Seeing and controlling chaos in the brain.

Just as the increasing sophistication of microscopy techniques have enabled neurobiologists to view nerve cells as they never could before, tools borrowed from mathematicians and physicists are now providing new insights into the electrical activity of these cells.

Predictable patterns do exist amidst the din of electrical impulses used by nerve cells to communicate with one another, says Steven J. Schiff of Children's National Medical Center in Washington, D.C. Until now, discerning such patterns has proved elusive. Although nerve cells sometimes fire regularly, they often change their timing unpredictably

Actually at least in an experimental setup, the patterns can be chaotic: They do have some predictability even though they look irregular, explains William L. Ditto, a physicist at the Georgia Institute of Technology in Atlanta. By anticipating the resulting patterns in the tested tissue and controlling the degree of irregularity, he and Schiff think they may one day successfully stifle the impulses that lead to epileptic seizures.

Working with Mark L. Spano at the Naval Surface Warfare Center in Silver Spring, Md., these researchers demonstrated this chaotic behavior - and control of that behavior - in thin slices of a part of the rat brain called the hippocampus. Bathed in a concentrated potassium solution, some cells in these slices synchronize their firing as "bursts." Parts of the slice then become epileptic.

With electrodes attached to the slice, the team tracked the timing between each impulse, or spike, and of changes in this timing and found nonlinear procedures for describing what they saw, they reported in the Aug. 25 Nature.

Once they determined the chaotic, albeit tractable, nature of the spiking patterns, they began fiddling with it. By electrically jolting the slice briefly at calculated moments, they tried to alter the firing of cells, Ditto says.

After 91 tests of 22 slices from 9 rats, they were convinced: "We can make [the slice pattern] chaotic; we can make it periodic; we can make it do almost anything," Ditto says. Seizures develop because too many nerve cells start to fire at once in too regular a fashion. Ditto suspects the so-called "anti-control" whereby they make the pattern more chaotic may help prevent seizures.

While an impressive demonstration of chaos and of the application of chaos controls to a complex biological system, the experiment still does not prove such chaos really exists in an intact brain, cautions Walter J. Freeman, a neuroscientist at the University of California, Berkeley. "The questions of whether [these dynamics] occur physiologically as part of normal brain function or are induced - these are all issues being hotly debated," he says.

Still, Schiff is unfazed. Before he surgically removes a part of a human brain that instigates seizures, he first monitors the electrical patterns to determine where these aberrant signals originate. The patterns he observes are more variable, but they still parallel measurements taken from the slices, he notes. Thus, he thinks studies in these slices will provide valuable information about human brains.

Also, chaos control looks promising in other organs. Two years ago, Spano and Ditto worked with cardiologists and demonstrated that hearts, too, exhibited chaotic behavior that the researchers could control (SN: 9/5/92, p.156). That data compares well with heartbeat data collected from people.

Schiff hopes a similar approach will help some epileptics avoid surgery to remove parts of their brain because drugs could not control their seizures. The idea of using electric fields and currents to alter brain activity is not new, but the use of chaos control procedures offers a chance for "more subtle intervention and more accurate control," says.