A controller working one of the 33 sectors of airspace at the Federal Aviation Administration's center at Leesburg jumped from his seat, swearing. "Look," he commanded. "Do you see anything moving?"

The blocks of numbers and figures that identify airplanes had stopped marching across the screen. The computer that drives the radar had "frozen," but the eight airplanes the controller was handling most certainly had not. They were hurtling through space at 600 mph, and the controller had lost the ability to tell where they were going.

The incident happened five years ago, to the FAA's embarrassment, with reporters in the room. But computers still freeze and controllers still swear, and when it happens it is stressful.

Most controllers have been complaining about stress for years. But the central issue in the current dispute, at least in public perception, is salary. The average controller makes $33,000 a year, and the Professional Air Traff Controllers Organization (PATCO) demanded a $10,000 across-the-board raise.

Dr. Robert M. Rose, chairman of the Department of Psychiatry and Behavioral Sciences at the University of Texas Medical Branch in Galveston, blames PATCO as well as the FAA for keeping stress and working conditions in the background during the past decade, a period that has seen numerous air traffic slowdowns and now one strike.Rose headed a team that studied controller stress in 1977 and 1978.

Controlling air traffic, Rose said in an interview yesterday, "is not uniquely stressful. However, we feel there is significant data . . . to support a conclusion that air traffic controlling does deserve to be considered among the more stressful occupations.

"The stress is not due so much to the frequency of high-pressure workload," Rose said, but because controllers know that "if an error is made, it may have uniquely grave consequences. It is the awareness of such a possibility, arising without much warning, that probably constitutes the greatest potential for stress for the air traffic controller."

The study was not looking at moral questions, Rose said, but "data showed this turned out to be a very important issue." More than 90 percent of the controllers in his sample of 400 from Northeast towers and centers, he said, felt the FAA does not reward or promote on the basis of good performance; three-fourths felt that the FAA was unconcerned about controllers' problems.

All controllers, he said, know other controllers who have been involved in a so-called "near-miss" or plane crash, and "they know what has happened to that man afterwards."

He cited the case, and PATCO's David Trick confirmed it, of the controller who handled a flight that crashed short of the runway in New York in 1976 and killed 112 people. Radio transcripts showed that preceding pilots had suggested that the runway be closed because of severe wind conditions. The controller was unable to work for nine months after the crash, Rose said, and died two years later of an aneurysm.

Another indication of job problems, Rose said, was the finding that 55 percent of the controllers in his sample were heavy drinkers, although "alcoholism is not high, about 9 percent," Rose said.

Rose's 1978 study has been rewritten, by Rose, "into layman's language," at the request of the FAA, PATCO officials said they have been denied the new version under a Freedom of Information Act request.

FAA spokesman Dennis Feldman said Rose's rewrite, delivered in May, was 140 pages long (the FAA expected 30 to 40 pages) and contained "new material which we're reviewing now."

Rose's primary concern, he said, is that "they'll fire these 8,000 or 10,000 controllers and then, eight to 10 years doen the road, they'll have to do it again because no one will attend to these long-running problems. There's a real cancer there."