The Invention of Memory

(Literary Masterpieces, Volume 7)

Though the mysterious inner workings of the mind have long been a source of fascination and investigation, two theories about how the brain works have for the most part never been challenged. The first is that memories exist in a permanent storage file in the brain. Much like a computer that holds all the data programmed into it, the mind, it is believed, holds a record of everything it has ever experienced. Wilder Penfield’s experiments in the 1930’s seemed to confirm this, when electrical stimulation of certain areas in his patients’ brains triggered memories of long-forgotten incidents. The second assumption is that brain function is localized—that is, certain brain activities can be traced to occur in specific areas of the brain. This idea was advanced by the dramatic brain research conducted by European scientists in the late nineteenth century, most notably by Paul Broca, who discovered in 1861 that a loss of speech could be correlated with the existence of a small lesion on the left side of the brain. Belief in permanent memory and localization of function has continued to form the basis for much of the brain research conducted in the twentieth century. In The Invention of Memory: A New View of the Brain, however, Israel Rosenfield argues that these two long-held assumptions may simply be wrong. As his subtitle suggests, he offers a new theory about how the brain works, basing his views on a careful reexamination of past research, in many cases arguing for different conclusions from the same set of data, and synthesizing that information with more recent neurobiological discoveries.

Rosenfield favors a theory known as “Neural Darwinism,” first proposed by Gerald Edelman, winner of the 1972 Nobel Prize for Medicine or Physiology. According to this theory, memory is not a fixed record, but an ever-renewing recreation, based on the stimulation and subsequent interaction of selected nerve networks, or “maps,” in the brain. A present stimulus, usually linked to an emotion, activates a past connection of nerve cells, and these interact with one another on the basis of a fluid system of categorization. The “shapes” of the neuron maps are continually altered by experience. Since it must adapt to a constantly changing world, the mind, Rosenfield argues, cannot afford to operate from a rigid filing system of stored information. Its survival depends upon the constant interaction of past experience with present environment; a category that was useful yesterday may no longer be useful today. Only the fittest neuronal connections—the ones that help the organism cope, the ones that are strengthened through useful repetition will survive; hence the “Darwinism” among the neurons.

Rosenfield builds his case for Neural Darwinism by first dismantling the theories of permanent memory and localization of function. Aside from the fact that neurobiologists have never been able to explain exactly how and where permanent memories are stored, at the theoretical level inconsistency and paradox abound. For example, if memories are a fixed record, why is it that, as Sigmund Freud noted, memory is often fragmentary and inexact? If a memory is fixed and filed, why does it come out in a slightly or sometimes radically different way each time it is recalled? The same question could be posed to the nineteenth century localizationists, who believed in word and symbol centers in the brain where fixed records of words and images were stored and that damage to these centers would inhibit a patient’s ability to recall words and symbols. Ask someone to draw several five-pointed stars in a row; rather than producing several identical images (presumably from his stored image of such a star), he will probably make stars that are all slightly different from one another. If the brain were merely copying a fixed image, why are all the drawings not identical replicas? Further, the idea that the brain “knows” a horse when it sees one because it has a stored image of a horse on file offers no explanation for how the brain copes the first time it sees an elephant. The localizationist view that perception depends upon recall of a stored image, contends Rosenfield, leads to the absurd conclusion that “the world is knowable . . . only if it is already known.”

For Rosenfield, these paradoxes point the way toward the importance of context in both recall and perception, something he believes the earlier researchers overlooked. In 1891, Jules Dejerine’s work with brain-damaged patients who were unable to recognize words or multidigit numbers confirmed his belief in the existence of specialized memory centers in the brain. Dejerine concluded that either his patients had suffered damage to the memory centers where words were stored or there was some disconnection between the center for visual images and the center for making sense of those symbols as language. Yet that these patients were able to recognize some letters or some single-digit numbers leads Rosenfield to a different conclusion. If a man...

(The entire section is 2053 words.)