Prompted by a plunge in the Consumer Confidence Index, which hit a near-five-year low in February, The New York Times dubbed confidence the "X factor that can save the day or push the economy over the brink into recession." Fed Chairman Alan Greenspan warned Congress that "changes in consumer confidence will require close scrutiny in the period ahead."

But in the age of the 24-hour news cycle, measures of confidence bring up a chicken-or-egg problem. Are consumers responding to their own lived experiences--say, a neighbor who lost a job, or a factory in town that shut down--or to the constant stream of hyped-up financial news on cable and the Internet?

By all rights, consumer confidence should not have been faltering--as it was--at the start of this year. Unemployment was at the low rate of 4.2 percent, and the latest payroll figures show a net gain of more than 125,000 new jobs in January. The Commerce Department also reports that American incomes and consumer spending rose in January--at the same time that consumer surveys showed confidence falling off a cliff.

There are two sources for information on consumer attitudes: The Consumer Confidence Index is the one most often quoted in the press. It is produced not by a government agency but by the Conference Board, a business advisory group. The University of Michigan's Index of Consumer Expectations tracks similar data, and is included in the Index of Leading Economic Indicators, also put out by the Conference Board.

Consumer confidence is considered a "leading indicator" because consumers may have access to information about an economic trend before it is picked up by more official statistics. If consumers report seeing Help Wanted signs in windows, the economic forecast is rosy; if they start mentioning layoffs, bad times may be ahead--or so the theory goes.

But the surveys don't ask about anything as specific as Help Wanted signs. The Conference Board's questionnaire, which is mailed out to a panel of 5,000 households, poses terse questions like: "Six months from now, do you think there will be [more/same/fewer] jobs available in your area?" The chattier Michigan survey, which is conducted by telephone using a randomly selected group, asks: "Looking ahead, which would you say is more likely--that in the country as a whole we'll have continuous good times during the next five years or so or that we'll have periods of widespread unemployment or depression, or what?"

With questions as open-ended as these, there is no way to distinguish concerns born of personal experience from those related to media coverage. Lynn Franco, director of the Consumer Research Center at the Conference Board, admits that "there is more of an information overload now than in the past. Media hype, especially talk of a pending recession, can affect consumers' outlook." Richard Curtin the director of the University of Michigan's Surveys of Consumers agrees. While he doubts that financial reporting prompted the latest decline in consumer confidence, he explains that "once the decline got started, consumers may have looked to the media to see if their perceptions were correct. The large stories about layoffs and rising unemployment may have had a significant impact on pushing confidence down even further."

Could it be that, when polled, consumers are responding not to their own economic experience but to dire reports that they've heard about how other consumers are faring? Once anxieties start to spread, confidence surveys may be little more than a reading taken inside the media echo chamber.