COVER

THE MYSTERIES OF MEMORY

MARK NICHOLS April 11 1988
COVER

THE MYSTERIES OF MEMORY

MARK NICHOLS April 11 1988

THE MYSTERIES OF MEMORY

COVER

An important early clue in the quest to solve one of the major mysteries of the human brain came almost by accident. Thirty-five years ago an American surgeon who was trying to treat a patient’s epileptic seizures turned to radical surgery—the removal of the curled bodies of brain tissue known as the hippocampus in the man’s temporal lobes. The operation relieved the patient’s seizures, but it also had a debilitating side effect. The man, known in medical literature only by his initials lost his ability to remember longer than fleetingly things that he had learned since the operation. As a result, H.M., who is now 60, has lived ever since in a present that has no recent past.

Stride: But the case did yield benefits for scientists struggling to understand memory.

Brenda Milner, a professor of neuropsychology at Montreal’s McGill University who has been studying H.M.’s case for more than 30 years, concluded that one of the functions of the hippocampus is to transform short-term memories for longterm retention. The discovery amounted to a major stride forward in the understanding of memory. Now a growing number of investigators are searching inside the brain and constructing new theories of memory, one of the least understood of the brain’s functions.

But progress in the field is slow. Just as everyone has experienced the infuriating vagaries of memory, scientists still are unable to answer some of the basic questions, such as how information is recorded in the brain or where it is stored. Memory, says Victor Shashoua, an associate professor of biological chemistry at McLean Hospital in Belmont, Mass., is “an incredible system, but we really don’t understand how it works yet.” Solving that riddle conceivably might open the way some day to developing treatments for such memory-loss disabilities as those that occur in Alzheimer’s disease, or even for less severe cases of forgetfulness. But until fairly recently scientists were not even sure exactly which parts of the brain are involved in memory. Now, through the study of animal brains and the examination of

disease-damaged human brains, a clearer picture is emerging. Researchers know that sensory input— anything from a face seen through a window to a flower’s scent—first travels through nerve circuits to the cerebral cortex, the structure that integrates incoming data and distributes it to various parts of the brain. These include the cerebellum, where learned responses are thought to be stored, and the hippocampus, which, besides its role in designating some impressions for long-term storage, probably also plays a role in help-

ing the brain to remember spatial layouts.

New ideas of how neural pathways operate are being advanced. After studying memory loss in victims of Parkinson’s disease at Toronto Western Hospital, psychologists Jean SaintCyr and Ann Taylor proposed recently that the caudate nucleus, a region in the front of the brain, has a pivotal role. They say that it is likely involved in the ability to learn simple tasks, solve new problems or plan future actions.

Tie: In a two-way system with the cortex, explains Saint-Cyr, the caudate is responsible for storing basic information for future use. That, he adds, would include instructions on how to perform tasks—“like, ‘How am I going to tie my shoelace?’ If you are a Parkinson’s victim and your caudate is not

working well, then you may not remember the best way to do that because the caudate is sending bad information to the cortex.”

Flux: Other scientists are searching for answers to one of the most baffling memory puzzles of all: how, in the brain’s flux of chemical and electrical events—where nothing seems to be permanent—can learning or memories be physically stored? During the 1950s the American psychologist Karl Lashley suggested that memory traces, which he called engrams, had to exist somewhere in the brain. But a search

of animal brains failed to reveal them. “I sometimes feel, in reviewing the evidence on the localization of the memory trace,” wrote Lashley, “that the necessary conclusion is that learning [memory] just is not possible.” Despite that, the search for engrams, in the form of long-lasting change in the brain, is continuing. “Some mathematicians and electronic engineers are thinking about how you might be able to have an electric circuit in the brain with stable properties to account for memory,” says biochemist Arthur Roach of Toronto’s Mount Sinai Hospital. “But we molecular biologists think there has to be some kind of physical thing underlying these circuits.”

In one approach to the problem, scientists studying simple organisms have been able to show that the formation of short-term memory does pro-

duce neurological changes. Dr. Eric Kandel, a professor of psychiatry and physiology at New York City’s Columbia University, found that when he tapped the siphon, or spout, of the sea snail aplysia californica, the animal— after first pulling its gill away—eventually became accustomed to the tapping and ceased to withdraw it. By studying the snail’s nervous system—

which, despite its simplicity, closely resembles basic elements of the human brain—Kandel and his co-workers made a significant discovery. As the snail became used to the tapping, nerve cells in its gill-withdrawal circuit began releasing smaller amounts of the neurotransmitter chemicals needed to send a withdrawal command to other cells in the circuit. More recent research, says Vincent Castellucci, who worked with Kandel for 19 years, has shown that the transition

from short-term to long-term memory in the snail can be prevented by blocking the synthesis of new proteins in its nervous system—a further clue to the type of physical change required for memory formation. The next step, says Castellucci, who took over as director of neurobiology and behavior at the Clinical Research Institute of Montreal in January, is to determine “What are

these new proteins, what do they do, and what are the chemical signals that tell the neuron: make me some new protein?”

Link: Other neuroscientists have put forward detailed hypotheses about what happens when memories are formed. Belmont’s Shashoua, for one, discovered that when goldfish are trained to recognize associated events, certain types of calcium-containing brain proteins, which he calls ependymins, collected in the spaces between

neurons. When calcium levels dropped during the fishes’ training periods, the ependymins solidified to form a fibrous “matrix” on parts of cell surfaces. Shashoua says that the matrixes help strengthen the links—or synapses— between neurons and may play a major role in memory formation. Adds Shashoua: “We believe that ependymin is deeply involved in the process of making permanent circuits, and that these are memories.”

Still, some investigators claim that memories are not stored in a specific place in the brain, but instead re-form as experiences modify old memories and as sets of memories compete with each other for survival in the brain. This is the system proposed by neurobiologist Dr. Gerald Edelman of the Neurosciences Institute at New York City’s Rockefeller University.

Grow: In Edelman’s theory, information entering the brain is sorted according to category by groups of cells that communicate among themselves. At the same time, Edelman maintains that brain development is governed not only by the genetic instructions contained inside brain cells—which determine the general structure of the brain—but by a process similar to Darwinian principles of natural selection. In the young human brain, trillions of possible connections can be made among neurons. But Edelman says that as an individual grows and contends with life, the connections between groups of linked cells that ensure the best response to the environment for that individual are strengthened—and less successful linkages are weakened. “Perception and memory do not consist just 9 of a passive reception of infor| mation,” said Edelman. “The ~ brain is involved in a creative, £ self-generative process.”

Until it is scientifically verified, Edelman’s elegant structure remains a hypothesis. As research proceeds, says Mount Sinai’s Roach, “I think we’ll find that every new piece of understanding about memory will lead to new levels where there will be many more questions to answer.” The emerging facts and theories at least help to explain why memory is sometimes unreliable. The miracle, of course, is that it exists at all.

MARK NICHOLS

MAUREEN ARGON