Researchers at the University of Bath and The London School of Economics and Political Science (LSE) believe they have solved a long-standing evolutionary mystery: why humans remain persistently overconfident despite the costly mistakes it can cause. Published in Psychological Review, the study argues that overconfidence—believing you are more capable than you really are—has endured through evolution not in spite of its costs, but because of them. The researchers compare overconfidence to a peacock’s tail, an elaborate but burdensome feature that survives because only the healthiest birds can afford to carry it. Likewise, they suggest that human self-belief acts as a costly signal of underlying ability.
“Overconfidence is strikingly common and often costly,” said Professor Chris Dawson, a behavioural economist at the University of Bath’s School of Management. It contributes to failed businesses, excessive risk-taking and poor decisions, yet evolution has not eliminated it. Instead, the researchers argue that the costs of overconfidence are unevenly distributed. People with greater ability are less likely to suffer severe consequences from inflated self-belief, allowing them to project confidence more successfully than those with lower ability. As a result, overconfidence remains a credible indicator of competence because only the most capable individuals can consistently sustain it.
The researchers also argue that confidence provides important social and professional advantages. “Confidence opens doors,” said Professor Dawson. Confident individuals are more likely to gain influence, attain leadership positions, secure promotions and form valuable relationships. However, confidence alone is not enough to guarantee success. Once opportunities arise, genuine ability determines who performs well. If overconfidence carried no cost, anyone could exaggerate their abilities without consequence, making self-belief meaningless as a signal. Its credibility depends on the fact that excessive confidence is more costly for less capable individuals.
The study builds on a theory proposed by evolutionary biologist Robert Trivers, who suggested that self-deception evolved because genuinely believing our own claims makes us more convincing to others by eliminating subtle signs of dishonesty, such as nervousness or vocal strain. However, the new research addresses a question left unanswered by Trivers: if overconfidence is so common, why do people not simply dismiss confident claims? Professor David de Meza of LSE explains that although people naturally discount others’ self-belief, those who are not overconfident risk being underestimated. Because inflated confidence remains costly, it continues to provide meaningful information about underlying ability.
The findings may also help explain why men generally display higher levels of overconfidence than women. According to Professor Dawson, historical differences in mating strategies meant that women often assessed qualities such as status, commitment and resourcefulness—traits that are difficult to observe directly. Overconfidence evolved as a credible way for men to signal these hidden qualities. The researchers stress that this reflects evolutionary pressures rather than conscious behaviour in modern society, offering a possible explanation for long-observed differences in confidence between the sexes.
The study further argues that loss aversion—the tendency to fear losses more than we value equivalent gains—works alongside overconfidence rather than against it. While people often project confidence, they frequently behave more cautiously in practice, preventing excessive risk-taking without undermining the confident image they present. “People talk confidently but act cautiously,” said Professor de Meza. The researchers conclude that overconfidence and loss aversion form a complementary pair, allowing people to seize opportunities while avoiding catastrophic mistakes. They caution that attempts to eliminate these biases could produce unintended consequences by reducing persuasion, ambition and opportunity-seeking, while encouraging greater risk-taking.
More information: Chris Dawson et al, Talking the Talk, Not Walking the Walk: The Coevolution of Overconfidence and Loss Aversion, Psychological Review. DOI: 10.1037/rev0000644
Journal information: Psychological Review Provided by University of Bath