Social psychology crystallized in the 19th century around a concern with crowd behavior: Why do otherwise reasonable individuals become irrational or even dangerous when placed in a mob of people? By the middle of the 20th century, social psychologists had widened their research to examine how people can be influenced to make incorrect judgments or cross moral boundaries. In the 1950s, Solomon Asch, a pioneer in social psychology, pitted naive test subjects against a group of strangers who made bizarre judgments about the relative lengths of lines. Pressured to conform to the group, subjects often disregarded the obvious visual evidence and adopted the prevailing judgment.

About the time of Milgram's experiment, Princeton University professor John Darley studied why bystanders, when confronted with strangers in distress, sometimes respond by walking away or closing the drapes. Inspired by the case of Kitty Genovese, a New York City murder victim whose cries for help failed to rouse her neighbors to action, Darley showed that test subjects were less likely to aid a stranger if they thought they were just one among several witnesses.

Despite evidence of sheeplike behavior, many researchers still assumed that individuals, on their own, could be counted on to be rational and moral. The sea change came in the 1970s, from insights gleaned through economics research. In a series of articles and books, psychologists Daniel Kahneman, who later won the Nobel Prize in Economics, and Amos Tversky rejected the long-held notion that humans are rational actors in a marketplace. Rather than using all the information available and calculating the best decision, they argued, the human mind relies on "quick and dirty" heuristics, mental shortcuts or rules of thumb, to make decisions.


Social psychologists, including Krueger, jumped in to investigate these rules of thumb. Because the rules aren't always rational, researchers thought they would be exposed in situations where test subjects were led to make mistakes. In effect, the psychologists started looking for errors—and for experiments that would prompt them to occur.

"Like many other graduate students, I thought this stuff was so cool," Krueger says, holding a book containing some of Kahneman and Tversky's work. "The task before us was to set up experiments that would show errors and biases, and those mistakes would tell us what was really going on with human cognition. Of course, what was really going on was always something bad—a departure from some researcher's idea of how the mind should work."

Krueger's interest was stereotyping. In the late 1980s and early 1990s, he published papers showing how people use arbitrary categories to make judgments. On hot August days, for instance, people look forward to the first day in September, as if turning a page on the calendar would suddenly make the weather cooler. Krueger found that people make two errors in this case: They underestimate temperature changes within a month (assuming, for instance, that August will be uniformly hot) and overestimate the changes in temperature that will occur when the month ends.

Since then, revelations of human misperception and bias have popped up in the social psychology studies like toadstools after a rain. We humans have a variety of ways of perceiving ourselves as smarter, more skilled, and more appealing than we are in reality. Most drivers, for example, say they drive more safely than the average person, even though that is a statistical impossibility. People also tend to consider themselves more attractive than others say they are. We tend to underestimate the chance that past events will reoccur, like winning two poker hands in a row (the "hot hand" fallacy). Likewise, we incorrectly assume that because a basketball player has made the last five shots he will make the sixth. We overestimate small risks, like being killed by a terrorist, yet underestimate much larger ones, like being killed in a traffic accident.

The list goes on: the "hindsight bias," the "systematic distortion effect," the "false uniqueness effect," the "just world bias," the "clouded judgment effect," and the "external agency illusion." And just in case you think you're hip to your own biases, researchers have unveiled the "bias blind spot," in which you see biases in others but overlook them in yourself.