Daniel T. Gilbert - US grants

Five years of support are requested for an ADAMHA Research Scientist Development Award (Level 11). The proposed research is part of an ongoing research program that investigates the effects of cognitive busyness (a state in which an individual's processing resources are depleted by concurrent demanding tasks) on social inference processes. My past research has argued for a multiple-stage model of trait inference in which successive stages require increasing amounts of conscious attention. As such, cognitive busyness renders perceivers unable to complete late stages in the information-processing sequence, and thus causes them to draw overly dispositional inferences about others. The present proposal describes four major areas of new research concentration. First, the proposal describes some new data which suggest that the original model is overly constrained, and offers a series of experiments that will allow me to expand the model in a variety of important ways. Second, the proposal describes a series of experiments that attempt to apply this model of trait inference in other-perception to a variety of problems in self-perception. Third, the proposal describes a series of experiments that seek to construe social comparison as the interaction of self-perception and other-perception processes. Finally, the proposal describes plans for a book that embeds this and other work in the overarching framework of Spinoza's theory of propositional representation.

DESCRIPTION (provided by investigator): The frequent experience of positive affect is a cornerstone of psychological and physical well-being. To maximize such experiences, people must be able to accurately forecast the affective consequences of future events. Our recent work on affective forecasting has investigated how such predictions are made, uncovered mechanisms that promote inaccuracy, and demonstrated some of the implications of these mechanisms for decision making. The present proposal argues that (a) unexpected and unusual events are particularly capable of evoking affective reactions, (b) people are driven to understand and explain such events so that they can control their reoccurrence, and (c) one consequence of understanding and explaining events is that these events lose their power to evoke affect. This simple conceptualization suggests several avenues for future research, and provides a general unifying theme for our past and proposed work. The current proposal describes 20 studies that explore the ways in which understanding and explaining events modulates affective experience, as well as the ordinary person's ability to predict these effects. Studies on "The Mechanics of Affective Forecasting" represent the next step in our effort to understand exactly how people make forecasts about the impact of future events. Studies on "Tactics of the Psychological Immune System" represent our continuing interest in understanding how people deal with negative events, with special emphasis on automatic processes. Studies on "Paradoxes of Immune Neglect" describe some of the potentially important ways in which the ordinary person's misconceptions about the causes of his or her own hedonic states can impact decision making. Finally, studies on "The Pleasures of Uncertainty" and "Reinvigorating Past Pleasures" deals with people's forecasts and experiences of positive events.

Most decisions are based on predictions about how much hedonic value -- that is, how much pleasure, satisfaction, utility, or reward -- different alternatives will bring. People make important decisions about which home to buy, which person to marry, which medical treatment to accept, and so on by estimating the hedonic value that each of these experiences will afford. But in the last decade, research by psychologists and behavioral economists has shown that these estimates are often wrong. That is, people often mispredict the hedonic value of future experiences and thus make suboptimal decisions that they later regret. Why does this happen?

Both the estimated and actual hedonic value of an experience depends on the other experiences to which it is compared. For example, chemotherapy may seem worse when compared with simple surgery than with a debilitating course of radiation. The comparisons people make are of two types. When a person compares an experience with other experiences he or she has had, will have, or could have had (e.g., At least this chemotherapy is not making me feel as bad as the radiation would have), the value of the experience is influenced by differences. But when a person compares an experience with the experience he or she was having in the previous moment(e.g., I am more nauseous than I was a few minutes ago), the value of the experience is influenced by changes.

The researchers suggest that as a general rule, when people are estimating the hedonic value of future experiences they tend to compute differences, but when they are actually having hedonic experiences they tend to compute changes. Because people make different comparisons at these two times, they often mispredict the hedonic value of future experiences. This proposal describes 13 studies that seek to investigate the causes and consequences of this phenomenon.

A decade of research has shown that people often can't predict whether something will make them happy, how happy it will make them, and how long that happiness will last. As a result, people make poor decisions in consequential domains ranging from the social and economic to the legal and medical. Researchers have worked hard over the past decade to find a way to improve the accuracy of peoples' predictions, and for the most part they've done this by trying to improve the accuracy with which people imagine or "mentally simulate" future events. The results of these attempts have been disappointing. Either they've failed completely, or they've been successful in limited circumstances.

The proposed research is based on the argument that the best way to improve the accuracy of affective forecasting is not to improve mental simulation, but to avoid it entirely. Instead, the investigators argue that to predict how happy you will be in the future you should follow the 17th century essayist François de La Rochefoucauld's advice: "Before we set our hearts too much upon anything, let us first examine how happy those are who already possess it." The investigators have found that this strategy for predicting happiness based solely on the report of another person who has already experienced that event, referred to as "surrogation," can improve predictions dramatically. Indeed, in many circumstances people are better off relying on the experience of a single randomly-selected stranger than on their own self-knowledge and imaginations! The current research involves a new series of studies designed to explore the conditions under which surrogation will and will not improve prediction accuracy as well as the conditions under which people are and are not likely to use it.

People make poor decisions in consequential domains ranging from the social and economic to the legal and medical, in part as a consequence of being unable to accurately predict how happy different actions will make them. The broader impact of this research is to provide concrete ways to help people make better predictions and, as a consequence, better choices.

The ability to occupy ourselves solely with our thoughts, paying no attention to the external world, is one of the signatures of what it is to be human. Although there is substantial research on many forms of reflection, such as problem solving and planning, there is relatively little research on deliberate attempts to improve one's subjective well being with one's own thoughts in the absence of any external stimulation. On the one hand this would seem easy to do so, given the richness of human memory (e.g., the ability to recall pleasant events from one's past) and the ability to construct enjoyable daydreams and fantasies. On the other hand, people do not often choose to entertain themselves solely with their minds. When at home with nothing to do, for example, most people choose to read, watch television, listen to music, or pursue their favorite hobby instead of engaging in reflection. The proposed research will investigate why this is, the conditions under which people can put their disengaged minds to good use, and the benefits of deliberative reflection. The overall goal of the research is to understand better the nature of human attention and mental control, and by so doing, discover new ways in which people can reduce stress and improve their well-being.

In preliminary research, Timothy Wilson (University of Virginia), Daniel Gilbert (Harvard University) and colleagues found that people typically do not enjoy spending time by themselves with nothing to do but think, that they enjoyed doing mundane external activities much more, and that many preferred to administer electric shocks to themselves instead of being left alone with their thoughts. Most people seem to prefer to be doing something rather than nothing, even if that something is negative. The proposed research will investigate three possible reasons why this is. First, if people attempt to think exclusively about positive thoughts for too long, negative thoughts begin to intrude. Second, in the absence of any external stimulation people continually scan the external environment. If there is nothing in the external world to engage the scanner it keeps searching for input, using up resources that could be devoted to internal thought. Third, people may not know how to engage in reverie effectively but can be taught to do so. The 12 proposed studies will involve laboratory experiments, field experiments, and experience sampling methodologies, and include college students and community members as participants.