How People Make Decisions That Involve Risk A Dual-Processes Approach

Many health and safety problems, including war and terrorism, are by-products of how people reason about risk. I describe a new approach to reasoning about risk that implements a modern dual-process model of memory called fuzzytrace theory. This approach posits encoding of both verbatim and gist representations, with reliance on the latter whenever possible; dependence of reasoning on retrieval cues that access stored values and principles; and vulnerability of reasoning to processing interference from overlapping classes of events, which causes denominator neglect in risk or probability judgments. These simple principles explain classic and new findings, for example, the finding that people overestimate small risks but ignore very small risks. Fuzzy-trace theory differs from other dual-process approaches to reasoning in that it places intuition at the apex of development, considering fuzzy intuitive processing more advanced than precise computational processing (e.g., trading off risks and rewards). The theory supplies a conception of rationality that distinguishes degrees of severity of errors in reasoning. It also includes a mechanism for achieving consistency in reasoning, a hallmark of rationality, by explaining how a person can treat superficially different reasoning problems in the same way if the problems share an underlying gist. KEYWORDS—risk perception; risky decision making; fuzzy-trace theory; intuition; dual processes in reasoning As I write these words, coalition troops are at war in Iraq. The decision to go to war with Iraq, like many decisions, was based on a perception of risk, that is, the perceived threat posed by Iraq under Saddam Hussein. Key government officials also believed that inaction with respect to Iraq posed a greater risk than taking action. Since September 11, 2001, foreign threats and risks of terrorism have loomed large in people’s perception, dwarfing perennial killers such as highway accidents and heart disease. In this article, I discuss the psychological factors that shape the perception of many kinds of risks and the decisions that involve those risks. A SHORT HISTORY OF RESEARCH ON MEMORY AND REASONING In the following section, I outline recent advances in research on the psychology of risk, drawing on modern concepts of memory representation, retrieval, and processing. New ideas about dual processes in memory and reasoning make it possible to predict risk perceptions and decisions that involve risks, rather than merely explain them after the fact. Before proceeding, however, it is useful to discuss the background of research on memory and reasoning, which led up to these new developments regarding risk. The conception of memory that has dominated psychology for decades is the computer metaphor. That is, information is considered to be held in a temporary store, called working memory, that has limited capacity. The idea that working memory has a limited capacity goes back to George Miller’s research on the ‘‘magical’’ number seven (the number of chunks of information that he estimated could be processed at one time). Miller’s work influenced many psychologists, notably Herbert Simon, who was a pioneer in research on judgment and decision making. Relying on the assumption that working memory was limited, Simon introduced the concept of bounded rationality. In short, Simon thought that human rationality is limited because of limitations in human information processing. The assumption that human information processing is limited continued to be the main motivating assumption behind subsequent approaches to judgment and decision making, including both heuristics-and-biases and fast-and-frugal approaches (see Gilovich, Griffin, & Kahneman, 2002). According to these views, humans use heuristics or fast-and-frugal reasoning strategies as mental shortcuts because of information processing limitations. The need to conserve limited mental resources was seen as the driving force in reasoning, judgment, and decision making. Although the computer metaphor guided research for decades, it has recently come under attack. Memory researchers have begun to criticize the concept of limited capacity in working memory (Nairne, 2002). Moreover, evidence indicates that working memory capacity is unrelated to the accuracy of reasoning, judgment, and decision Address correspondence to Valerie Reyna, Department of Psychology, University of Texas at Arlington, Box 19528, Arlington, TX 76019-0528; e-mail: [email protected]. CURRENT DIRECTIONS IN PSYCHOLOGICAL SCIENCE 60 Volume 13—Number 2 Copyright r 2004 American Psychological Society making. Researchers have studied many tasks, and each has shown the same result: Reasoning is independent of memory (Reyna & Brainerd, 1995). In addition, researchers have noted that dramatic errors in reasoning occur on tasks that impose few demands on memory capacity (Shafir & LeBoeuf, 2002). These findings of independence between reasoning and remembering prompted the development of fuzzy-trace theory (Reyna & Brainerd, 1995). Fuzzy-trace theory accounts for this independence through the assumption that people form two kinds of mental representations, verbatim and gist representations, but rely primarily on gist. Gist representations are fuzzy (less precise than verbatim representations) traces of experience in memory, hence the name fuzzy-trace theory. Because two kinds of representations are posited, fuzzy-trace theory is an example of a dual-process model of memory. Fuzzy-trace theory explains findings of reasoning-remembering independence because responses to memory tests often require the details found in verbatim representations, whereas responses to reasoning tests often require only gist representations. Thus, reasoning accuracy is independent of memory accuracy because gist representations are independent of verbatim representations. This explanation was confirmed by experiments in which the reliance on verbatim versus gist representations was actively manipulated (e.g., through instructions or by varying the time delay prior to the memory test), producing positive dependency, negative dependency, and independence between memory and reasoning under theoretically predicted conditions (Reyna & Brainerd, 1995). Finally, assumptions of fuzzy-trace theory have been modeled mathematically to secure quantitative estimates of the contributions of verbatim and gist representations, and associated judgment processes, in a variety of tasks (Reyna, Lloyd, & Brainerd, 2003). Fuzzy-trace theory is a dual-process theory of reasoning as well as memory. People can use either verbatim or gist representations to solve reasoning problems (although they mainly use gist). Researchers who study fuzzy-trace theory treat judgment-and-decision-making tasks as examples of reasoning problems, but work from a conception of reasoning that is different from traditional approaches. Traditional theories of reasoning are modeled on logic or computation; reasoning is said to occur in a series of ordered steps (e.g., premises are first understood and then integrated to draw conclusions), and precision is considered a hallmark of good reasoning. In contrast, according to fuzzy-trace theory, reasoning processes unfold in parallel rather than in series, often operating on the barest senses of ideas (the gist of a problem), and are fuzzy or qualitative rather than precise. Thus, a person presented with a reasoning problem encodes multiple representations of the same problem facts, retrieves reasoning principles from his or her stored knowledge (e.g., the principle that probability depends on the number of wins out of the total number of plays), and applies the reasoning principles to the mental representations of the problem facts. In this view, human reasoning is a messy process: Multiple perceptions of the problem are encoded, the right reasoning principle might or might not be retrieved, and the execution of processing (applying principles to problem representations) is unreliable. Processing is considered to be unreliable because of interference (getting bogged down in the execution of processing; discussed in the next section), as opposed to logical incompetence or memory overload. Fuzzy-trace theory differs from other dual-process approaches to reasoning in some important respects. One example of these differences is that fuzzy-trace theorists place intuition at the apex of development rather than at the nadir. This view of intuition is supported by developmental studies of children’s learning and of adults’ acquisition of expertise, which have demonstrated a progression from detail-oriented and computational processes (e.g., trading off the magnitudes of risk and reward) to fuzzy and intuitive processing (people process less information more qualitatively as development progresses; Jacobs & Klaczynski, 2002; Reyna & Ellis, 1994). For example, my colleagues and I found that in deciding whether to admit a patient with chest pain to the hospital, expert cardiologists processed fewer dimensions of information than less expert physicians did and also processed those dimensions in a cruder all-or-none fashion (patients were either at risk or not at risk of an imminent heart attack; Reyna et al., 2003). Another difference between fuzzy-trace theory and other recent dual-process approaches is the role they give emotion in decisions that involve risk. Some dual-process theorists elevate emotion above reason, arguing that decision makers ought to rely on their gut feelings. As in other dual-process theories, emotion is important in fuzzy-trace theory: Gist representations capture the meaning of experience, including its emotional meaning. However, emotion is not viewed as an unerring signal of what is adaptive (Reyna et al., 2003). Fuzzy-trace theory provides an alternative conception of rationality, compared with traditional theories of judgment and decision making. Any theory of rationality must provide a mechanism for achieving what is called the consistency criterion: Superficially different reasoning problems should be treated consistently. For example, if you choose surgery when it is described as having an 80% survival rate, you should still choose it if it is described as having a 20% mortality rate. In fuzzy-trace theory, the mechanism for achieving consistency is found in the sharing of gist. To the rational decision maker who is contemplating surgery, the gist of the risk of surgery is the same whether it is described in terms of survival or mortality rates. Fuzzy processing is also a source of irrational biases and inconsistencies, however. My colleagues and I (Reyna et al., 2003) have developed a taxonomy of these biases and inconsistencies in which degrees of rationality are specified. That is, errors in reasoning, judgment, and decision making are categorized as more or less advanced according to the type of processing that produced them. Some kinds of errors are seen as more irrational—more serious failings—than others. In sum, ideas about working memory capacity that were traditionally invoked in judgment-and-decision-making research have been challenged by findings that reasoning and remembering are independent. Researchers such as Simon simply assumed that working memory capacity was an important factor, but research that has actually examined the relation between memory and reasoning does not support this assumption. Fuzzy-trace theory emphasizes alternatives to capacity explanations, independence of reasoning performance from memory constraints, and dual-process assumptions about memory and reasoning. Intuition has a special place in fuzzy-trace theory, and is considered an advanced form of reasoning because of developmental evidence about the typical sequence of errors as reasoners gain expertise in reasoning. Intuitions in reasoning come about as a result of parallel processing of multiple representations, uncertain retrieval of reasoning principles, and an overarching preference for gist representations (as opposed to verbatim representations). What do these assumptions predict about risk perception and decision making? Volume 13—Number 2 61 Valerie F. Reyna