President Reagan, responding to a secret protest from the Joint Chiefs of Staff, has dropped the approach he took at the December summit meeting with Soviet leader Mikhail Gorbachev to bridge the longstanding U.S.-Soviet differences over the Strategic Defense Initiative (SDI) research program and the 1972 Antiballistic Missile (ABM) Treaty, U.S. officials said yesterday.

On Wednesday, Reagan secretly approved instructions to U.S. negotiators in Geneva telling them to demand Soviet acquiescence to the disputed "broad interpretation" of the ABM Treaty that would allow realistic testing in space of elements of SDI. Gorbachev refused to accept any such formulation when he was in Washington.

Reagan's new instructions were a response to a recent letter from the chairman of the Joint Chiefs, Adm. William J. Crowe Jr. to national security adviser Colin L. Powell. In that secret letter, Crowe told Powell that the vague formula on SDI research agreed to by Reagan and Gorbachev at the summit was unacceptable.

The summit formula, produced after heated argument between U.S. and Soviet officials, allowed each side to conduct missile defense research "as required" within the constraints of the 1972 ABM Treaty. Both sides acknowledged that this was, in effect, an agreement to continue to disagree on SDI.

Crowe told Powell that the ambiguity of the summit statement would likely encourage future disagreements between the United States and the Soviet Union over antimissile research. Speaking for the joint chiefs, Crowe said in his letter that the United States should either negotiate with Moscow a specific list of acceptable SDI tests, or agree that the ABM Treaty would remain in force for only five more years, clearly permitting unlimited testing and development when that period ends.

Reagan rejected both of Crowe's alternatives, instead instructing his negotiators in Geneva to seek Soviet approval for the administration's longstanding view that the ABM Treaty should be read as allowing extensive realistic SDI testing in space. Majorities in both houses of Congress and Nixon administration officials who negotiated the ABM Treaty have disputed this reading of the treaty.

News of Reagan's latest shift on the ABM issue left numerous administration officials pessimistic that the two sides will be able to circumvent the acrimonious dispute over missile defenses and press forward with a treaty sharply reducing strategic offensive nuclear weapons, the Reagan administration's chief arms control objective.

"This is a real strategic arms 'treaty-buster,' " one official said yesterday. "It means the United States and the Soviet Union may never come to an agreement" on missile defenses, which the Soviets have insisted is a precondition for reductions in offensive nuclear weapons.

Crowe registered the chiefs' dissent in response to a draft U.S. treaty on missile defenses that incorporated the Reagan-Gorbachev summit formula. The four-page draft treaty adopted the summit language committing both sides to observe the ABM Treaty "as signed" in 1972 for an agreed period, while permitting research, development, and testing "as required."

Powell, appearing at a news conference moments after the summit ended, declared that Soviet acceptance of this formula allowed unrestricted SDI tests under the disputed "broad" interpretation of the ABM Treaty. Soviet spokesmen firmly disputed that view, and several weeks later, Powell conceded in a letter to Congress that the Soviets disagreed, and had explicitly reserved the right to "suspend implementation" of a strategic arms treaty if the United States conducted such tests.

Crowe's concern, expressed in the letter, was that a treaty incorporating the vague summit formula might leave the Soviets free to resume building up their offensive forces if they decided that U.S. SDI tests violated the ABM Treaty "as signed."

At the same time, Crowe expressed concern that the Soviets might exploit the vague formula to test and develop defensive weapons in ways Congress would not allow the Pentagon to match. Congress has insisted that all U.S. tests conform to the traditional, "narrow" interpretation of the treaty.

Crowe's suggestion that the administration negotiate a detailed list of acceptable space tests is the same position long advocated by senior State Department arms control adviser Paul H. Nitze. The Soviets also have expressed interest in agreeing on a list of permitted research activities, but Reagan has repeatedly rejected the idea.

Crowe's second alternative, to propose U.S. adherence to the ABM Treaty for a five-year period, was based on his prediction that little or no weapons technology is likely to be ready for realistic space testing in the next five years.

The administration's formal position in Geneva is to adhere to the treaty for seven years, while the Soviets have demanded that each side adhere to it for 10 years.

In rejecting Crowe's proposals, Reagan ordered U.S. negotiators not to discuss the length of time that both sides must adhere to the ABM Treaty until the Soviets have flatly accepted the so-called broad interpretation.

Reagan's objective was apparently to leave open the possibility that he would propose the five-year period if the Soviets again spurn the "broad interpretation."

At the same time, Reagan accepted a recommendation by the Joint Chiefs of Staff and the State Department that the draft U.S. missile defense treaty require both sides to provide six months' notice of withdrawal from the ABM Treaty at the end of an agreed period of commitment to it.

The civilian leadership of the Defense Department has long argued against requiring such a notice on the grounds that announcing an intention to withdraw from the treaty would be politically unpopular. In a U.S.-Soviet compromise, the summit communique stated that "each side will be free to decide its own course of action" at the end of an agreed period.

But Reagan accepted the chiefs' view that U.S. security could be endangered without at least six months notice of Soviet intentions to deploy a ballistic missile defense.