Individuals with a certain type of brain damage live a kind of shadow life. They can converse, get around perfectly well, and even learn many types of skills that may be new for them: how to ride a bike, for example.

But these individuals can’t remember a single fact from day to day, hour to hour, and even minute to minute. In effect–and even though some retain memories dating back before the damage occurred–they’re imprisoned in the present.

These individuals have lost the use of a small but critical paired structure buried deep in the brain. Called the hippocampus because it’s vaguely shaped like a seahorse, the structure is critical in forming memories, says MIT’s Matthew Wilson.

“Patients with serious hippocampal damage can learn and remember new physical skills,” he notes, “but they can’t form memories of the people, places and events in their lives.”

Wilson, an assistant professor of brain and cognitive sciences, has worked hard to puzzle out how the hippocampus works as catalyst for memory formation. Much of his current research is focused on the intriguing question of whether the hippocampus, in concert with memory machinery elsewhere in the brain, actually helps ensure that particular memories will become enduring ones as we sleep.

Feats of memory

Such research is part of a vital enterprise. The memory is a powerful system, and while the feats of a select few–like the individuals who can flawlessly recite lists of 100 or more numbers heard decades earlier–draw the spotlight, the memories of average individuals are impressive, too. Many of us, for instance, have memories that can store images of an estimated 10,000 or more faces.

When memory breaks down, though, the results can be tragic. Alzheimer’s disease, which afflicts 4 million in the U.S. alone, often so damages victims’ memories they lose any sense of identity.

What’s to be done? Wilson thinks we need explore just how it is that tiny electrical and chemical events in the brain can translate into remembering that song, that fragrance, or that long-ago experience.

The faculty member, who majored in electrical engineering as an undergraduate, says he was “always interested in cognition and intelligence.” So, as a grad student at Caltech, he set out to create a computer model of one part of the brain.

“What I discovered,” he says, “is that building such a model required a degree of understanding we didn’t possess.”

Coordination matters

He thus decided he needed to probe the brain at the level of its individual cells, and of the minuscule links among those cells. (One of the brain’s striking features is that each of its billions of neurons-the organ’s computing cells-may have thousands of links to other neurons.) He put his engineering skills to work again, creating a unique instrument that can detect which of100 or more brain cells packed into a pinhead-sized space are firing.

Working with Nobel laureate MIT biologist Susumu Tonegawa, Wilson used the device to probe cells in the hippocampus. The researchers found that when the lab mice under scrutiny navigated a maze, the targeted cells fired in predictable succession like bulbs on a Las Vegas billboard. “We could basically track where the animals were by monitoring which cells were firing,” says Wilson.

The researchers also tested animals where those same cells lacked a key neurochemical–an approach made possible by Tonegawa’s heralded technique for “knocking out” individual genes in selected cells. In this group, the cells fired helter-skelter, and the animals, unlike their untreated counterparts, failed hopelessly at learning the lay-out of mazes.

The study, among other lessons, highlights the extent to which memory is about coordination. Wilson, noting that his mother was trained as an opera singer, makes the point in a vivid way. “If you went to the same opera 1,000 times,” he says, “but your brain couldn’t coordinate your memories of different aspects of the work, you wouldn’t remember that opera.”

Dream discoveries

Wilson has since begun exploring the extent to which dreams play a role in memory. “The idea,” he says, “is that we often dream about things that are significant for us, and that this can lead to the formation of long-term memories.”

That doesn’t mean a bone-chilling nightmare is an example of memory consolidation at work. Wilson says, in fact, that the dreams involved in consolidating memories are unlikely to be the kind that wake us up, or even enter our conscious thoughts.

He also notes, though, that the hippocampus actively communicates during dreams with parts of the brain where long-term memories are lodged. There are signs, he adds, at least some of this back-and-forth represents a sorting process that leads us to retain some memories while letting others slip away.

Wilson is quick to concede that finding patterns in the way selected brain cells fire during sleep is a long way from defining a specific role for dreams in memory, much less understanding the workings of the memory as a whole. Still, he says science doesn’t have to pin down the role of all memory-related brain cells–which doubtless number in the billions–to help those with impaired memories.

“One working definition of understanding memory is to ask, Do we know enough to intervene in a way that improves the memories of people like Alzheimer’s patients?” he says. “I believe that that will happen well within my professional lifetime, and maybe within the next several years.