“For example, why is it that you think of yourself as a talented person? (C’mon, give it up. You know you do.) To answer this question, researchers asked some volunteers (definers) to write down their definition of talented and then to estimate their talent using that definition as a guide. Next, some other volunteers (nondefiners) were given the definitions that the first group had written down and were asked to estimate their own talent using those definitions as a guide. Interestingly, the definers rated themselves as more talented than the nondefiners did. Because definers were given the liberty to define the word talented any way they wished, they defined it exactly the way they wished—namely, in terms of some activity at which they just so happened to excel. ("I think talent usually refers to exceptional artistic achievement like, for example, this painting I just finished,” or “Talent means an ability you’re born with, such as being much stronger than other people. Shall I put you down now?”).

Of course, if you’ve ever discussed a football game, a political debate, or the six o’clock newscast with someone from the other side of the aisle, you’ve already discovered that even when people do encounter facts that disconfirm their favored conclusions, they have a knack for ignoring them, forgetting them, or seeing them differently than the rest of us do.

Consider a study in which volunteers were shown some quiz-show questions and asked to estimate the likelihood that they could answer them correctly. Some volunteers were shown only the questions (the question-only group), while others were shown both the questions and the answers (the question-and-answer group). Volunteers in the question-only group thought the questions were quite difficult, while those in the question-and-answer group—who saw both the questions (“What did Philo T. Farnsworth invent?”) and the answers (“The television set”)—believed that they could have answered the questions easily had they never seen the answers at all. Apparently, once volunteers knew the answers, the questions seemed simple (“Of course it was the television—everyone knows that!”), and the volunteers were no longer able to judge how difficult the questions would seem to someone who did not share their knowledge of the answers.

we spend countless hours and countless dollars carefully arranging our lives to ensure that we are surrounded by people who like us, and people who are like us. It isn’t surprising, then, that when we turn to the folks we know for advice and opinions, they tend to confirm our favored conclusions—either because they share them or because they don’t want to hurt our feelings by telling us otherwise.35 Should the people in our lives occasionally fail to tell us what we want to hear, we have some clever ways of helping them. For example, studies reveal that people have a penchant for asking questions that are subtly engineered to manipulate the answers they receive.36 A question such as “Am I the best lover you’ve ever had?” is dangerous because it has only one answer that can make us truly happy, but a question such as “What do you like best about my lovemaking?” is brilliant because it has only one answer that can make us truly miserable (or two if you count “It reminds me of Wilt Chamberlain”). Studies show that people intuitively lean toward asking the questions that are most likely to elicit the answers they want to hear. And when they hear those answers, they tend to believe what they’ve nudged others to say, which is why “Tell me you love me” remains such a popular request.37 In short, we derive support for our preferred conclusions by listening to the words that we put in the mouths of people who have already been preselected for their willingness to say what we want to hear.

Volunteers in one study were asked to evaluate the intelligence of another person, and they required considerable evidence before they were willing to conclude that the person was truly smart. But interestingly, they required much more evidence when the person was an unbearable pain in the ass than when the person was funny, kind, and friendly. When we want to believe that someone is smart, then a single letter of recommendation may suffice; but when we don’t want to believe that person is smart, we may demand a thick manila folder full of transcripts, tests, and testimony.

Precisely the same thing happens when we want or don’t want to believe something about ourselves.

For instance, volunteers in one study were invited to take a medical test that would supposedly tell them whether they did or did not have a dangerous enzyme deficiency that would predispose them to pancreatic disorders. The volunteers placed a drop of their saliva on a strip of ordinary paper that the researchers falsely claimed was a medical test strip. Some volunteers (positive-testers) were told that if the strip turned green in ten to sixty seconds, then they had the enzyme deficiency. Other volunteers (negative-testers) were told that if the strip turned green in ten to sixty seconds, then they didn’t have the enzyme deficiency. Although the strip was an ordinary piece of paper and hence never turned green, the negative-testers waited much longer than the positive-testers before deciding that the test was complete. In other words, the volunteers gave the test strip plenty of time to prove that they were well but much less time to prove that they were ill. Apparently it doesn’t take much to convince us that we are smart and healthy, but it takes a whole lotta facts to convince us of the opposite. We ask whether facts allow us to believe our favored conclusions and whether they compel us to believe our disfavored conclusions. Not surprisingly, disfavored conclusions have a much tougher time meeting this more rigorous standard of proof.

Distorted views of reality are made possible by the fact that experiences are ambiguous—that is, they can be credibly viewed in many ways, some of which are more positive than others. To ensure that our views are credible, our brain accepts what our eye sees. To ensure that our views are positive, our eye looks for what our brain wants ... For positive views to be credible, they must be based on facts that we believe we have come upon honestly. We accomplish this by unconsciously cooking the facts and then consciously consuming them.

If you make a test you will compare to people who scored worse making test and comparing to people who scored worse

For example, volunteers in one study took a test that ostensibly measured their social sensitivity and were then told that they had flubbed the majority of the questions. When these volunteers were then given an opportunity to look over the test results of other people who had performed better or worse than they had, they ignored the tests of the people who had done better and instead spent their time looking over the tests of the people who had done worse. Getting a C– isn’t so bad if one compares oneself exclusively to those who got a D. This tendency to seek information about those who have done more poorly than we have is especially pronounced when the stakes are high.

Friends will not help other friends when the test is described as important intellectual ability

Volunteers in one study took a test and were then given the opportunity to provide hints that would either help or hinder a friend’s performance on the same test. Although volunteers helped their friends when the test was described as a game, they actively hindered their friends when the test was described as an important measure of intellectual ability. Apparently, when our friends do not have the good taste to come in last so that we can enjoy the good taste of coming in first, we give them a friendly push in the appropriate direction. Once we’ve successfully sabotaged their performances and ensured their failure, they become the perfect standard for comparison.

When volunteers in another study were given a glowing evaluation by a supervisor, they were more interested in reading background information that praised the supervisor’s competence and acumen than background information that impeached it. By controlling the sample of information to which they were exposed, these people indirectly controlled the conclusions they would draw.

For example, in one study, some volunteers were shown evidence indicating that extraverts receive higher salaries and more promotions than introverts do (successful-extravert group) and other volunteers were shown evidence indicating the opposite (successful-introvert group). When the volunteers were asked to recall specific behaviors from their pasts that would help determine whether they were extraverted or introverted, volunteers in the successful-extravert group tended to recall the time when they’d brazenly walked up to a complete stranger and introduced themseves, whereas volunteers in the successful-introvert group tended to recall the time when they saw someone they liked but had been too shy to say hello.

If you’ve managed to forgive your spouse for some egregious transgression but still find yourself miffed about the dent in the garage door or the trail of dirty socks on the staircase, then you have experienced this paradox.

For example, volunteers in one study were students who were invited to join an extracurricular club whose initiation ritual required that they receive three electric shocks. Some of the volunteers had a truly dreadful experience because the shocks they received were quite severe (severe-initiation group), and others had a slightly unpleasant experience because the shocks they received were relatively mild (mild-initiation group). Although you might expect people to dislike anything associated with physical pain, the volunteers in the severe-initiation group actually liked the club more. Because these volunteers suffered greatly, the intensity of their suffering triggered their defensive systems, which immediately began working to help them achieve a credible and positive view of their experience"