ANRV364-PS60-03 ARI 27 October 2008 16:10 Mindful Judgment and Decision Making Elke U. Weber and Eric J. Johnson Center for the Decision Sciences (CDS), Columbia University, New York, New York 10027; email: [email protected] Annu. Rev. Psychol. 2009. 60:53–85 Key Words First published online as a Review in Advance on choice, preference, inference, cognition, emotion, attention, memory, September 17, 2008 learning, process models The Annual Review of Psychology is online at psych.annualreviews.org Abstract This article’s doi: A full range of psychological processes has been put into play to explain 10.1146/annurev.psych.60.110707.163633 judgment and choice phenomena. Complementing work on attention, by Columbia University on 04/23/09. For personal use only. Copyright c 2009 by Annual Reviews. information integration, and learning, decision research over the past All rights reserved 10 years has also examined the effects of goals, mental representation, 0066-4308/09/0110-0053$20.00 and memory processes. In addition to deliberative processes, automatic Annu. Rev. Psychol. 2009.60:53-85. Downloaded from arjournals.annualreviews.org processes have gotten closer attention, and the emotions revolution has put affective processes on a footing equal to cognitive ones. Psy- chological process models provide natural predictions about individual differences and lifespan changes and integrate across judgment and deci- sion making ( JDM) phenomena. “Mindful” JDM research leverages our knowledge about psychological processes into causal explanations for important judgment and choice regularities, emphasizing the adaptive use of an abundance of processing alternatives. Such explanations sup- plement and support existing mathematical descriptions of phenomena such as loss aversion or hyperbolic discounting. Unlike such descrip- tions, they also provide entry points for interventions designed to help people overcome judgments or choices considered undesirable. 53 ANRV364-PS60-03 ARI 27 October 2008 16:10 of neuroscience methodologies to complement Contents behavioral research, the field has started to re- alize, however, that the brain that decides how INTRODUCTION .................. 54 to invest pension money and what car to buy ATTENTION ....................... 56 is the same brain that also learns to recognize Exogenous Influences .............. 57 and categorize sounds and faces, resolves per- Endogenous Influences............. 58 ceptual conflicts, acquires motor skills such as ENCODING AND EVALUATION . 59 those used in playing tennis, and remembers (or Evaluation is Relative .............. 59 fails to remember) episodic and semantic infor- Choice from External Search ....... 60 mation. In this review, we make a strong case Inferences from External Search .... 62 for the utility of this realization. Goal and Framing Effects .......... 62 JDM reviews are often structured by task MEMORY PROCESSES ............. 62 categories, with section headings such as “pref- Memory Storage and Retrieval ..... 62 erences,” “beliefs,” and “decisions under risk Memory and Inference ............. 63 and uncertainty” (Payne et al. 1992), and “risky MULTIPLE INFORMATION choice,” “intertemporal choice,” and “social PROCESSES...................... 65 decisions” (Loewenstein et al. 2007). In con- The Emotions Revolution .......... 65 trast, our review employs headings that might Affective Processes ................. 65 be found in a cognitive psychology textbook. Dual-Process Explanations ......... 67 It capitalizes on the 50 years of research on Dual-Representation Models ....... 69 cognitive and motivational processes that have LEARNING ......................... 70 followed Simon’s (1957) depiction of human Predictive Accuracy ................ 71 decision makers as finite-capacity information CHARACTERISTICS OF THE processors and decision satisficers. Attentional DECISION MAKER .............. 72 (in particular, perceptual) and learning pro- Gender ............................ 72 cesses have a longer history of consideration, Age............................... 73 with phenomena such as “diminishing sensi- Personality ........................ 73 tivity of outcomes” or “reference point encod- Cognitive Traits/Styles ............. 73 ing” for perception and the “illusion of validity” INCREASING POLICY for learning. Affective, memory, and predic- RELEVANCE..................... 74 tion processes have only more recently emerged Health ............................ 74 as explanations of judgment and choice Wealth ............................ 75 by Columbia University on 04/23/09. For personal use only. phenomena. Implications: The Behavioral We retain some task category distinctions Advantage ...................... 75 to organize specific content where appropriate. CONCLUSIONS .................... 75 Annu. Rev. Psychol. 2009.60:53-85. Downloaded from arjournals.annualreviews.org Thus, we distinguish between preference and inference. Preferences involve value judgments and are therefore subjective, such as deciding how much to charge for an item on eBay. Infer- INTRODUCTION ences are about beliefs, such as the judged likeli- Since its origins in the 1950s, judgment and de- hood that a political candidate will win the next cision making ( JDM) research has been dom- election, and typically have objectively verifi- inated by mathematical functional relationship able answers. Although this distinction reflects models that were its point of departure in the tradition, it may not reflect psychological real- form of normative models. This focus on eco- ity. Preferences and inferences seem to draw on nomics and statistics may have led JDM re- the same cognitive processes. JDM: judgment and decision making search to underutilize the insights and meth- Our ability to organize our review by psy- ods of psychology. Aided by the recent arrival chological processes is a sign of the growing 54 Weber · Johnson ANRV364-PS60-03 ARI 27 October 2008 16:10 maturity of the field. JDM research no longer simply generates a growing list of phenomena PROCESS MODELS AND PROCESS TRACING that show deviations from the predictions of normative models. Instead, it has been devel- Early models in decision research attempted to explain changes oping and testing hypotheses about the psycho- in judgments or decisions (the “output”) as a result of changes logical processes that give rise to judgments and in information considered (the “inputs”) using tools such as re- choices and about the mental representations gression and analysis of variance. This approach is problematic used by these processes. Although the number because it considers only a subset of observable behavior and be- of JDM articles in major social psychology jour- cause different models can predict one set of outputs from a given nals remained constant over the past 10 years, set of inputs. Process models help because they consider more the number of JDM articles in major cognitive variables and add multiple constraints. By virtue of hypothesiz- psychology journals increased by 50% over that ing a series of psychological processes that precede a judgment period, reflecting the increased interest in inte- or choice, they make predictions about intermediate states of the grating judgment and choice phenomena with decision maker, between the start and end of the decision (“What the frameworks of hot and cold cognition. external information is sought out? What facts are recalled from New tools have undoubtedly contributed to memory?”). Process models also make predictions about the tem- this trend. This includes functional imaging and poral order of these states (“What will a decision maker think other neural and physiological recordings, pro- about first, second, etc.?”). Process data are the data used to test cess tracing tools (see sidebar Process Models hypotheses about these intervening processes and intermediate and Process Tracing), and, increasingly, mod- states. They include functional imaging and other measures of eling tools such as mediation (Shrout & Bolger localized brain activation, response times, verbal protocols, eye- 2002) and multilevel analysis (Gelman & Hill movement tracking, and other information-acquisition tools (see 2007). A focus on psychological mechanisms www.mouselabweb.org). has guided the decomposition of JDM task behavior into contributing cognitive processes and their variation across groups (Busemeyer tion. The recognition that preferences are typ- & Diederich 2002, Stout et al. 2004, Wallsten ically constructed rather than stored and re- et al. 2005, Yechiam et al. 2005). An increased trieved (Lichtenstein & Slovic 2006) may be focus on individual differences has been a no- psychology’s most successful export to behav- ticeable feature of behavioral decision research ioral economics and the policy community and over the past decade. Increased use of Web- illustrates the utility of psychological process based experimentation (Birnbaum & Bahra explanations. We now know how, and increas- by Columbia University on 04/23/09. For personal use only. 2007) allows access to respondents with much ingly why, characteristics of choice options and broader and representative variation on demo- task guide attention, and how internal memory graphic and cognitive variables, with new in- or external information search and option com- Annu. Rev. Psychol. 2009.60:53-85. Downloaded from arjournals.annualreviews.org sights about individual, group, and life-span dif- parison affect choice in path-dependent ways. Preferences: in ferences