Refuting the old saw that your first guess is always best, 33 studies over 70 years suggest sticking with your first instinct is not always a smart tack. But because getting an answer wrong after going against our first instinct is so frustrating, we tend to believe that changing answers is generally a foolish practice that will result in more wrong answers, according to recent studies by a research team at the University of Illinois at Urbana-Champaign, Northern Arizona University and Stanford University.

In an article in May's Journal of Personality and Social Psychology (Vol. 88, No. 5), the team presents its finding that people buy into the first-instinct myth because it feels worse to change a correct answer to an incorrect one than to stick with an original incorrect answer. And that feeling makes changing right answers to wrong more memorable than a wrong-to-right change and therefore seemingly more probable.

"Our first thought was that this is just an old wives' tale that got propagated," says University of Illinois psychologist and study co-author Justin Kruger, PhD. "But that doesn't explain just how ubiquitous it is. With our research, it became clear that there is a fundamental asymmetry to how people react to getting a problem wrong when they've changed their answer, as opposed to failing to switch from a wrong answer to the right answer."

Take the example of switching into a grocery store line that appears to move faster, says Kruger: Most people have the intuition that as soon as they switch, their new line slows down and their old one speeds up.

"Are the gods punishing us for our impulsiveness?" Kruger asks. "Probably not."

A better explanation might be that moving over to an even slower line is more frustrating and memorable than just staying put in a dud line, says Kruger, who theorized that it's much the same for test-taking.

To establish the first-instinct fallacy, the researchers examined the introductory psychology midterm exams of 1,561 University of Illinois students for eraser marks. They counted the number of times students changed answers and found that 51 percent of the changes were from wrong to right, 25 percent were from right to wrong and 23 percent were from wrong to wrong. Changes from wrong to right outnumbered changes from right to wrong 2-to-1, Kruger points out.

When the researchers surveyed 51 of those students for their intuitions on answer-changing, 75 percent expected changes from right to wrong to outnumber changes from wrong to right--a sentiment proven false by the eraser marks on the students' tests, Kruger says.

In further tests, students indicated that switching a correct answer to an incorrect answer is more regrettable, frustrating and memorable than failing to switch from an incorrect answer.

"Getting the answer wrong on your first instinct is just nowhere near as bad as seeing that you had the answer right and changed it to wrong," he says. "That's just incredibly frustrating."

To test the power of such potential regret and frustration to influence people's test-taking behavior, the researchers mimicked the game show "Who Wants to Be a Millionaire?"

"The game was perfect for this research because, with the life lines, Regis manages to create precisely the dilemma that we're arguing test-takers face all the time: Do I take the life-line's advice or should I stick with my first instinct?" Kruger says. "If you watch the show, you know that it creates a lot of tension and sets up exactly the hedonic asymmetry we wanted to test."

Through a team version of the game, participants again showed that they were more frustrated with wrong answers when a teammate switched from a first instinct than when they stuck. They also remembered their teammate having better luck when they stuck with their first instincts than when they changed--even though the two conditions produced even results. And finally, students were more critical of their teammate's test-taking strategy when they changed their answers than when they stuck with a first instinct, demonstrating that the frustration with and memory of correct answers being changed to wrong answers causes people to think answer-changing is bad, Kruger says.

"It's not that first instincts are generally bad or wrong," he says, "just that people are too conservative in changing answers."

The fallacy could apply to many situations, and Kruger hopes to test some of them, such as choosing between cars, career paths and potential dates or marriage partners.

"In these situations, people may also be willing to place greater trust in their first instinct than they should," he adds. "The results of that sort of trust could prove quite illuminating too."