Illusions About Ourselves: How We Overrate What We Know and Control

Start with the simplest version of the pattern. Ola Svenson's 1981 driving study found that 93 percent of a U.S. sample and 69 percent of a Swedish sample put themselves in the top half for driving skill, which cannot be true of a group judged honestly.

Here the focus moves from rating oneself against others to trusting one's own answers. In experiments where people state how confident they are, a stated ninety percent certainty repeatedly exceeds the actual hit rate, and researchers separate overestimation, overplacement and a third form of the bias.

A bias about biases. Emily Pronin and colleagues reported in 2002 that people rated themselves as less susceptible to common biases than the average person while readily spotting the same biases in others, so knowing about bias does not protect us from it.

Confidence in our own knowledge can be tested directly. Leonid Rozenblit and Frank Keil found in 2002 that people who rated their understanding of a zipper or a flush toilet highly gave lower ratings after trying to write out how it works.

Knowing a subject well has its own cost. Colin Camerer, George Loewenstein and Martin Weber introduced the term in 1989, and it names how experts who have forgotten what it is like not to know often struggle to explain things clearly to beginners.

The same assumption turned inward. Thomas Gilovich, Victoria Husted Medvec and Kenneth Savitsky published studies in 1998 showing that people overestimate how easily others can tell they are nervous or lying, because their own inner experience feels so obvious to them.

From what others see to what we can influence. In 1975 Ellen Langer found that people would sell a lottery ticket they had chosen themselves for a higher price than an assigned one, although the odds were identical, and features such as choice and familiarity strengthen the effect.

Self-assessment now turns toward the future. Neil Weinstein's 1980 study first systematically showed that people think they are less likely than others to suffer negative events, and giving accurate statistics about risk often fails to change personal estimates.

Optimism meets schedules. Daniel Kahneman and Amos Tversky proposed the planning fallacy in 1979, and the Sydney Opera House, expected in 1963 and opened in scaled-down form in 1973, is a frequently cited example of underestimated time and cost.

The path ends with the forecast about ourselves. Jordi Quoidbach, Daniel Gilbert and Timothy Wilson reported in 2013 that people of all ages recall considerable change in the past decade yet predict little change in the next, as if the present self were the finished one.