Why do only 15 percent of us use seat belts when we ride in cars? This question was raised by two psychologists, Paul Slovic of Decision Research in Eugene, Ore., and Norman Schwalm of Perceptronics in Woodland Hills, Calif., who conducted two studies designed to convince people of the risks of not using seat belts.

The psychologists, whose findings were written up by the American Psychological Association, concluded that most people simply don't understand the risks of riding without seat belts.

The National Highway Traffic Safety Administration estimates that half of the 29,000 people killed in light truck and car accidents last year might be alive today had they worn seat belts. But the odds of anyone being killed on a single trip -- about 1 in 4 million, according to the APA report -- are, in fact, minuscule. "So, if you don't use a seat belt on a given trip and you don't have an accident, that pattern of behavior is immediately rewarded," notes the APA report.

"However," it continues, "over a lifetime most of us take an average of 50,000 trips in a motor vehicle. That raises the probability, over an entire lifetime, to one in 100 of being killed in an accident and one in three of being seriously injured." Raising motorists' awareness of their lifetime odds of being killed or seriously injured, then, could be a key to greater voluntary use of seat belts.

The researchers concluded, however, that various messages used to encourage seat belt use had no effect on the people they studied. As a result, Slovic supports mandatory seat belt laws, which have been adopted in five states. The Department of Transportation will require air bags in cars unless states representing two-thirds of the population approve mandatory seat belt laws by 1989.

The statistics from other countries with mandatory seat belt laws suggest they work to some extent. According to James Nichols, chief of the program evaluation and support division of NHTSA, seat belt usage has risen by 50 percent in the more than 30 countries that have mandatory usage laws. The death toll from accidents has dropped between 20 and 30 percent.

A study done several years ago by D.F. Huelke, of the University of Michigan Medical School and the Highway Safety Research Institute in Ann Arbor, concluded that fatalities would have been reduced by 91 percent in rollover accidents had seat belts been used and by 71 percent in frontal crashes. Injuries would have been reduced in both by about half. His study also found that persons involved in rollover crashes were five times more likely to sustain serious or fatal injuries when they were ejected from the car than those who were not.

From a study of rural, high-speed accidents, he concluded that a combination of lap-shoulder seat belts and air bags would reduce fatalities by 34 percent, saving 9,248 lives a year, while use of a seat belt alone would reduce them by 32 percent, saving 8,704 lives.

The most recent DOT estimate, issued in 1983, was that death and injuries attributed to motor vehicle accidents cost society $57 billion a year in medical care, property loss and loss of productivity -- a significantly larger sum than the $37 billion annual trade deficit with Japan that is dominating the headlines.

The Insurance Institute for Highway Safety, a research and communications enterprise supported by the casualty and property insurance industry, has endorsed mandatory seat belt laws and passive restraints. "The key here tends to be that there has to be enforcement of the law and some reasonable degree of public support," said Brian O'Neill, the institute's executive vice president. "The experience elsewhere is if there is not some reasonable degree of public support it will be a failure.

"A mandatory seat belt law is not a sufficient answer to the problem," he said. "It's a partial answer. We need improved restraint systems such as air bags. It would multiply the benefits of a seat belt law. You get protection at higher speeds for those who wear the belts and you get some protection for those who won't. We won't ever get 100 percent of the people wearing belts."

The research on why people won't wear seat belts strongly suggests that O'Neill is right: Some people, perhaps a great many, simply won't wear them. But if the public can be persuaded to think of its odds spread out over a lifetime, voluntary compliance might rise and support for mandatory laws and passive restraints is bound to follow. As things stand now, however, 85 percent of us routinely gamble with lives, and society as a whole pays the bill.