Summary: Researchers offer an evolutionary explanation for why human overconfidence endures despite producing costly mistakes. They argue that overconfidence can act like a costly biological signal—similar to a peacock’s tail—whose expense is precisely what makes it a credible indicator of real ability.
Because the harms of inflated self-belief fall more heavily on low-ability individuals than on high-ability ones, overconfidence can reliably convey otherwise hidden competence. To prevent reckless consequences, the authors show that overconfidence is likely co-adapted with loss aversion, an internal brake that lets people display confidence publicly while behaving more cautiously in private.
Key Facts
- Handicap principle applied to cognition: Overconfidence functions like a biological handicap—costly to maintain—so only those with sufficient ability can sustain it without dire consequences, making the signal informative.
- Uneven cost distribution: High-ability individuals tend to suffer fewer or less severe costs from overconfidence, so inflated self-belief becomes a cheaper and more sustainable signal for the capable.
- Extension of Trivers’ theory: Builds on Robert Trivers’ idea that self-deception evolved to help deceive others by removing behavioral tells, explaining why confident assertions are sometimes accepted rather than dismissed as mere bluster.
- Loss aversion as an internal safeguard: Rather than a standalone cognitive error, loss aversion may act as a concealed handbrake that reduces the most dangerous consequences of overconfidence without eliminating its outward signaling value.
- Evolutionary explanation for sex differences: The model accounts for higher average overconfidence in men as a way to advertise hard-to-observe traits—status, commitment, and resource acquisition—that matter in mating competition.
Source: University of Bath
Researchers from the University of Bath and the London School of Economics and Political Science (LSE) address a long-standing puzzle: why does overconfidence persist in humans even though it often causes costly decisions? Their answer reframes the costs as part of the explanation rather than a contradiction.
Published in the Psychological Review, the study proposes that overconfidence works like an evolutionary signal. In nature, a peacock’s ornate tail is costly to grow and maintain, but its expense makes it reliable: only the fittest males can afford it. Similarly, costly self-belief can serve as an honest indicator of underlying competence.
Professor Chris Dawson, a behavioural economist at the University of Bath’s School of Management, notes: “Overconfidence is common and often costly. It helps explain failed ventures, excessive risk-taking and poor judgments. The puzzle is why evolution has not eliminated it. Our work suggests the costs contribute to its signaling value.”
The research highlights that the negative consequences of overconfidence are not evenly spread. When confident beliefs lead to mistakes, those mistakes typically damage low-ability individuals more severely. Conversely, capable people often avoid the worst outcomes or can recover more effectively, so their confident displays are sustainable.
Because confidence can open opportunities—from leadership roles to promotions and social influence—the benefits also scale with ability. “Confidence opens doors,” Dawson explains. “Once inside, real ability determines success. That alignment is what preserves the signal.”
The study extends Robert Trivers’ influential account that self-deception evolved to conceal the cues of intentional bluffing, making false claims more persuasive. The new contribution addresses why observers do not simply treat confident claims as empty rhetoric: bold self-belief carries material penalties that make it costly for the incapable to fake consistently.
Professor David de Meza of LSE explains: “Observers discount confident claims to some extent, but overconfidence has real costs. Unless someone is actually overconfident, they risk being underestimated. Only those who can afford the price of excessive self-belief will stake those claims—so the signal retains informational value.”
The model also accounts for the pairing of overconfidence with loss aversion. Instead of seeing loss aversion purely as a bias, the authors argue it often functions as a hidden safety mechanism: it allows individuals to project confidence publicly while curbing the most dangerous actions privately. This concealment preserves the signal’s credibility while protecting against catastrophic risk.
“People can talk boldly but act cautiously,” says de Meza. “That combination preserves status signaling without inviting fatal errors.”
The authors warn that well-intentioned efforts to eradicate these biases wholesale may be counterproductive. Eliminating overconfidence could reduce people’s ability to persuade or seize opportunities, while removing loss aversion could eliminate the brake that prevents confidence from turning into reckless behavior.
Key Questions Answered:
A: Observers partially discount confident statements, but because inflated self-belief imposes real, material costs that disproportionately hurt low-ability people, bold claims still carry informative weight. The cost asymmetry keeps confident signaling meaningful.
A: Overconfidence projects status and secures opportunities, acting as the outward signal, while loss aversion serves as an internal restraint. Together they allow individuals to claim advantages publicly while avoiding catastrophic risks privately, preserving both persuasion and survival.
A: Policies that aim to eliminate overconfidence and loss aversion entirely could backfire: without some overconfidence, people may fail to secure opportunities or persuade others; without loss aversion, people might expose themselves to excessive danger. The traits may be symbiotic and functional when balanced.
Editorial Notes:
- This article was edited by a Neuroscience News editor.
- The journal paper was reviewed in full.
- Additional context was added by staff.
About this evolutionary neuroscience and psychology research news
Author: Lynn Li
Source: University of Bath
Contact: Lynn Li – University of Bath
Image: Image credit to Neuroscience News
Original Research: Open access. “Talking the Talk, Not Walking the Walk: The Coevolution of Overconfidence and Loss Aversion” by Chris Dawson and David de Meza. Published in Psychological Review. DOI: 10.1037/rev0000644
Abstract
Talking the Talk, Not Walking the Walk: The Coevolution of Overconfidence and Loss Aversion
A long-standing puzzle in evolutionary theory is why two apparently opposing behavioral tendencies—overconfidence and loss aversion—coexist. Overconfidence pushes toward action and assertion, while loss aversion restrains initiative. Traditional accounts, notably Trivers’ proposal that self-deception evolved to help deceive others, explain why people who genuinely believe their claims are more persuasive. Yet this leaves open why confident claims are often taken seriously rather than dismissed as cheap talk, and why overconfidence is partially offset by loss aversion.
The authors propose a signaling framework: self-deception and inflated self-belief serve to communicate ability. Because the decision errors produced by high self-belief are less costly for more able individuals, and because the benefits of being perceived as able increase with actual ability, stable overconfidence can act as a credible signal. Loss aversion emerges as part of the equilibrium: it reduces the most damaging costs of overconfidence but usually remains concealed, so it does not remove the external signal entirely. From this perspective, the two biases are complementary—together they improve payoff compared with their absence.
This interpretation challenges advice that individuals should broadly eliminate both overconfidence and loss aversion, suggesting instead that these traits may have evolved as a functional paired system that balances persuasion and prudence.