In the Palaces of Memory

(Critical Survey of Contemporary Fiction)

George Johnson, a NEW YORK TIMES journalist and author of MACHINERY OF THE MIND: INSIDE THE NEW SCIENCE OF ARTIFICIAL INTELLIGENCE (1986), begins his new book with a statement so arresting it should be enshrined in writing classes as an example of how to hook a reader: “Whenever you read a book or have a conversation, the experience causes physical changes in your brain. In a matter of seconds, new circuits are formed, memories that can change forever the way you think about the world.” As the implications of that remarkable fact sink in, it becomes easy to believe Johnson’s assertion that, for the three years he spent working on this project, it was difficult for him “to maintain” much of an interest in anything else.”

Unfortunately, IN THE PALACES OF MEMORY doesn’t live up to the promise of its provocative opening. Johnson tried to write several books a the same time. He wanted to show “science in the making,” the necessarily messy process of theory-building and testing, the many false starts and the rate breakthroughs—all of which gets radically simplified and cosmeticized in after-the-fact textbook accounts. Yet by tying his account to work-in-progress, Johnson runs into the same problem that besets filmmakers who, seeking to capture the quality of ordinary life, create films that are unbearably tedious.

At the same time, Johnson wanted to tell his story via profiles of individual scientist—a strategy employed...

(The entire section is 472 words.)

In the Palaces of Memory

(Literary Masterpieces, Critical Compilation)

It is astonishing, and perhaps even pleasing, to be reminded that, for all of humankind’s scientific advances, something so simple and basic as human memory remains a mystery. Despite intensive research, no one is able to say precisely how and where memories are stored in the brain. The mechanism that allows one to remember that George Washington was the first president of the United States, that one’s best friend from fifth grade had a dog named Susie, or that the local coffee shop serves a blue-plate special on Thursdays is still one of nature’s best-kept secrets.

In In the Palaces of Memory, the scientific race to uncover this secret is documented in exquisite detail. George Johnson, a New York Times journalist, gives a rich (and, for the nonscientifically inclined, sometimes dizzying) account of the latest theories and experiments. His book is not an encyclopedia of all that is current in the field of brain and memory research, nor is it an argument for one particular theory over another; rather, In the Palaces of Memory is an action-adventure story. It is the heat of the race to uncover the workings of memory that makes Johnson’s book so fascinating. In the Palaces of Memory is as much about the way scientists actually discover things as it is about the knowledge produced by science itself.

For the uninitiated, Johnson’s book is an eye-opener. One meets scientists with egos, blind spots, petty jealousies—and the same kind of idiosyncratic genius and creative flair one usually expects only from bohemians and artists. One observes the course of scientific research flowing not, as one might have presumed, relentlessly and logically forward, building upon good ideas while discarding bad, but tracing a winding path, up, down, and around the politics of funding, almost as arbitrarily as the vicissitudes of fashion. One theory may fall into disrepute only to be revived again thirty years later, or a new and quite plausible theory might pop up but be virtually ignored because it does not incorporate the currently fashionable synapse or the chemical compound of the day. This is not to say that Johnson debunks science or scientists; rather, he humanizes them.

Johnson focuses on three researchers in particular: Gary Lynch, a biochemist at the University of California, Irvine, who studies synapses (the gaps neurotransmitters cross when transferring nerve impulses from one cell to another); Leon Cooper, a physicist at Brown University, who shared the 1972 Nobel Prize for the theory of superconductivity and who now works with artificial intelligence, simulating the operation of nerve nets in large computing systems; and Patricia Churchland, a professor of philosophy at the University of California, San Diego, who, having studied the mind for years in the abstract, took classes at medical school to get a closer look at the physical brain and now attempts to provide the neuroscience movement with philosophical underpinnings.

With each of his protagonists, Johnson drops the reader into the center of a scientific controversy. Among neuroscientists such as Gary Lynch, for example, there is agreement that learning affects the synaptic connections between nerve cells—but whether the change occurs at the sending neuron or the receiving neuron is in hot debate. Johnson details the waning hypotheses, experiments, and conclusions of the postsynapticists and the presynapticists, then introduces a physicist, Leon Cooper, who thinks such slavish attention to the neuron is beside the point.

Cooper believes that it is not so much what goes on inside a neuron that matters as how each neuron interacts with others, the web of neural connections that yields the full thinking and remembering capacity of the brain. He has little interest in the passive biochemical act of recording information. Rather, Cooper wonders how the brain generates its information, how it codes, stores, categorizes, and how it perceives and retrieves patterns. A theorist who believes that theorizing is as much a fine art as music or painting, Cooper is most enthralled with how the brain performs this function, its actual physical mechanism for theorizing. Cooper’s approach to solving this problem is also no doubt annoying to the biologists. He is trying to construct a workable model of a neural network on computer in an attempt to simulate the mapping procedure of the brain. It is the “software,” the very program the brain runs, that Cooper endeavors to duplicate; the “hardware” (whether biochemical neurons or a machine) assumes less significance....

(The entire section is 1883 words.)