Two new studies have raised urgent questions about how well highway construction can relieve Washington's severe congestion, finding that adding road capacity by itself generates significantly more traffic.

The findings, presented yesterday at a national conference of transportation experts, showed that between a quarter and half of the new capacity in the region was used up simply because more motorists were attracted to it.

This represents the first research that examines the Washington area to gauge an effect called "induced travel." This is the amount of added traffic that occurs, for instance, when motorists take advantage of a new lane to make more or longer trips or to switch from transit to driving. In the long run, induced travel also can occur as the wider road allows nearby development to accelerate, producing even more traffic.

Although the studies do not conclude that building or widening roads is fruitless, they show that it is more difficult to build your way out of congestion--and more expensive--than many traffic engineers have thought.

"If you're simply widening congested roadways with the idea you'll get rid of congestion, that's naive," said Lewis M. Fulton, an American policy analyst based at the International Energy Agency in Paris and author of one of the reports.

His study used 26 years of data from every county in Maryland, Virginia and North Carolina. In the Washington-Baltimore metropolitan area, the analysis concluded that about one-third of the added road capacity on main highways--whether new lanes or entirely new roads--was used up by induced travel; every 10 percent expansion in roads led directly to a 3.3 percent rise in the number of vehicles driving on them.

The statewide results from Maryland and Virginia were even more dramatic, showing that about half of added road capacity was filled up because of induced travel. A 10 percent expansion in roads led to a 4.5 percent increase in motorists in Maryland and a 5.1 percent increase in Virginia.

The second study, conducted by Robert B. Noland, a former transportation analyst at the Environmental Protection Agency, reviewed 15 years of data from major American metropolitan areas. He found a somewhat smaller--but still highly significant--induced effect. The research showed that a 10 percent expansion in roads produced a 2.8 percent rise in travel.

Some of the effects identified in both studies were immediate, and the full traffic increase occurred in two to four years.

Neither study claimed that most of the region's traffic has been caused by the induced effect. Instead, the research acknowledged the dominant role played by growth in population, wealth and other economic and demographic factors.

But the question remains whether expanding roads will ease the congestion caused by all these factors or simply make it worse, according to experts at the meeting yesterday of the Transportation Research Board, an arm of the National Academy of Sciences.

The two studies are part of an escalating discussion in the last two years among planners, engineers and scholars over induced travel.

In November, the EPA convened an advisory panel of academics to review previous drafts of the studies by Fulton and Noland. Several scholars warned that the research might not adequately handle the chicken-and-egg question of whether building roads causes traffic or whether planners build roads in anticipation of new traffic. Fulton and Noland said yesterday that they have revised their work and were confident of their conclusions.

Though the concept of induced travel has often been challenged by road builders and highway engineers, the findings yesterday were readily accepted by two top Washington area planners who participated in the discussion.

"I think this is quite reasonable," said Ronald F. Kirby, chief transportation planner for the Metropolitan Washington Council of Governments. "Clearly, highway construction has an impact on land-use effects." He said his agency is grappling with how to better gauge the effect of expanding highways, in particular how it fuels development along the widened road and in turn produces more traffic.

Neil J. Pedersen, Maryland's chief highway planner said: "In Maryland, from a policy perspective, we accept induced travel does occur when additional highway capacity is added." He said the proper lesson is not to forgo road building but to direct it toward areas where the state wants to foster economic development.

Indeed, researchers agreed that building roads can still be crucial for fostering economic development, even if the potential for easing congestion is limited.

The studies' conclusions got a frosty reception from Bob Chase, of the Northern Virginia Transportation Alliance, which advocates road building. "You may need more capacity than is currently thought, but that doesn't lead to the conclusion that it isn't still the best solution," he said.

But Chris Miller, president of the Piedmont Environmental Council, said the latest findings demonstrate that new roads are likely to promote sprawl rather than ease traffic: "This confirms that it doesn't solve the problem. It stimulates more of the problem."

CAPTION: Aerial view shows backup on Interstate 95 north of Occoquan. The studies concluded it is more difficult to build your way out of congestion than engineers have thought.