New studies suggest that people infected with the AIDS virus may suffer a loss of mental function long before they experience any other symptoms of the fatal disease.

Tests conducted on infected but apparently healthy men showed an unusually high number had impaired coordination, cognitive difficulties or abnormal results from images taken by magnetic resonance devices. Until now, specialists thought the virus caused no deleterious effects until other symptoms of acquired immune deficiency syndrome appeared.

The findings have alarmed public health officials and recently prompted the Defense Department to remove service personnel who have tested positive for the AIDS virus from stressful, high performance jobs such as flying aircraft, handling nuclear weapons or working on sophisticated machinery.

Health experts expect the results, which scientists describe as preliminary, to ignite a new round of debate over the need for widespread AIDS testing, particularly for those whose jobs may require unusual dexterity.

"This could look like Alzheimer's disease in a young person," said Dr. Edmund Tramont, director of AIDS investigators at the Walter Reed Army Institute of Research. "We just aren't sure when the clinical manifestations of HIV infection begin, so the smartest thing to do seems to exclude those who are infected from critical tasks. On balance it is just better to be safe than to take a risk."

Scientists have known for some time that the human immunodeficiency virus (HIV) that causes AIDS enters the cental nervous system and remains there. The great majority of AIDS patients experience some mental impairment, and AIDS dementia has recently been included by the Centers for Disease Control as a manifestation of the illness.

But researchers were surprised to find that even apparently healthy individuals show definite signs of problems that could harm their judgment severely.

In a study published in the current issue of the Annals of Internal Medicine, Dr. Igor Grant and colleagues from the University of California at San Diego recruited a sample of 55 patients and controls from a group of homosexual men in San Diego.

They broke the sample into four groups: 15 with AIDS, 13 who had a less severe form of the disease, AIDS-related complex (ARC); 16 who tested positive for the AIDS virus but who showed no other symptoms, and 11 healthy men who were not infected.

The researchers gave each subject a battery of nine separate neurological and psychological examinations designed to assess their mental capabilities. They then rated each man's performance as definitely impaired, probably impaired, or unimpaired by standardized criteria.

The rate of impairment was 9 percent in the healthy control subjects, but 87 percent in the patients with AIDS. The ARC group had 54 percent impaired. Most surprising to scientists, the rate for the group that was infected but not ill was 44 percent.

"I was stunned by the results," said Grant, professor of psychiatry at UCSD and a senior researcher at the San Diego Veterans Administration Medical Center. "I expected the large number of AIDS patients to show deficiencies. But to have 44 percent of the seropositive people show problems, that's dramatic."

Grant said that studies have shown that as much as 10 percent of randomly selected people show impairment.

"If the number had come up 20 percent I would have said that is no major difference in such a small study," he said. "But we had more than a fourfold increase and that is extremely significant."

Grant said that larger studies confirming the results should be carried out before any specific policies are written as a result.

But senior military officials and Public Health Service officers have been meeting for months to discuss the implications of Grant's study and several others that have produced similar results. Military officials, who have said in the past they are concerned that stressful jobs would be harmful to the health of infected service personnel, have argued that infected people should not hold sensitive jobs.

"Given the current trend of scientific data . . . the working group was extremely concerned about the continued presence of HIV positive personnel in high-risk, technically complex or sensitive positions," a working group in the office of the chief of naval operations wrote in a Sept. 24 confidential memo that was sent to senior military and Public Health Service officials.

Public Health Service officials said they are concerned that the results from the Grant study and several others under way will be used to justify calls for wide testing of the population.

"This is going to be the next great area of debate," said a senior administration official who asked not to be named. "It's going to cause a fight no matter how you do it. If the military tells its pilots they can't fly if they test positive, how can TWA say anything different. Before you know it, every trucking company in the United States is going to say we need those AIDS tests, too."

Researchers have begun long-term studies with greater numbers of subjects. And they have urged caution until the results are in.

"This would be a very upsetting way to make national policy," said Dr. Bradford Navia, a fellow in neurology at the Massachusetts General Hospital, who has done research on HIV infection in the brain. "We need to go slowly on this subject with careful studies, not just rush off with sentiment."