When Marcel Proust dipped a piece of toast into a cup of tea in January 1909, one thing led to another:

“And suddenly the memory revealed itself … and the whole of Combray and its surroundings, taking shape and solidity, sprang into being, town and garden alike, from my cup of tea.”

“The classic case of recovered memory is Proust,” says William P. Banks, professor of psychology and co-editor of a 1996 book titled The Recovered Memory/False Memory Debate.

Banks, co-founder and editor of the journal Consciousness and Cognition, believes the mechanism of such recollections may be that sensory stimuli trigger a “splash of associations.” From these associations, the best matches may become conscious memories in a process that, he marvels, is faster than Google.

“Imagine being able to report within a second whether a book was in the Honnold/Mudd Library collection,” says Banks. “That is the scale and speed of our memory.”

The investigation of human memory has come a long way since Proust, who experienced not just a splash but a flood of associations, recounted in 16 volumes titled A la Recherche du Temps Perdu, originally translated as Remembrance of Things Past, or, more recently, In Search of Lost Time.

“We don’t really talk about memory any more as a unitary function,” says Nicole Y. Weekes, associate professor of psychology, who, with Richard S. Lewis, associate professor of psychology and neuroscience, is analyzing data from a joint study they conducted on stress and memory, funded by the National Science Foundation. “There are several different types of memory. You can talk about memory for how to do things, memory for facts, memory for events. There is memory that you are aware of, and memory of which you are not aware—such as how to ride a bicycle. There is little question at this point that the different types of memory are represented differently in the brain.”

In their study, Weekes and Lewis investigated psychological and hormonal levels of stress among students during examination periods and in the exam-free summer. “We’re just beginning to analyze the data,” says Weekes. “Thus far, if we can conclude anything, it looks like memory actually is improving during times of higher stress. And that raises a question: If memory is improving, what’s getting better? Is it that they’re getting better at processing or encoding information from the environment, or better at consolidating memory in the brain, or does stress help them to find a cue that provides easier access to the memory?”

The poignant case of Henry M., a Connecticut man who suffered from epileptic seizures as a youth and young adult, has been instrumental in advancing researchers’ understanding of the multilayered way in which memory processes work. After his hippocampus was surgically removed in 1953, when he was 27, he lost the ability to create new long-term memories—those that are accessible after more than a matter of minutes. He retained his slightly above-average IQ, as well as memories of his life before the surgery. Now elderly, he lives in a kind of Proustian netherworld, knowing only a distant past and momentary present, always in search of lost time.

“We know that the hippocampus is critical in consolidating sensory information and putting it into long-term storage,” says Karen Parfitt, associate professor of biology. “So the big question in neuroscience is: How do the neurons in the hippocampus do that?”
Tens of thousands of papers already have been published in response to that overarching question, but a complete explanation has yet to unfold.

“The human brain has about a hundred billion neurons, and each of those neurons has a thousand to 10,000 synapses, so the complexity is staggering,” says Parfitt, who, with her students, studies the chemical signaling between neurons in the rat hippocampus. “Most neuroscientists believe that memories are stored via alterations in the strength of chemical synapses. A fascinating property of synapses in the hippocampus is that when they are used repetitively, the signal transmitted from one neuron to the next is strengthened somehow.”
“This increase in synaptic efficacy is more or less permanent, and is therefore referred to as ‘long-term potentiation,’ or LTP,” she says. “Our central goals in the lab are to elucidate some of the molecular mechanisms underlying LTP, and to reach a better understanding of how such mechanisms change with aging.”

Previous work by Parfitt has shown that hippocampal neurons of older rats fail to respond properly to norepinephrine, a neurotransmitter involved in attention and arousal. Such neurotransmitters help to initiate a cascade of intracellular chemical events that leads to the production of signaling molecules known as “second messengers.” One such second messenger is called cyclic AMP.

“Cyclic AMP seems to be extremely important in learning and memory, whether you look at a fruit fly, or a sea slug, or a rat, or a human,” says Parfitt. “The enzyme that makes it is highly conserved evolutionarily. That points to it being a very important molecule for survival. In fruit flies, if you overexpress the enzyme that is turned on by cyclic AMP, those animals learn a task in fewer trials; in other words, they become super learners. Drug companies are hot on the trail of cyclic AMP.”

So are Parfitt’s students, who work with high-tech electrophysiology and microscopy equipment in the new Seaver Biology Building. The study of age-related changes in the hippocampus is essential in understanding the memory decrements that occur with normal human aging, as well as with pathological conditions such as Alzheimer’s disease, Parfitt notes. “Clearly, a better understanding of this critical brain region may enhance our ability to learn and retain new information in our later years,” she says.

Another memory-related part of the brain, a group of cerebral convolutions called the insula, is a focus of investigation by Deborah Burke, W.M. Keck Distinguished Service Professor and professor of psychology. Burke’s groundbreaking work on tip-of-the-tongue experiences and other word-finding problems associated with aging has drawn a succession of major funding grants and has made her a widely quoted expert in the national and international media.
Recent studies have suggested that the left insula is important in the production of phonology, or memory for the sounds of words. Other studies have correlated advancing age with a loss of gray matter in the insula, even among people whose behavior is considered normal and who show no signs of dementia.

“For a long time I’ve been looking at word-finding problems, and our interest was whether the tip-of-the-tongue experience may in any way be related to atrophy in the insula,” says Burke.

At the Centre for Speech and Language at the University of Cambridge in England, with the help of Pomona alumni Meredith Shafto ’96 and Phyllis Tam ’04, Burke tested a range of young to elderly adults for tip-of-the-tongue experiences—or TOTs. Using pictures of famous people, such as British soccer stars, and descriptions of palindromes, which are words or phrases spelled the same way forward and backward, they found, as expected, that TOTs in the study group increased with age. They also administered a nonverbal cognitive test to the subjects. Functional magnetic resonance imaging was used to assess the extent of atrophy in the left insula as well as in other brain regions in the research participants.

“We wanted to compare the number of tip-of-the-tongue experiences that a person had with what the scan showed us about the amount of atrophy in their brain,” Burke says. “We found that the more atrophy a person had in the left insula, the more TOTs that person would have. And it’s only on the left side, the hemisphere involved with language. We also found that the size of the insula has no relation to how they did on the nonverbal test. So it seems that this isn’t an area that’s correlated with cognitive performance in general.

“It supports a model I’ve worked on for a long time that suggests that the problem here is in retrieving the sounds of words,” Burke says. “It’s not a problem with one’s store of semantic information or conceptual information.”

Correlating a phonological retrieval deficit in older adults with atrophy in a specific place in the brain is a significant advance. Still a mystery is what causes atrophy of brain structures such as the insula in people who seem to be aging normally.

Abnormal aging continues to exact a devastating toll through Alzheimer’s disease, a condition in which amyloid proteins form plaques and tangles that block the normal functioning of neurons, resulting in progressive cognitive dysfunction and death.
Laurel Beckett ’68, a professor in the Department of Public Health Sciences of the School of Medicine at the University of California, Davis, took part in the first broad population-based study of Alzheimer’s, in the late 1980s.

“It was thought at the time that Alzheimer’s disease was this unusual condition that happens to a relatively few people when they’re older,” she says.

Beckett, a biostatistician who works at the juncture of mathematics and medicine, estimated in that study that four million Americans had Alzheimer’s. The shocking number propelled the disease to the front ranks of national public health concerns. The total is now believed to be closer to five million, and it is estimated that 16 million Americans may have Alzheimer’s by 2050, when the upper bound of Beckett’s 1990 figure is correlated with data on anticipated life spans and the expanding population of older age groups.

“The fastest-growing age group is 85 and older, and our estimate suggested that as many as 47 percent of this group might in fact have Alzheimer’s disease,” she says. “This also includes a range from mild to moderate disease—it’s not all people who have serious dementia and can’t function.”

Beckett is among the leaders of the Alzheimer’s Disease Neuroimaging Initiative, a $60 million public and private effort to standardize methods of tracking the illness from its earliest clinical stages to reduce the time and cost of clinical trials of possible treatments. Danielle Harvey ’96, who works with Beckett at UC Davis, has a role in the initiative as well.

Beckett also has been involved since 1994 in the Religious Orders Study in Chicago, which is closely following about 1,000 priests, nuns and brothers from Catholic religious orders, all of whom have agreed to donate their brains for autopsies after death.

“These are people for whom we will have a longtime, very detailed history, so the hope is to determine what was going on in their brains during all that time in which they were or were not declining in function, what may have been driving the onset of pathology, and what may have been protecting those who show little or no decline,” she says.

Old age, Beckett stresses, is not inevitably shadowed by a twilight of memory.

She says that maintaining elevated levels of cognitive activity and physical fitness may help to create a “neural reserve” that counteracts memory deterioration.

“If you look at individuals, the most striking thing is the very, very wide variation in patterns,” she says. “There are people who stay mentally and physically fit into their 90s. Even among those with Alzheimer’s, the paths can be quite different. A better understanding of the disease can only help us, and there has been a real increase in knowledge. There’s a great deal of promise.”

Probing the mysteries of memory might help answer questions raised by Proust when he sipped his tea a century ago and felt a sudden surge of emotion that gave rise to remembrance:

“Whence could it have come to me, this all-powerful joy? … What did it mean? How could I seize and apprehend it? … It is plain that the truth I am seeking lies not in the cup but in myself.”

Michael Balchunas is a freelance writer living in Claremont, California.