Illusory Superiority: The Lake Wobegon Effect in Human Cognition

Above Average, All of Us

In 1977, a survey of one million high school students found that 70% rated themselves above average in leadership ability. Eighty-five percent rated themselves above average in their ability to get along with others — an ability on which 25% rated themselves in the top 1%. Since only 50% of any group can be above the median by definition, these numbers describe a systematic and impossible collective self-assessment. College Board was not surveying unusually self-aggrandizing students. They were documenting a cognitive pattern that social psychology research has since confirmed across cultures, demographics, and ability domains.

The phenomenon has acquired several names: the above-average effect, the better-than-average effect, illusory superiority, and — after the fictional Minnesota town in Garrison Keillor's radio program where "all the children are above average" — the Lake Wobegon effect. Whatever the label, the phenomenon is the same: humans systematically and incorrectly rate themselves as superior to the average person across a wide range of traits and abilities.

The Evidence Base

The above-average effect has been documented across a remarkable range of domains:

The cross-cultural replication is particularly important. Early critics suggested the effect was a product of Western individualistic culture. While the magnitude of the effect is indeed smaller in East Asian samples — where self-effacement norms are stronger — the effect is not eliminated. It appears in Japanese, Chinese, Korean, Indian, and South American samples, though often with different domains (self-serving comparisons in areas valued by the local culture).

Why Illusory Superiority Occurs

Multiple mechanisms contribute to the above-average effect:

• Differential weighting of attributes — People define the trait being compared in ways that favor their own strengths. When asked if they are a good driver, most people implicitly emphasize the dimensions of driving they happen to excel at (safety, smoothness, knowledge of rules) and de-emphasize those they don't (parallel parking, speed).
• Motivated cognition — Positive self-views serve psychological functions: they support self-esteem, maintain motivation, and reduce anxiety. There is evolutionary logic to moderate self-enhancement — individuals who believe they have a reasonable chance of success attempt more challenges than those who don't, and probabilistically do better even if the belief is mildly inflated.
• Social comparison bias — People compare themselves to specific others rather than to the true statistical distribution. These comparison targets are often those immediately around them rather than the full population, and people naturally select favorable comparison targets.
• Dunning-Kruger overlap — Incompetence impairs not only performance but also the metacognitive ability to recognize one's own incompetence. People with limited skill in an area often lack the framework to evaluate how limited their skill actually is.

The Dunning-Kruger Effect: A Related but Distinct Phenomenon

The Dunning-Kruger effect, described by David Dunning and Justin Kruger in their 1999 paper "Unskilled and Unaware of It," is often conflated with illusory superiority but describes a more specific pattern. Their study found that individuals who scored in the bottom quartile on tests of logical reasoning, grammar, and humor significantly overestimated their own performance — not just claiming to be above average, but claiming actual test scores far higher than their actual performance. Meanwhile, high performers tended to slightly underestimate their relative standing, assuming others performed about as well as they did.

Subsequent research has complicated the original Dunning-Kruger findings. A 2020 paper by Gignac and Zajenkowski replicated the phenomenon but found the magnitude to be smaller than originally reported, with some of the original findings attributable to statistical artifacts (regression to the mean). The phenomenon is real — low performers overestimate, high performers are relatively well-calibrated — but the original framing as a sharp discontinuity may have been overstated.

Domain-Specificity and the Limits of Illusory Superiority

Illusory superiority is not uniform across situations. Several moderating factors reduce or reverse the effect:

• Easy vs. difficult tasks — For easy tasks, most people correctly believe they perform above average. For difficult tasks, the above-average bias may reverse: people underestimate their performance because they assume the task is hard for everyone else too. This pattern, called the hard-easy effect, produces systematic miscalibration in both directions.
• Proximal vs. distal comparison — People show stronger above-average bias when comparing to "the average person" than to specific, known individuals. The generic comparison target is easy to inflate against; a specific person you know well is not.
• Objective vs. subjective traits — The effect is stronger for ambiguously defined traits (intelligence, leadership) than for objectively measurable ones (how many push-ups you can do). Objective feedback reduces illusory superiority by anchoring self-assessment in reality.

Consequences in Real-World Contexts

Illusory superiority has measurable consequences across several domains:

Terrance Odean's analysis of 10,000 discount brokerage accounts found that investors who traded most frequently — driven by overconfidence in their market timing ability — earned average annual returns 5.5 percentage points lower than the market, while the least active traders nearly matched market performance. Overconfidence was literally costly.

Calibration as a Cognitive Virtue

The problem with illusory superiority is not self-confidence per se, but miscalibration — believing one's abilities are more accurate than they are. Research on expert forecasters by Philip Tetlock found that well-calibrated individuals — those whose confidence in their predictions accurately matched their actual accuracy rates — significantly outperformed poorly calibrated peers, even when both groups had similar baseline knowledge.

Calibration can be improved. Training in probabilistic thinking, regular exposure to objective feedback, and deliberate practice at estimating one's own performance before receiving external evaluation all reduce overconfidence. Intelligence analysts, weather forecasters, and poker players who regularly receive quick, accurate feedback on their predictions show significantly better calibration than populations insulated from such feedback.

Lake Wobegon is a fictional town. The cognitive tendency it represents is real — and consequential. Recognizing that one is almost certainly subject to it, without knowing exactly where or to what degree, is itself the beginning of better calibration.