Summary: A new study demonstrates that the celebrated Dunning-Kruger effect is the result of statistical artifact rather than human psychology.
The study applies advanced statistical modeling to massive replication datasets, correcting for test volatility and regression to the mean. The team revealed that when statistical distortions created by sorting participants purely by test scores are eliminated, overconfidence emerges as a universal trait, with the highest-performing individuals exhibiting the greatest degree of overconfidence, effectively inverting the classic Dunning-Kruger effect.
Key Facts
- Statistical Artifact Identification: The classic Dunning-Kruger pattern—where low performers overpredict their scores by large margins, is shown to be a mathematical artifact caused by noise, luck, and uncorrected regression to the mean in test scoring.
- Inversion of Overconfidence Trends: Once statistical models account for random variance in test performance and self-assessments across massive datasets, higher-ability individuals consistently demonstrate the highest levels of overconfidence.
- Performance Volatility Distortion: Sorting participants strictly by test scores distorts the data because low test scores often reflect bad luck or ambiguous questions, artificially inflating estimated overconfidence, while high performers experience the inverse mathematical bias.
- Universality of Overconfidence: The research shows that overconfidence is a widespread human tendency across all skill tiers, but its magnitude scales positively with actual competence.
- Strategic Ability Signaling: The authors propose that high performers have the strongest incentives to display overconfidence as a behavioral strategy to signal unobservable competence to others.
Source: University of Bath
Researchers at the University of Bath and The London School of Economics and Political Science (LSE) suggest that the celebrated Dunning-Kruger effect, often taken to mean the least competent people are the most overconfident, may actually be the wrong way round.
For decades, the Dunning-Kruger effect has been understood to mean that people with low ability, those who know the least, are the most overconfident, while high performers are more realistic. Hundreds of studies have replicated the classic finding, illustrating that people who do well on multiple-choice tests overpredict their performance by less than those who do badly.
Backed by this mountain of evidence, the Dunning-Kruger effect has become the internet’s ultimate shorthand for calling out clueless confidence—explaining everything from bad bosses to Twitter trolls with a simple ‘truth’: low performers are ‘unskilled and unaware of it.’
However, this new study, published in Psychological Review, reveals how the Dunning-Kruger effect is an artefact of flawed statistical analysis rather than human nature.
When this flawed statistical analysis is corrected, the researchers found that high-ability people tend to be the most overconfident – reversing the Dunning-Kruger claim, widely understood as “stupid people don’t know they’re stupid”.
Professor Chris Dawson, from the University of Bath’s School of Management, said: “Our research demonstrates that the original methodology fell victim to a statistical illusion. The popular image of the incompetent person brimming with confidence tells us less about human psychology than it does about the hidden traps in statistical analysis.
“When we clear away the statistical problems, the apparent pattern disappears. Instead, we see that overconfidence is a universal human trait, but it is the most capable among us who exhibit it the most.”
Professor David de Meza from LSE’s Department of Management, said: “Test performance is naturally volatile- impacted by factors such as luck or bad question phrasing-and sorting participants purely by their test scores mathematically distorts the data to manufacture a false psychological phenomenon.
“When a test score is exceptionally low, it’s often because the person just got unlucky with the questions—meaning their estimated performance naturally ends up looking way too high. The exact opposite happens to top performers, who benefit from a lucky break.”
To address it, the researchers applied improved statistical techniques to data from thousands of participants across massive replication datasets. This approach manages to handle randomness in both people’s test scores and their predictions about those scores. Once a statistical model respects this reality, the classic story flips. It now turns out that the more able tend to be the most overconfident, the opposite of Dunning-Kruger.
It is plausible that poor performers may sometimes lack the cognitive ability to recognise their own incompetence. There are, though, other reasons why overconfidence arises. The researchers go on to show that it can result from attempts to signal otherwise hidden ability, a mechanism that high-ability individuals have the most incentive to engage in.
Key Questions Answered:
A: The original methodology sorted participants strictly by their test scores without accounting for test volatility or random luck. Low scores often resulted from bad luck, making participants’ self-assessments look artificially overconfident due to mathematical regression to the mean rather than psychological ignorance.
A: When statistical models control for random noise in test scores and predictions, the classic pattern flips: high-ability individuals actually display the highest levels of overconfidence, though overconfidence remains present across all performance levels.
A: Beyond cognitive factors, researchers suggest overconfidence can function as a social signaling mechanism. High-ability individuals have the greatest incentive to project confidence to communicate hidden competence and talent to others.
Editorial Notes:
- This article was edited by a Neuroscience News editor.
- Journal paper reviewed in full.
- Additional context added by our staff.
About this psychology research news
Author: Lynn Li
Source: University of Bath
Contact: Lynn Li – University of Bath
Image: The image is credited to Neuroscience News
Original Research: Open access.
“Talking the Talk, Not Walking the Walk: The Coevolution of Overconfidence and Loss Aversion” byChris Dawson, David de Meza. Psychological Review
DOI:10.1037/rev0000644
Abstract
Talking the Talk, Not Walking the Walk: The Coevolution of Overconfidence and Loss Aversion
A puzzle for evolutionary theory is the existence of two seemingly offsetting behavioral “biases,” overconfidence and loss aversion. Overconfidence is a call to action, while loss aversion curbs initiative.
The most prominent evolutionary explanation of overconfidence, proposed by Trivers (1976), is that self-deceit arises to better deceive others. Missing from this account is why sincere messages are believed, especially given the widespread prevalence of self-deception. Moreover, if overconfidence is adaptive, why is it at least partially canceled by loss aversion?
We propose a signaling theory according to which the role of self-deception is to better inform others. Since the decision error associated with high self-belief is less burdensome for the more able, and the benefit of being perceived as able increases with ability, hardwired overconfidence is a credible signal of true ability. Evidence supports this interpretation.
A further implication of signaling is that loss aversion is part of the equilibrium. It partially ameliorates the decision costs of overconfidence, but as it is usually hidden, it does not eliminate its signaling role. “Biases” are thus symbiotic—the payoff to agents from an integrated set of biases is higher than would be the case in their absence.
From this perspective, Kahneman’s advice that individuals eliminate both overconfidence and loss aversion is poorly founded.
