The results, published today in the journal Nature, showed that overconfidence pays off only when there is uncertainty about opponents' real strengths, and when the benefits of the prize at stake is sufficiently larger than the costs."Evolution of Narcissism: Why We're Overconfident, and Why It Works"
"So let's say you and I are fighting over some resource," Johnson said. "As long as there is some uncertainty about the outcome and the resource is valuable compared with the costs incurred in fighting for it, then overconfidence is the best strategy."
For instance, if people are fighting over an island with oil reserves, the benefit of accessing the oil might be a hundred billion dollars, while the costs of the war might be ten billion.
But if "if the cost of conflict or competition is high, and all for a fairly worthless prize—you're much better off being cautious."
Johnson compares the survival of overconfidence as a primary human trait to our outmoded cravings for high-calorie food—"a consequence of a preference that was useful to us in the past" when calories were scarce "but [that] goes wrong when there's a McDonald's on every corner."