The way older people view themselves affects how they walk, a Harvard study has found, with new evidence showing that boosting seniors' self-image could result in a mind-over-matter improvement in their stride.

"Our study suggests the importance of psychological factors in physical performance," said Jeffrey M. Hausdorff, an assistant professor of medicine at Harvard Medical School and a researcher at Beth Israel Deaconess Medical Center in Boston.

In the study, one simple intervention--tapping into participants' subconscious by flashing words on a computer screen so fast they cannot be read--increased walking speed by nearly 10 percent in older people. Hausdorff called that finding a "surprisingly large benefit."

"The societal stereotypes of aging may in fact have a direct impact on physical functioning in old age," said Becca Levy, a social psychologist who collaborated on the new research. The findings suggest that positive reinforcement helps to "improve functioning in old age" and to soften the stereotype that "physical and mental decline is inevitable" with old age, she said.

When elderly people's walking speed slows down significantly, numerous studies show, they become much more likely to require nursing home care and much more susceptible to serious illness and earlier death.

An important walking characteristic that affects the ability to balance is known as "swing time," which is the amount of time spent with one foot off the ground. Those who lose the spring in their step and shuffle their feet rather than raising them have less swing time.

Studies of seniors' walking performance have traditionally focused on diseases and physical problems, not mental attitude, said Hausdorff. His new study, published in the Journal of the American Geriatrics Society, sought to test how self-held stereotypes about aging may affect the walking pattern, or gait, in older persons.

He studied 47 men and women aged 63 to 82 who described themselves as healthy. The participants, randomly assigned to one of two groups, were asked to play a 30-minute computer game in which words flashed briefly on the screen. While the words appeared too quickly to be read (they looked like dark flashes), they may have stayed long enough to deliver a subliminal, or subconscious, message to the participants. Before and after the computer game, each participant was timed walking 300 feet to determine walking speed. Swing time was also measured for two minutes with special equipment.

When the researchers compared walking patterns, the group exposed to positive words about aging, such as "wise," "astute" and "accomplished," showed a significant increase in walking speed and swing time. The other group, which got negative words such as "senile," "dependent" and "diseased," showed no significant differences in either walking speed or swing time.

The team concluded that "negative self-stereotypes of aging" likely play an important role in the loss of walking speed and swing time as people grow older.

The experimental approach in this study--testing the effects of stereotypes in the elderly by tapping into the subconscious--was developed by Levy, who was at Harvard when the study was conducted and is now an assistant professor at Yale University School of Medicine. She had earlier used the same technique to show that positive subliminal words improved memory while negative ones led to memory decline in the elderly, but not in younger participants. More recently, Levy has shown that this approach can influence handwriting, self-confidence and even the will to live. She said that a soon-to-be-published study found that elderly individuals who received positive psychological reinforcement without being aware of it "had a stronger will to live than those shown negative images."

Hausdorff noted that perceptions of aging have already changed in recent years. "Older people live very healthy lives, and stereotypes are being broken, from John Glenn going back into space to George Bush jumping from a plane again," he said.

He hopes that the new study, funded in part by the National Institute on Aging and the National Institute of Mental Health, will promote further efforts to improve self-image and maximize walking ability in the elderly by using psychological as well as physical resources. Previous efforts have targeted exercise to help walking performance. The improvement found in the short computer experiment is similar to the amount of walking speed improvement seen in studies in which the elderly participated in weeks or months of rigorous physical exercise and weight training to improve gait or walking speed.

Follow-up research may seek to identify how long the improvement in walking lasts after positive input and what benefits there might be with less healthy groups, such as Parkinson's disease patients.

The computer study is the latest in a series Hausdorff has conducted to unravel the mysteries of walking. His gait laboratory has developed an unusual method for measuring walking patterns, using thin, pressure-sensitive shoe pads attached to a small recorder worn on the ankle. The device can monitor a person's natural walking rhythm by measuring hundreds or thousands of stride-to-stride variations.

Using computer analysis, the Harvard group has examined stride patterns over time with a technique borrowed from physics to see how much variability exists in walking rhythms.

A Hausdorff study of elderly persons with a history of falling for no apparent reason found that they had more variation in their walking patterns compared with those with no history of falling. The study examined 35 men and women over age 70 who were able to walk independently for at least six minutes. The walking monitors detected two to three times more variability in the walking patterns of those with a history of falling, indicating a significant but subtle increase in walking unsteadiness. However, the study found that "elderly fallers did not walk more slowly than elderly nonfallers."

Ultimately the goal of such research is screening of elderly patients to predict who is more likely to fall and to find ways to try to prevent such falls. Falls are a severe problem in the elderly, since they often lead to broken hips or other injuries and may result in a loss of independence and increase the chance of their being placed in a nursing home. While walking speed does not appear to be a predictor of falls, said Hausdorff, variability in walking rhythm may hold some clues. More research is needed to confirm whether gait variability or unsteadiness precedes or follows falls, since it is possible that once people fall they change the way they walk because of fear of falling again. At the moment, Hausdorff's gait-measuring technology is only available in the research setting and not in doctors' offices.

Growing evidence shows that "changes in walking are not simply due to aging and should not be accepted as an inevitable part of aging," Hausdorff says. "Walking problems could be a sign something else is wrong. If you fall, check with your doctor," he advises. Abnormal slowness or unsteadiness could be related to diseases with a neurological basis (such as Parkinson's or Alzheimer's), heart disease, arthritis, vision problems, medication effects, obesity or inactivity leading to muscle weakness.

For those who are healthy, he recommends "exercise, exercise, exercise to maintain muscle strength and balance" as one way to reduce the risk of falls and walking problems.