the monkeys moved their fingers had increased in size. As the animals conquered successively smaller cups, the area got bigger; the number of cells that participated in the task increased. But after the neurons in the cortex mastered the fourth cup, the area shrank again; as the skill became more automatic, it was delegated to other parts of the brain lower down in the chain of command. The expanded portion of the executive part of the brain, the cerebral cortex, was no longer needed to carry out the skill and guide the hand. This commanding part of the brain, the control center, reverted back to its original size, freeing up neurons to learn other things. There is evidence that the same thing happens in humans as in Merzenich's monkeys. Alvaro Pascual-Leone of Harvard Medical School, Boston, and Avi Karni of Hadassah University Hospital, Israel, have independently shown, using mapping techniques-for example, magnetic resonance imaging (MRI) and transcranial magnetic stimulation-in living human subjects, that skill acquisition recruits more cortical neurons to master the skill, and that as the skill becomes more automatic, less of the recruited cortex is used.

Thus, the brain has a tremendous ability to compensate and rewire with practice. Temple trained at the Safeway doors for hours each day for several months until the skill became automatic. At first it was incredibly difficult; by the end she did it with little concentration. Once she had mastered the initially higher-order activity, it was probably pushed down into lower regions of the brain, freeing her cortex to learn a new skill. The same would seem to be true about her rehearsing the Sadat-Begin tapes and applying them to her own conversations.

Practice counts.

What we learn from Temple's story and Merzenich's monkeys is that our brains are wonderfully plastic throughout adulthood. Brain structure is not predetermined and fixed. We can alter the ongoing development of our brains and thus our capabilities. This is not always beneficial, however, as sometimes in the brain's attempt to adapt, the rewiring can make things worse.

Excerpted from A User's Guide to the Brain by John J. Ratey, M.D. Copyright © 2002 by John J. Ratey, M.D.. Excerpted by permission of Vintage, a division of Random House, Inc. All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.

Tags: Neuropsychology, Neurological Disorders

About the Author

John J. Ratey, M.D., is an associate professor of psychiatry at Harvard Medical School. He has lectured extensively and published many articles on the topic of treating adults with ADD. Dr. Ratey is the author of A User's Guide to the Brain and the co-author of Driven to Distraction. He lives in Cambridge, Massachusetts, where he has a private practice. More