(FORTUNE Magazine) – The more I look at the amazing rise of the U.S. stock market, the more I'm convinced we are looking at a mammoth psychological problem. I don't mean mammoth as in "huge" (though maybe that too), but as in "elephant." Let me explain.

If you follow trends in psychology, you know that Freud is out and Darwin is in. The basic idea of "evolutionary psych" is that our brains are exquisitely designed to help us cope with our environment--but unfortunately, the environment they're designed for is the one we have lived in for the past two million years, not the alleged civilization we created just a couple of centuries ago. We are, all of us, hunter-gatherers lost in the big city. And therein, say the theorists, lie the roots of many of our bad habits. Our craving for sweets and fatty food evolved in a world without ice cream and french fries; our interest in gossip evolved in a world without tabloids; our taste for music evolved in a world without Celine Dion. And we have investment instincts designed for hunting mammoths, not capital gains.

Imagine the situation back in what ev-psych types call the Ancestral Adaptive Environment. Suppose that two tribes--the Clan of the Cave Bear and its neighbor, the Clan of the Cave Bull--live in close proximity but follow different hunting strategies. The Cave Bears tend to hunt rabbits--a safe strategy, since you can be pretty sure of finding a rabbit every day, but one with a limited upside, since a rabbit is only a rabbit. The Cave Bulls, on the other hand, go after mammoths--risky, since you never know when or if you'll find one, but potentially very rewarding, since one felled mammoth provides a yield of, well, elephantine proportions.

Now suppose that for a year or two the Cave Bulls have been doing very well, making a killing practically every week. After a while, the natural instinct of the Cave Bears is to feel jealous, and to try to share in the good fortune by starting to act like Cave Bulls themselves. In the ancestral environment, that instinct was entirely appropriate: The kinds of events that would produce a good run of mammoths--favorable weather producing a lush crop of grass, migration patterns bringing large numbers of beasts into the district--tended to be persistent, so it was a sound idea to emulate whatever strategy had worked in the recent past.

But transplant our tribes into the world of modern finance, and those instincts aren't appropriate at all. Efficient-markets theory tells us that all the available information about a company is supposed to be already built into its current stock price, so that any future movement is inherently unpredictable. Rational investors, by this logic, should treat bygones as bygones: The fact that your neighbor made a lot of money in stocks last year while you stayed in cash is no reason to get into stocks now. But suppose that, for whatever reason, the market goes up month after month; your MBA-honed intellect may say, "Gosh, those P/Es look pretty unreasonable," but your prehistoric programming is shrieking, "Me want mammoth meat!"

Those instincts can be mighty hard to deny--and self-reinforcing, at least for a while. After all, an increase in the number of people acting like Cave Bulls tended to mean fewer mammoths per hunter, but an increase in the number of modern bulls tends to produce even bigger capital gains, as long as the run lasts. Any broker can tell you that in recent months the market has been rising, despite mediocre earnings news, because of fresh purchases by people desperate to get in on the action. But sooner or later the supply of such people will run out; then what?

Okay, okay, I know that sophisticated investors are supposed to take the long view and arbitrage away these boom-bust cycles. And maybe, just maybe, the market is where it is because wise and far-seeing folks have understood that the New Economy can produce growing profits forever, and that the rise of mutual funds has eliminated the need for old-fashioned risk premiums. But my sense is that people who take a long view have been driven to the edge of extinction by the sheer scale of recent gains, and that explanations we now hear of why current prices make sense are rationalizations rather than serious theories.

The whole situation gives me the chills. It could be that I'm just a Neanderthal too thick-skulled to understand the new era. But if you ask me, there's an Ice Age just over the horizon.

PAUL KRUGMAN is a professor of economics at the Massachusetts Institute of Technology.