Strangers to Ourselves

by Timothy D Wilson

  And finally, I should report on what happened to my psychologist friends who carried their exhaustive list of 7-point scales to every house they visited. After dutifully filling out the scales for a few houses, they found that they were even more confused than before about which houses they liked and why. "We finally threw away the list," they said, "and went with our gut feelings about which house we liked the best." They bought a lovely house in which they have been living happily for the past fifteen years. Perhaps introspection is not always fruitful, and may even mislead people about how they feel. As the poet Theodore Roethke put it, "Self-contemplation is a curse / That makes an old confusion worse.""

  Does this mean that introspection is a useless exercise that is best avoided? that we should advise against all navel-gazing, tell insight therapists to take down their shingles, and recommend that people focus on anything but themselves? It would be odd for a psychologist to tell people never to think about themselves, and this is not my message. The key is to understand that introspection does not open magic doors to the unconscious, but is a process of construction and inference. Once this is understood, the question becomes when this process of construction is likely to be helpful and when it is not.

  OURS IS NOT TO REASON WHY

  Consider what happens when people engage in the Franklinesque type of introspection, whereby they analyze the reasons for their preferences. Sometimes people do this formally, as Franklin suggested, by making lists of the pluses and minuses of the alternatives. At other times they do it less formally, such as when they think, "Why do I feel the way I do about this person I'm dating, anyway?" My colleagues and I have investigated what happens when people introspect in this manner. We typically ask people to spend about ten minutes writing down their reasons for a particular feeling. We tell them that the purpose of this exercise is to organize their thoughts and that no one will read what they write, and then see what effect this introspection has on their subsequent attitudes.

  We have asked people to analyze a wide range of attitudes, including their feelings toward someone they have just met, romantic partners, political candidates, social issues, consumer products, works of art, and college courses. We have been struck by the fact that people have no difficulty in coming up with a list of reasons for their feelings. Almost never has anyone said, "Sorry, I just don't know why I feel the way I do." Instead, people freely and readily write quite detailed reasons for their feelings.

  The accuracy of people's reasons, however, is suspect. People are not always wrong-if they say they love their romantic partner because he is extremely kind, or because he has a great sense of humor, they might be right. People do not have access to all the determinants of their feelings, however, and their reasons are often a function of cultural or personal theories that can be wrong or, at best, incomplete. In the panty-hose study discussed in Chapter 5, for example, people did not recognize that the order in which they examined four pairs of panty hose helped determine which one they liked the best. Instead, people constructed stories to explain their feelings, and these stories were often incorrect. As Immanuel Kant put it, "We can never, even by the strictest examination, get completely behind the secret springs of action. 1112

  If people recognized that their explanations were sometimes inaccurate, there would be no danger in making a list of the reasons why they felt the way they did. "I'll do the best I can," they might say, "but keep in mind that my list is undoubtedly incomplete and that some of the things I put down are probably wrong. Hey, I took psychology in college, Doc." As seen in Chapter 5, however, there is an illusion of authenticity such that the reasons people give feel more accurate than they are.

  Because people have too much faith in their explanations, they come to believe that their feelings match the reasons they list. If they generate several reasons why their dating partner is pretty unexciting ("He has really nice taste in upholstery"), they infer that they are not all that in love-even if they were in love before. In other words, they construct a story about how they feel that is based on reasons that are not entirely trustworthy. The story has the ring of truth to people, but because they have used faulty information (reasons that happened to be on their minds), it often misrepresents how they really feel.

  We have found evidence of just this sequence of events. For example, Dolores Kraft and I asked college students involved in dating relationships to write down, privately and anonymously, why their relationship was going the way it was, and then to rate how happy they were with their relationship. Compared to people in a control condition who did not analyze reasons, these students tended to change their attitudes toward their relationship. Some became happier with it, some less happy.

  Why? First, we assumed that people did not know exactly why they felt the way they did. It is not as if people can say with any accuracy, "Okay, here are nay reasons: her basic integrity and kindness account for 43 percent of my love, her sense of humor 16 percent, her political views 12 percent, that endearing way she brushes the hair out of her eyes 2 percent, and the rest is pheromones." Instead, people brought to mind reasons that conformed to their cultural and personal theories about why people love others and that happened to be on their minds ("I was just looking at the paisley pattern on his couch and thinking about what a great decorator he is"). Because there is a certain arbitrariness to these reasons, they often do not match people's prior feelings perfectly. In fact the reasons people gave bore almost no relationship to how happy they said they were with their relationship a few weeks earlier. But because people do not recognize this fact, they assume that their reasons are an accurate reflection of their feelings, leading to attitude change. In short, people construct a new story about their feelings based on the reasons that happen to come to mind."

  This is what seems to have happened to Proust's Marcel in Remembrance of Things Past. As seen in Chapter 1, Marcel becomes convinced that he no longer loves Albertine, after analyzing and introspecting about his feelings: "As I compared the mediocrity of the pleasures that Albertine afforded me with the richness of the desires which she prevented me from realizing ... [ I ] concluded that I did not wish to see her again, that I no longer loved her."

  I should point out that analyzing reasons does not always lead to attitude change in a negative direction. In our study of dating couples, not all people who listed reasons became more negative toward their relationship. Rather, the direction of the attitude change depended on the nature of the reasons that happened to come to each person's mind. People who found it easiest to think of positive reasons ("He's a great friend and easy to talk to") changed their attitude in a positive direction, whereas those who thought of lukewarm or negative reasons ("He has a fine fashion sense, though it would be nice if he didn't wear that pink shirt quite so often") changed in a negative direction. Marcel found it easiest to think of negative aspects of his relationship with Albertine, and thus concluded that he no longer loved her.

  If Benjamin Franklin picked up a psychology journal and read about these findings, he might respond, "Just as I thought-when people step back and think about the pros and cons, they come up with a betterinformed, more reasoned point of view. After people analyze reasons, their attitude is superior to the quick, possibly rash judgments they would have otherwise made"

  The story people construct on the basis of their reasons analysis, however, can misrepresent how they really feel. Such was the case with Marcel, who discovers, only after learning that Albertine has left him, how wrong he was about his overanalyzed feelings. We have found that the feelings people report after analyzing reasons are often incorrect, in the sense that they lead to decisions that people later regret, do not predict their later behavior very well, and correspond poorly with the opinion of experts.

  For example, in another study we compared people who were asked to list reasons about why their relationship was going the way it was with people who did not list reasons. Whose feelings did the best job of predicting the longevity of the
relationship? It was the latter group, who did not analyze reasons. This is consistent with the notion that when people analyzed reasons, they constructed stories based on faulty data, such as which aspects of the relationship were easiest to put into words, were on their minds, or were consistent with their theories about how they should feel, leading to attitudes that were less well informed than those of people in the control group, who just gave their unanalyzed, gut feelings. As Goethe put it, "He who deliberates lengthily will not always choose the best."

  A study of people's attitudes toward works of art tested Goethe's hunch. Some people analyzed exactly why they liked or disliked five art posters and some did not. Then, all participants chose one of the posters to take home. Two weeks later, we called people up and asked them how happy they were with the poster they had chosen. Benjamin Franklin might predict that the people who analyzed their reasons would make the best choices, by carefully laying out the pros and cons of each option. We found the opposite: the people who did not list reasons, and presumably based their choices on their unanalyzed gut feelings, were happier with their posters than were the people who had listed reasons. Like Mario Vargas Llosa, who found it difficult to tell how he felt about the films when he analyzed each one, the students in the reasons analysis group seemed to lose sight of which poster they really liked the best."

  A few years ago I was interviewed by a reporter about this line of research. After we chatted for a while the reporter said she had one final question: "So, Dr. Wilson, I gather you are saying that people should never think about why they feel the way they do and should simply act on their first impulses?" I was horrified and had images of people following the reporter's conclusions about my research, leading to increases in teen pregnancy, drug relapses, and fistfights.

  It is important to distinguish between informed and uninformed gut feelings. We should gather as much information as possible, to allow our adaptive unconscious to make a stable, informed evaluation rather than an ill-informed one. Most of us would agree that it would not be wise to marry the first person we are attracted to. If we spend a lot of time with someone and get to know him or her very well, and still have a very positive gut feeling, that is a good sign.

  The trick is to gather enough information to develop an informed gut feeling and then not analyze that feeling too much. There is a great deal of information we need in order to know whether someone would make a good partner, much of it processed by our adaptive unconscious. The point is that we should not analyze the information in an overly deliberate, conscious manner, constantly making explicit lists of pluses and minuses. We should let our adaptive unconscious do the job of forming reliable feelings and then trust those feelings, even if we cannot explain them entirely.

  IS IT ALWAYS SO BAD TO THINK ABOUT REASONS?

  Another thing I told the reporter is that there are some exceptions to the danger of analyzing reasons, which follow from our explanation of why it can be harmful. As we have seen, people often change their minds about how they feel because the reasons they think of do not match their prior feelings very well. There is a group of people for whom this is not true, namely people who are quite knowledgeable about the topic they are analyzing. In the study with the art posters, for example, people who knew a lot about art-those who had taken high school and college art courses-tended to list reasons that matched their prior feelings well. Consequently, the act of listing reasons did not lead to any attitude change in this group. It was the unknowledgeable people who were most likely to bring to mind reasons that conflicted with their initial feelings, causing them to revise their stories about how they felt. Contrary to Benjamin Franklin's advice, the knowledgeable people in our studies do not seem to gain anything by analyzing reasons. The art experts did not like the posters they chose more than unknowledgeable people, but neither did they like them more.

  But surely, you might argue, we have not done a fair test of the kind of introspection Franklin recommended. He suggested that people write down pros and cons "during three or four days consideration," whereas in our studies people typically write about reasons only once for ten minutes or so. Might people better decipher their feelings with a longer self-analysis? To find out, Dolores Kraft and I asked the people in our dating couples study to come back to our lab and analyze reasons again, once a week for four weeks. We found that a fair amount of attitude change occurred the first time people analyzed reasons (as discussed earlier), and then people tended to stick to this new attitude when they came back and analyzed reasons again. There did not seem to be any advantage to analyzing reasons more than once; rather, people brought to mind reasons that conflicted with their initial attitude, changed their attitudes to match those reasons, and then stuck to that new attitude.

  It is possible, of course, that people would have benefited from an even longer reasons analysis or from one that was not spread out over so much time. My hunch, though, is that if people are not very knowledgeable about the topic they are analyzing, it is an exercise best avoided-at least in the way we have studied it, whereby people sit down by themselves and think about why they feel the way they do.

  RECOGNIZING GUT FEELINGS

  Suppose you take my advice and let your adaptive unconscious develop feelings about somebody or something, and avoid the kind of introspection in which you try to put into words exactly why you feel the way they do. What if you are still not certain how you feel? Sometimes people have mistaken beliefs about the nature of their feelings, particularly when their feelings conflict with cultural feeling rules ("people love their ponies,I'll my wedding day will be the happiest time of my life"), personal standards ("I am not prejudiced at all toward African Americans"), or conscious theories ("I must love him because he conforms to my idea of Mr. Right"). Is there a kind of introspection by which you can gain access to feelings that are hidden in this manner?

  Introspection should not be viewed as a process whereby people open the door to a hidden room, giving them direct access to something they could not see before. The trick is to allow the feelings to surface and to see them through the haze of one's theories and expectations.

  A recent study by Oliver Schultheiss and Joachim Brunstein suggests one way people might accomplish this. They measured people's implicit motives, using the Thematic Apperception Test technique described in Chapter 4, whereby people make up stories about a set of standard pictures and these stories are coded for how people express motives such as the need for affiliation or power. They then told participants that they would play the role of a therapist who would use directive techniques to counsel a client. Because people were instructed to be directive and keep control of the situation, and to focus on ways of helping the client, those who were high in both the need for power and the need for affiliation were expected to react especially positively.

  The question is, did people know that this was a situation that was well suited or not well suited to their implicit motives? The answer was no when the researchers simply described the counseling situation to participants, and then asked them how they would feel. Consistent with many studies that find that people are not very aware of their implicit motives, people who were high in the need for affiliation and power did not anticipate that the counseling session would make them any happier or feel more engaged than other participants.

  In another condition, however, people first underwent a goal imagery procedure whereby they listened to a detailed, tape recorded description of the counseling session and imagined how they would be likely to feel in that situation. Under these circumstances people high in the need for affiliation and power were more likely to recognize that the situation would be one that they would enjoy, and they reported that they would be much happier and more engaged in that situation than other participants did.''

  Thus, hearing a detailed, image-laden description of the situation was sufficient to trigger feelings generated by people's implicit motives, and people were able to pay attention to these feelings and use th
em to predict how they would feel in the real situation. I would not call this "introspection" as normally defined, because people were not opening doors to hidden rooms in order to see feelings of which they were unaware. Instead, they were able to imagine a future situation well enough that the feelings it would invoke were actually experienced, and were able to avoid the kinds of introspection we have studied (analyzing reasons) that might obscure how they would really feel.

  It remains to be seen how well people can use this technique in everyday life. The suggestion, at least, is that if people took the time to imagine future situations in great detail (e.g., "How would I feel if my housekeeper rushed in with the news that Albertine has left me?"), they might be better able to recognize the feelings generated by their adaptive unconscious, and to see through the smoke screen created by analyzing reasons or by the adoption of cultural feeling rules and conscious theories. They would have better data on which to base their narrative about their feelings and reactions.

  Introspecting about Personal Problems

  Although some of the studies on introspection considered so far dealt with topics that were very important to people, such as why a romantic relationship is going the way it is, they are generally not topics that are causing people distress (most of the participants in our studies were reasonably happy with their relationships). Perhaps people are more adept at introspecting about things that have gone wrong in their lives. There are many ways to introspect about one's source of distress, however, some of which are more helpful than others.

  RUMINATING WHEN DISTRESSED

  One way is to ruminate about a problem, which Susan Nolen-Hoeksema defines as thinking about one's feelings and their causes repetitively without taking action to improve one's situation. In numerous studies, she has found that rumination leads to a negative, self-defeating pattern of thought that makes matters worse, especially when people are depressed or in bad moods to start with. Ruminators are worse at solving problems related to their distress, focus more on negative aspects of their past, explain their behavior in more self-defeating ways, and predict a more negative future for themselves.