A photo caption yesterday transposed the identifications of television network officials Warren Mitofsky of CBS and George Watson of ABC.

The heads of the three major television news organizations told Congress yesterday that they will project the outcome of the 1984 presidential election on the basis of exit polls and sample precincts, despite complaints that similar projections in 1980 may have cut down the vote in western states and affected results of other elections.

Saying it went against the grain to withhold or delay reporting of the vote projections, CBS News President Van Gordon Sauter and ABC News Vice President George Watson said that if Congress is worried about the problem it should set uniform poll-closing times across the country.

NBC News President Reuven Frank declined to suggest a remedy, but agreed with the others that "it would be inconsistent with traditional journalistic standards" to delay the presidential vote projections until the polls have closed in the West.

Their position--voiced at a hearing of two House subcommittees--left western representatives of both parties fuming. Rep. Al Swift (D-Wash.) told them, "People all over this country just think you're wrong."

The issue arose in the 1980 election, when Ronald Reagan rolled up such a wide margin over Jimmy Carter that Reagan's probable victory became apparent early in the evening.

NBC News, using "exit polls," interviews with voters leaving polling places, called the election for Reagan at 5:15 p.m. Pacific time, almost three hours before the polls closed. The other networks, relying on partial voting returns from scientifically selected sample precincts, came along a bit later--but still before the polls closed--with similar projections.

There were reports of people leaving the voting lines or turning back on their way to the polls when they heard the news, and some candidates who were defeated in close races in the West blamed their losses on the networks.

All three spokesmen yesterday said that although one study partially financed by ABC News suggested a 20 to 25 percent falloff among prospective voters who had heard the projections, the survey research on voter-discouragement factors is scanty and conflicting.

But when Rep. Timothy E. Wirth (D-Colo.), chairing the hearing along with Swift, asked the network executives what they would do if there were clear evidence of a 20 percent falloff, they demurred. "I'd be very, very troubled by it," Sauter said. "The tension would be acute. But it would not be fair to my organization to give an answer off the top of my head."

Watson said, "I don't know what we would do, but once we start putting a filter on what it's good for people to know, I worry."

Frank said, "I would support any effort to find a legislative remedy, but my responsibility is to give people the news."

The favored remedy of the broadcasters is a system that would lengthen the voting period and close polls at the same time across the country. The policy at all three networks, they said, is not to broadcast results of exit polls or sample votes in state races until the majority of the polls have closed in that state.

But a statement given to the panel by the League of Women Voters and the Committee for the Study of the American Electorate said that the networks violated that policy in 1982. The report said, and the officials confirmed, that in 1982, CBS and ABC joined NBC in making projections on the basis of exit polls, rather than waiting for returns from the selected precincts.

Saying that "augurs ill for 1984," the statement said that "it is possible that even if the networks adhered to their stated, but not practiced, policy of declaring winners only after the polls close in a given state, they would likely be projecting a presidential winner as early as or even earlier than in 1980."

No legislation threatening the practice is pending, but Rep. William M. Thomas (R-Calif.) said, "I am still hoping we get a voluntary response."

But the network executives all repeated variants of Sauter's statement that "if we temper our efforts to report certain information to our audience on the theory that it would be better for them not to know, we start down a very dangerous road."