Journal of Behavioral Decision Making J. Behav. Dec. Making, 16: 93���106 (2003) Published online 19 February 2003 in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/bdm.434 Searching for Certainty in an Uncertain World: The Difficulty of Giving Up the Experiential for the Rational Mode of Thinking YAACOV SCHUL* and RUTH MAYO HebrewUniversity ofJerusalem, Israel ABSTRACT Our research explores predictions that people make in a simple environment consisting of sequences of a binary signal followed by two possible outcomes. In order to optimize their prediction success, respondents should use a very simple decision rule, called maximization, whereby they consistently predict according to the signal. In line with past research, our findings show that even respondents who realized after the experi- ment that maximization is optimal failed to use it during the experiment itself. We dis- cuss conditions that weaken or reinforce behaving according to the optimal rule in a repeated choice situation. Experiment 1 shows that individuals who are forced to plan their strategy and justify their actions are more likely to discover and use the optimal rule than those not forced to do so. Thinking about the appropriateness of one���s perfor- mance can be done in two different orientations: focusing on the past (justifying past actions) or on the future (planning future action). Experiment 2 shows that planning induces rule-base thinking, while justifying fails to do so. These findings are discussed within a theoretical framework which suggest an interplay between the experiential and the rational modes of processing. Copyright # 2003 John Wiley & Sons, Ltd. key words prediction maximization rational and experiential modes of processing Recently there has been an increase of interest in the suggestion that there are two distinct modes of infor- mation processing: experiential and rational (Epstein, 1994 Lieberman, 2000 Nisbett et al., 2001 Sloman, 1996 Smith & DeCoster, 2000). Although they vary in terminology, focus, and some important details, the different models agree that the rational mode of processing uses rules to process information abstractly and analytically, whereas the experiential mode engages associative processes that are applied to particular epi- sodes, concrete images, and narratives. One of the significant differences between the two modes of proces- sing has to do with selectivity. Associative processing in the experiential mode takes a form of pattern completion cued by the salient features of the input (Smith & DeCoster, 2000), with little or no control Copyright # 2003 John Wiley & Sons, Ltd. * Correspondence to: Professor Yaacov Schul, Department of Psychology, Hebrew University, Mount Scopus, Jerusalem 91905, Israel. E-mail: yschul@mscc.huji.ac.il Contract/grant sponsor: US-Israel Binational Science Foundation (BSF).

of the individual (Lieberman et al., in press). The rational mode of thinking, in contrast, can selectively focus on relevant cues from the input and use them in symbolically represented rules. Donovan and Epstein (1997) argue that the experiential mode is the default mode of information proces- sing, so that everyday behavior is primarily determined by it (see also Lieberman et al., in press). The dom- inance of experiential thinking is consistent with many demonstrations which show that people are sensitive to minute details associated with the information assembly, tending to utilize concrete, often irrelevant, individuating information when making predictions and immediate choices (Trope & Liberman, 2000 Ross & Nisbett, 1991). In an analogy to the pattern-completion process attributed to the experiential mode of processing, likelihood estimates are effected by how ���good��� the story is. A detailed scenario consisting of what appears to be a causally linked chain of events can give rise to a higher likelihood estimate than a scenario containing a disconnected subset of these events (Tversky & Kahneman, 1983 Pennington & Hastie, 1992). It might not be surprising that the experiential mode is used for processing highly complex narratives and events. Our study explores very impoverished events and tests whether, in line with the hypothesized dom- inance of the experiential over the rational mode, individuals fail to use optimal decision rules consistently in processing such events and using them in making predictions. Moreover, we investigate whether the failure to apply optimal decision rules reflects respondents��� ignorance, namely, their failure to appreciate that the rules are optimal. We hypothesize that the answer is negative. That is, we hypothesize that the inclination to use the experiential mode of processing can dominate decision making even when individuals realize that rule-based predictions are optimal. In such cases individuals may deviate from rule-based predictions or choices because of the temptation to use all the available individuating information. The use of such information is appealing because it allows individuals to be responsive to minute details and sensitive to the configuration of informational cues. More often than not, however, decision makers are not cognizant of the possibility that minute details may merely reflect error variance, so that using such details might lead them astray (Einhorn, 1986). In fact, as people attempt to construct stories that explain more of the variance of a to-be-predicted event, they tend to devaluate the significance of the error compo- nent in making predictions (Dawes, 1994). Moreover, using the individuating information gives decision makers an illusion of control. They are in a unique position to make a decision by considering the various informational cues, weighting and combining them in a way that uniquely reflects the interactions between them (Dawes, 1994). Hence they believe that they can be correct in every decision they make. In contrast, rule-based decisions are more mechanized. In making decisions in this fashion, one loses the ability to consider the unique aspects of the individual case (Einhorn, 1986). Giving up control over one���s action may be aversive, leading individuals to try to discover non-existent contingencies so they can out-guess the opportunities afforded by using the rule alone (Braverman & Fischer, 1968). The contrast between the rule-based and the experiential modes of processing is particularly interesting when people have the opportunity to experience the same situation several times. In such cases people can react to the multitude of cues associated with each particular trial or select only those specified by an abstract rule. The repetition may allow people to understand that there is no foolproof course of action and that rule- based performance is optimal. Are people able to completely ignore most of the trial-specific information and select the relevant cues, thus consistently making their decisions according to the optimal rule in such cases? To illustrate, assume that every morning Linda has to decide whether or not to take an umbrella when she goes to the bank. Assume further that Linda listens to the forecast because she knows that weather forecasts are more likely than not to be correct. We can ask whether Linda follows the rule of taking an umbrella if and only if the forecast predicts rain, or whether she sniffs the air, examines the clouds, ponders the history of rain, and the like, believing that she has individuating information that makes it possible to predict rain on a particular day over and above what the forecast affords. 94 Journal of Behavioral Decision Making Copyright # 2003 John Wiley & Sons, Ltd. Journal of Behavioral Decision Making, 16: 93���106 (2003)