Emotion and Rationality 1

Emotion and Rationality: A Critical Review and Interpretation of Empirical Evidence

Michel Tuan Pham Columbia University

The relation between emotion and rationality is assessed by reviewing empirical findings from multiple disciplines. Two types of emotional phenomena are examined—incidental emotional states and integral emotional responses—and three conceptions of rationality are considered—logical, material, and ecological. Emotional states influence reasoning processes, are often misattributed to focal objects, distort beliefs in an assimilative fashion, disrupt self-control when intensely negative but do not necessarily increase risk-taking. Integral emotional responses are often used as proxies for values, and valuations based on these responses exhibit distinct properties: efficiency, consistency, polarization, myopia, scale- insensitivity, and reference-dependence. Emotions seem to promote social and moral behavior. Conjectures about the design features of the affective system that give rise to seeming sources of rationality or irrationality are proposed. It is concluded that any categorical statement about the overall rationality or irrationality of emotion would be misleading.

The relation between emotion and rationality, affect emotional states and reviews their effects on and reason, is an ageless question. This question has reasoning, belief accuracy, self-control and risk- preoccupied philosophers, commoners, and classical taking. The third section focuses on the role of writers for many centuries. It is only recently, integral affective responses in judgment and decision however, that it has become the subject of scientific making. This section identifies distinct properties of inquiry and empirical investigations. In the past 20 affective responses as proxies for value and evaluates years, investigations related to this question have the “somatic marker hypothesis.” The fourth section been conducted across a wide range of scientific examines the role of emotions in social and economic disciplines including cognitive and social interactions. The concluding section discusses psychology, economics, decision research, consumer identified empirical regularities and advances research, and neuroscience. Unfortunately, because theoretical conjectures about the principles of an empirical studies are necessarily grounded in a affective system of judgment and behavioral certain theoretical, substantive, and methodological regulation that gives rise to seeming sources of context, any one study can provide, at best, only a rationality and irrationality. It is concluded that any very partial answer to the extremely complex categorical statement about the overall rationality or question of emotion and rationality. The empirical irrationality of emotion may be simplistic and literature on emotion and rationality is thus very misleading. fragmented and sometimes seemingly inconsistent. What is needed, therefore, is a comprehensive review Types of Emotional Phenomena and Types of of the wide range of empirical findings that have Rationality emerged across various literatures about the relation Emotions refer to complex states of the between emotion and reason. This is the object of this organism characterized by changes in autonomic article. nervous system arousal accompanied by distinct The article is structured in five sections. The physiological expressions, specific action tendencies, first section introduces distinctions between two and subjective feeling experiences of a certain types of emotional phenomena—incidental emotional valence (see Strongman, 1987). Emotions generally, states and integral affective responses—and three though not always, arise from a cognitive appraisal of conceptions of rationality—logical, material, and the emotional object or situation in terms of its ecological. The next section focuses on incidental meaning for one’s well-being (Lazarus, 1991). In this

This article may not exactly replicate the final Review of General Psychology, Vol. 11 (2), 155-158 version published in the APA journal. It is not the 1 copy of record. Copyright American PsychologyPsychological Association. Emotion and Rationality 2 review, the term “emotion” will be used somewhat choice, rationality requires that preferences be broadly to refer to the presence of affect in general. It transitive: If a person prefers A over B and prefers B will be used not only in reference to emotions over C, then this person must also prefer A over C. proper—that is, intense affective experiences such as Similarly, according to normative (Bayesian) rules of anger, fear, joy, and love that have clear emotional inference, if a person has to guess which of two types referents—but also in relation to milder affective of taxis was more likely involved in a hit-and-run responses, feelings, and states, including moods that accident, it would be rational to take into account the do not have clear referents. This wide-ranging use of relative proportion of each type of taxi in the area. the term emotion is intentional. If one is to have a full This first conception of rationality has been referred appreciation of the rationality or irrationality of to as logical (Kahneman, 1994). emotional phenomena, it is important not to restrict A second conception of rationality one’s analysis to the most intense emotional emphasizes the consistency between a person’s experiences.1 decisions and actions and this person’s objectives and When studying the effects of emotion on self-interests. According to renowned economist judgment, decision, and behavior, two types of Amartya Sen (1990, p. 210), “rationality … demands emotional phenomena should be distinguished: cogent relations between aims and objectives actually incidental emotional states and integral emotional entertained by the person and the choices that the responses (Bodenhausen, 1993). Incidental emotional person makes.” This conception is central to standard states are those whose source is unrelated to the economic theorizing where it is posited that rational object of judgment or decision. These states include individuals choose courses of actions in a way that current emotions not caused by the target object, maximizes these individuals’ own utility. Choices of preexisting mood states, and enduring emotional inferior alternatives are irrational, so are behaviors dispositions such as chronic anxiety. Integral that are not in the person’s self-interest (e.g., emotional responses are those experienced in relation compulsive gambling, excessive smoking, to the object of judgment or decision. More unprotected sex with strangers). This second specifically, integral affective responses are emotions conception of rationality may be referred to as and feelings that are elicited by features of the target material. object, whether these features are real, perceived, or The study of emotion raises a third type of only imagined (Cohen, Pham, & Andrade, in press).2 rationality. Certain types of behaviors and actions are Three conceptions of rationality also need to “rational” not because of they are logically consistent be distinguished in discussing the relation between or serve the person’s self-interest, but because they emotion and rationality. The first conception fulfill broader societal goals, meet higher moral emphasizes reasoning, consistency, and logic. standards, or serve greater evolutionary purposes. According to the Webster’s New World Dictionary Some of these behaviors and actions, in fact, may be (Neufeldt, 1991, p. 1115), the word rational implies against the person’s material self-interest. For “the ability to reason logically, as by drawing example, it would not be in a bystander’s self-interest conclusions from inferences.” People are rational to take on an armed mugger and attempt to rescue the (irrational) if their beliefs, judgments, choices, and mugger’s victim. However, if the bystander elects to actions respect (violate) certain standards of logic. do so, one could hardly call this act irrational. Such For example, in the standard economic theory of benevolent, altruistic acts are quite reasonable, even desirable, from a societal or moral standpoint, even if 1 For instance, in his treaty on emotion and they seem irrational from a strictly material rationality, Elster (1999) concentrates his analysis on standpoint. Similarly, people’s almost universal intense emotional experiences of the kind discussed by attraction to certain ideals of beauty may seem classical writers. This analytical strategy introduces two irrational from a logical standpoint and could also be major sampling problems. First, it is unlikely that materially irrational if it leads to unfortunate classical writings, however insightful, are statistically outcomes (e.g., heartbreak). However, there is representative of human reality. Second, even if they evidence that the attraction to certain standards of were, an exclusive focus on intense emotions is bound beauty is sensible from an evolutionary standpoint to overstate their general consequences. (see Etcoff, 1999). Certain behaviors and attitudes 2 Unfortunately, space constraints prevent a may therefore be “rational,” not in the logical or discussion of the extensive work on memory for material sense, but in terms of their consistency with affective experiences and affective forecast of future societal goals, moral standards, or evolutionary experiences (see, e.g., Kahneman, 1994).

Review of General Psychology, Vol. 11 (2), 155-158 2 Emotion and Rationality 3 purposes. This third form of rationality can be termed analogical problems (Keinan, 1987; Leon & Revelle, ecological in that it reflects humans’ ability to relate 1985), and (f) process persuasion arguments less to their environment, whether social, cultural, or thoroughly (Sanbonmatsu & Kardes, 1988, but see natural.3 A primary function of emotions may in fact Pham, 1996). Intense emotional states such as anxiety be to support this ecological form of rationality. therefore appear to produce deficits in people’s reasoning abilities. However, this conclusion needs to Rationality/Irrationality of Incidental Emotional be qualified in several respects. States Because most of these findings pertain to the effects of high anxiety, it is not clear whether they Emotional states are incidental if their source generalize to other intense emotions (e.g., joy, anger, is unrelated to the object of judgment or decision. intense pride). For example, unlike other intense Incidental emotional states have a variety of rational emotions, anxiety involves a cognitive element of and irrational influences on judgments, decisions, and worry that could be driving some of the above- behaviors. They influence people’s reasoning described deficits. In addition, the effects of intense processes, the accuracy of their beliefs, their ability arousal on cognitive performance are not always to exert self-control, and their tendency to take risks. negative (Humphreys & Revelle, 1984). Finally, They are also often misattributed to target objects. states of intense emotional arousal appear to benefit reasoning in at least one respect. In task settings Effects of Emotional States on Reasoning where multiple cues are available, emotionally aroused individuals seem to adjust to their reduced Emotional states influence people’s processing capacity by narrowing down their cue- reasoning processes, and therefore their logical utilization to the more diagnostic cues at the expense rationality. The desirability of these influences seems of the less diagnostic cues (e.g., Bacon, 1974; R. to be a function of the intensity of the states, their Hockey, 1973). As a result, states of high emotional valence, and their appraisal content. arousal tend to increase the relative reliance on Most intense emotional states, except diagnostic versus nondiagnostic information in sadness, are accompanied by high levels of judgment (M. T. Pham, 1996). autonomic arousal, which is known to impair Milder emotional states also influence working memory capacity (S. Darke, 1988a; reasoning processes. Compared to neutral moods, Humphreys & Revelle, 1984). This decrement in good moods have been found to lead individuals to processing capacity has a variety of consequences (a) categorize objects more broadly (Isen & that seem detrimental to sound reasoning. For Daubman, 1984; Isen, Niedenthal, & Cantor, 1992), example, compared to nonanxious participants, (b) generate more creative answers in response- anxious participants tend to (a) have lower ability to generation tasks (T. R. Greene & Noice, 1988; Hirt, recall information and organize this information in Melton, McDonald, & Harackiewicz, 1996), (c) memory (Mueller, 1977, , 1978), (b) take longer to perform better in problem-solving tasks that require verify the validity of logical inferences (S. Darke, ingenuity (T. R. Greene & Noice, 1988; Isen, 1988b), (c) scan alternatives in a more haphazard Daubman, & Nowicki, 1987), and (d) solve multi- fashion (Keinan, 1987), (d) select an option without attribute decision problems more efficiently (Isen & considering every alternative (Keinan, 1987), (e) Means, 1983). According to Isen (2001), these and commit more errors in geometric and semantic other findings show that positive moods have generally beneficial effects on reasoning, problem 3 Clore (2005) recently proposed that, in the solving, judgment, and decision making. This study of emotion and rationality, the notion of “value” conclusion needs to be tempered because other should be defined more broadly than in typical studies suggest that positive moods lead to poorer economic discourse. In addition to (a) the desirability of reasoning performance in a variety of respects. outcomes (captured by the notion of material Positive mood individuals are more likely to rationality), Clore suggests including (b) the consistency overestimate the degree to which others’ actions are of actions with standards and (c) the consistency of driven by personals disposition as opposed to attributes of objects with tastes and attitudes. The situational factors, a bias known as the “fundamental ecological rationality proposed here subsumes Clore’s attribution error” (Forgas, 1998). Positive mood has second category (the consistency with standards) and also been found to decrease performance in deductive part of his third category, namely the consistency with reasoning tasks (Oaksford, Morris, Grainger, & tastes and attitudes rooted in evolutionary logic.

Review of General Psychology, Vol. 11 (2), 155-158 3 Emotion and Rationality 4

Williams, 1996) and result in more intransitive beneficial effects. More intense states of sadness such preferences (Fiedler, 1988). Numerous attitude and as chronic depression seem to interfere with persuasion studies also indicate that positive moods reasoning and effortful processing (Conway & decrease the depth with which people process Giannopoulos, 1993; Hartlage, Alloy, Vazquez, & substantive information (Batra & Stayman, 1990; Dykman, 1993; Silberman, Weingartner, & Post, Bless, Bohner, Schwarz, & Strack, 1990; Bless, 1983). Mackie, & Schwarz, 1992; Mackie & Worth, 1989; Worth & Mackie, 1987). Positive mood individuals Misattribution of Incidental Affective States seem to rely instead on global knowledge structures and internal cues including scripts (Bless, Schwarz, A robust result about the psychology of Clore, Golisano, & Rabe, 1996), stereotypes emotion is that people tend to attribute their affective (Bodenhausen, Kramer, & Suesser, 1994), and states to whatever object is the current focus of their judgmental heuristics such as ease of retrieval (Ruder attention (N. Schwarz & Clore, 1996). When the & Bless, 2003). Overall, positive moods seem to have object of attention is indeed the source of feelings, mixed effects on people’s reasoning. On the one this is logically rational. However, people tend to hand, they seem to promote greater flexibility and make this attribution even when the actual source of creativity in problem solving, which appears logically the feelings is totally unrelated to the object of desirable; on the other hand, they seem to promote a attention. In a classic study, Schwarz and Clore more top-down, less data-driven, and less thorough (1983) found that respondents who were in a good mode of processing, which appears logically less mood as a result of being surveyed on a sunny day desirable. reported higher levels of life satisfaction than Negative moods, especially those of the respondents who were in a bad mood as a result of sadness kind, have generally been found to have being surveyed on a rainy day. Respondents effects that mirror those described above. Compared mistakenly inferred that their weather-induced moods to neutral and pleasant moods, sad moods have been reflected how they felt about their personal lives. found to increase the care with which people process Similar logically irrational misattributions have been substantive information in persuasion (Bless, Bohner, found in numerous studies showing that incidental Schwarz, & Strack, 1990; Sinclair, Mark, & Clore, mood states generally have assimilative (mood- 1994), decrease the reliance on general knowledge congruent) effects on object evaluation (Albarracin & structures such as scripts and stereotypes (Bless, Kumkale, 2003; Ottati & Isbell, 1996; Siemer & Schwarz, Clore, Golisano, & Rabe, 1996; Reisenzein, 1998). Bodenhausen, Kramer, & Suesser, 1994), increase The widespread misattribution of incidental the ability to estimate covariation from scatter plot mood states may explain the puzzling effects of data (Sinclair & Mark, 1995), reduce the weather on the stock market, the presumed citadel of susceptibility to halo effects (Sinclair, 1988), reduce rationality. In a challenge to the hypothesis that fundamental attribution errors (Forgas, 1998), and financial markets are efficient, a number of studies increase the transitivity of preferences (Fiedler, have recorded above-average stock market 1988). Overall, sad moods seem to trigger a more performance on sunny days and below-average systematic, data-driven, and analytical form of performance on rainy and winter days (Hirshleifer & reasoning consistent with logical rationality. One Shumway, 2003; Kamstra, Kramer, & Levi, 2003; possible explanation, based on the “affect-as- Saunders, 1993). A plausible explanation is that a information” hypothesis (N. Schwarz, 1990), is that sunny weather puts investors in a good mood that sad moods signal to the individual that the situation is they misinterpret as optimism about the stock market, problematic and therefore requires a more vigilant therefore taking more risk; rainy or winter weather form of processing (N. Schwarz, 2002). Not all puts investors in a depressed mood that they negative moods trigger this vigilant form of misinterpret as pessimism about the stock market, processing. States of anger and disgust seem to therefore taking less risk. decrease the depth of processing and increase the People misattribute not only the valence of reliance on stereotyping and other heuristic cues, their incidental affective states, but also their arousal apparently because these states trigger a sense of and cognitive appraisal components. Numerous certainty (Bodenhausen, Sheppard, & Kramer, 1994; studies have shown that incidental emotional arousal Tiedens & Linton, 2001). Note also that only mild is often misinterpreted as an integral affective states of sadness (sad moods) seem to produce these response to a target, resulting in more polarized

Review of General Psychology, Vol. 11 (2), 155-158 4 Emotion and Rationality 5 evaluations of this target (Dutton & Aron, 1974; (Dommermuth & Millard, 1967). These assimilative Foster, Witcher, Campbell, & Green, 1998; Gorn, influences extend beyond strict evaluative judgments. Pham, & Sin, 2001; Mattes & Cantor, 1982; White, For example, risks are perceived to be higher under Fishbein, & Rutsein, 1981; Zillmann, 1971). negative moods than under positive moods (Johnson Incidental arousal is misconstrued as an integral & Tversky, 1983; Wright & Bower, 1992) and under response to the target, resulting in polarization incidental states of fear than under incidental states of because people “feel strongly about” the target anger (Lerner & Keltner, 2001). Sad events are also (Gorn, Pham, & Sin, 2001). People also seem to perceived to be more likely under incidental states of interpret the appraisal content of their incidental sadness and angering events more likely under emotional states as if the states were related to the incidental states of anger (DeSteno, Petty, Wegener, target. For example, Lerner, Small, and Loewenstein & Rucker, 2000). These assimilative influences can (2004) found that incidental states of disgust be explained in terms of misattribution: incidental decreased the perceived value attached to possessions affective states may be misinterpreted as integral that were objectively unrelated to the source of affective responses to the target (N. Schwarz, 1990). disgust. This is presumably because disgust typically However, other processes may also be at work signals an internal source of discomfort, which (Forgas, 1995), including selective encoding or encourages the rejection of possessions. retrieval of affect-consistent information under The degree to which people misattribute incidental emotional states (Bower, 1981; Isen, their affective states is not absolute, however. Shalker, Clark, & Karp, 1978). Misattributions usually disappear when people are While momentary emotional states often made aware of the true source of their affective states distort people’s perceptions, it has been hypothesized (e.g., Gorn, Goldberg, & Basu, 1993; N. Schwarz & that chronically depressed individuals have more Clore, 1983; Siemer & Reisenzein, 1998). This accurate perceptions of reality than nondepressed explains why intense emotions, whose source tends individuals, whose perceptions tend to be self- to be salient, are less likely to be misattributed. enhancing—a hypothesis known as depressive Attributions seem to depend on the perceived realism (Alloy & Abramson, 1988). In a seminal representativeness of the affective state with respect study, Alloy and Abramson (1979) found that to the target (M. T. Pham, 1998; Strack, 1992). For depressed individuals had more accurate perceptions example, incidental affective states are more likely to of the contingency between their behavior and some be misattributed when there is a surface resemblance environmental outcome than did nondepressed between the source of the affective state and the individuals who tended to overestimate this domain of the decision (Raghunathan, Pham, & contingency when the outcome was desirable and Corfman, 2006). underestimate it when the outcome was undesirable. Seemingly more objective perceptions among Effects of Emotional States on Belief Accuracy depressed individuals have been observed in several other studies (e.g., Alloy & Ahrens, 1987; Gotlib, A basic requirement of logical rationality is McLachlan, & Katz, 1988; Keller, Lipkus, & Rimer, an accuracy of perceptions and beliefs. A large body 2002; Lewinsohn, Mischel, Chaplin, & Barton, 1980; of evidence indicates that incidental affective states Martin, Abramson, & Alloy, 1984; see Dobson & tend to distort people’s perceptions and beliefs about Franche, 1989 for a review). For example, compared objects in an assimilative fashion (Isen, Shalker, to nondepressive individuals, depressive individuals Clark, & Karp, 1978; Mayer, Gaschke, Braverman, & have been found to attend more evenly to positive, Evans, 1992), especially if the target is ambiguous neutral, and negative words (Gotlib, McLachlan, & (Gorn, Pham, & Sin, 2001; Isen & Shalker, 1982; Katz, 1988) and revise their estimates of health risks Miniard, Bhatla, & Sirdeshmukh, 1992). For more accurately after receiving medical feedback example, participants who watched a commercial (Keller, Lipkus, & Rimer, 2002). However, other appearing in a happy television program perceived it studies have failed to support the depressive realism to be more effective than participants who watched hypothesis (Benassi & Mahler, 1985; Dunning & the same commercial embedded in a sad program Story, 1991) and suggest that the phenomenon might (Goldberg & Gorn, 1987), and consumers who tasted not generalize to more meaningful and consequential a soft drink after watching a pleasant movie rated the tasks (Pacini, Muir, & Epstein, 1998). For example, beverage’s attributes more favorably than participants Dunning and Story (1991) found that depressed who tasted it after watching an unpleasant movie individuals were less accurate and more

Review of General Psychology, Vol. 11 (2), 155-158 5 Emotion and Rationality 6 overconfident in predicting the probability future phenomenon is not due to a decrease in motivation or personal events than nondepressed individuals. Some ability to self-regulate under negative affective states, analyses suggest that that the original Alloy and but to a shift in priority among distressed individuals Abramson (1979) findings may have been a who seem to place the immediate goal of feeling methodological and statistical artifact (Allan, Siegel, better ahead of other goals (Tice, Bratslavsky, & & Hannah, (in press); Msetfi, Murphy, Simpson, & Baumeister, 2001). Kornbrot, 2005). Empirical support for the depressive Given the disruptive effects that intense realism hypothesis is therefore very mixed. One negative affective states have on self-control, one possible explanation for this inconsistency may be would intuitively predict that these states should also that depression only enhances realism in tasks that make people more risk-risking. The empirical involve some reassessment of the current situation. evidence is in fact very mixed. Several studies This is because, as explained in the concluding indicate that negative emotional states with strong section, sadness-related states may serve as a signal arousal increase risk-seeking (Fessler, Pillsworth, & for situational reassessment. Flamson, 2004; Leith & Baumeister, 1996; Mano, In summary, it seems that incidental 1992, , 1994). For example, Leith and Baumeister affective states undermine logical rationality by (1996) found that participants who were angry or distorting beliefs in an assimilative fashion, anticipated an impending embarrassment were more especially if the target is ambiguous and if the state is likely to choose economically inferior “long-shot” improperly attributed to the target. One debated gambles over superior “safe-bet” gambles, whereas exception could be chronic states of depression, sad participants did not exhibit this bias. Fessler, which may enhance belief accuracy under certain Pillsworth, and Flamson (2004) also found that anger conditions. triggered more risk-seeking in gambling, especially among men. Similarly, Mano (1994) found that Effects of Emotional States on Self-Control and Risk- intense emotional arousal increased the willingness to Taking pay for lotteries and decreased the willingness to pay Improper self-control and excessive risk- for insurance, in other words increased risk-taking for taking (or avoidance) are primarily matters of both potential gains and potential losses. material rationality (although they also raise issues of However, other findings indicate that ecological rationality). It is well established that people’s attitude toward risk under negative emotions intense drive states such as hunger, pain, sexual is not just a function of the level of arousal associated arousal, drug cravings, and sleep deprivation produce with the emotion, but also a function of the appraisal breakdowns in self-control and increase people’s content of the emotion (Lerner & Keltner, 2001; willingness to take risks in order to alleviate the drive Raghunathan & Pham, 1999). Raghunathan and state (Loewenstein, 1996). For example, states of Pham (1999) observed that in risk-reward-tradeoff high sexual arousal increases people’s willingness to situations, anxious individuals tend to prefer low- use unethical means of getting sex and decrease their risk/low-reward options, whereas sad individuals willingness to practice safe sex (Ariely & tend to prefer high-risk/high-reward options—a Loewenstein, 2005; Bouffard, 2002). seeming reversal of the high-arousal/high-risk pattern Negative emotional states, especially intense observed elsewhere (see also Raghunathan, Pham, & ones, produce similar breakdowns in self-control. Corfman, 2006). According to Raghunathan and Negative affective states have been found to (a) Pham (1999), this is because anxiety, which is reduce the ability to resist temptation and delay typically associated with situations of low control and gratification among children (Fry, 1975; J. C. high uncertainty, activates a goal of risk and Schwarz & Pollack, 1977; Seeman & Schwarz, uncertainty minimization, whereas sadness, which is 1974), (b) increase the tendency to overeat among typically experienced in response to the loss of a dieters (Ruderman, 1986), (c) increase the chance of source of reward, activates a goal of reward relapse among people trying to quit smoking maximization. Other studies have also uncovered (Shiffman & Waters, 2004), (d) encourage shopping decreased risk-seeking under anxiety when the level among compulsive buyers (Faber & Christenson, of arousal is held constant (Mano, 1992, , 1994). 1996), (e) increase procrastination (Tice, Bratslavsky, Lerner and Keltner (2001) similarly observed that & Baumeister, 2001), and (f) produce over- fear tends to trigger risk-aversion, whereas anger consumption of limited collective resources (Knapp tends to trigger risk-seeking even though both are & Clark, 1991). Recent studies suggest that this high-arousal negative emotions. This is because fear,

Review of General Psychology, Vol. 11 (2), 155-158 6 Emotion and Rationality 7 like anxiety, is typically associated with situations of are generally reached more rapidly than those based uncertainty and low control, whereas anger is on descriptive inputs, both in stimulus-based tasks typically associated with situations of certainty and (M. T. Pham, Cohen, Pracejus, & Hughes, 2001; high control. Disgust, another high arousal emotion, Zajonc, 1980) and in memory-based tasks has also been found to decrease risk-seeking in (Verplanken, Hofstee, & Janssen, 1998). This seems gambling among women (Fessler, Pillsworth, & to be because integral affective responses can arise Flamson, 2004). very rapidly (LeDoux, 1996; Zajonc, 1980) and enter In summary, unlike drive states, intense evaluations through simple associations (De Houwer, negative emotions do not have a uniformly positive Thomas, & Baeyens, 2001) or straightforward effect on risk-seeking (see G. R. J. Hockey, Maule, interpretation (M. T. Pham, Cohen, Pracejus, & Clough, & Bdzola, 2000). High emotional arousal Hughes, 2001; Strack, 1992). Judgments and seems neither necessary, nor sufficient to explain decisions based on integral affective responses also risk-seeking under negative emotions. Rather, the generally require less processing resources (Epstein, effects of negative emotions on risk-seeking seem to 1990). This is evident from the robust finding that depend on complex interactions between the goals constraints on processing resources such as time activated by the emotional state and the nature of the pressure, distraction, or cognitive load generally risks to be taken. This may explain why a meta- increase the reliance on integral affective responses analysis of published studies relating chronic states of in judgment and choice (e.g., Avnet & Pham, 2004; anger, sadness, and anxiety to risky sexual behavior Nowlis & Shiv, 2005; M. T. Pham, Cohen, Pracejus, found virtually no correlation (Crepaz & Marks, & Hughes, 2001; Shiv & Fedorikhin, 1999). Overall, 2001).4 it seems that integral affective responses provide fast and resource-efficient assessments of value. Whether Rationality/Irrationality of Integral Emotional these assessments are logically and materially Responses as Proxies for Value rational depends on a variety of considerations— some of which are evaluated below. However, from Unlike incidental affective states, integral an ecological standpoint, judgmental speed and affective responses are those elicited by perceived or resource-efficiency seem to be desirable properties. imagined features of the target object. Integral Extremity and Polarization. Everything else affective responses play a major role in people’s equal, judgments, decisions, and behaviors based on evaluations of, decisions about, and behavior toward integral affective responses tend to be more extreme objects, even if the attributes of the objects are held and polarized than those based on more descriptive constant (e.g., Abelson, Kinder, Peters, & Fiske, inputs—a phenomenon related to the so-called 1982; Hsee & Kunreuther, 2000; M. T. Pham, 1998; “vividness effect” in judgment (Nisbett & Ross, M. T. Pham, Cohen, Pracejus, & Hughes, 2001). The 1980). For example, in France, newspaper articles primary reason seems to be that integral affective using the emotional label “Mad Cow disease” responses are often used as proxies for value: Things resulted in more dramatic decreases in beef that feel good must be desirable and things that feel consumption than comparable articles using the bad must be undesirable (see, e.g., Damasio, 1994; scientific label “Creutzfeldt-Jakob disease” M. T. Pham, 2004; N. Schwarz & Clore, 1996; (Sinaceur, Heath, & Cole, 2005). In general, Slovic, Finucane, Peters, & MacGregor, 2002), a responses to risks seem to be stronger when the risks pervasive inference that accounts for the often- are conveyed in a emotion-provoking manner than observed assimilative effect of incidental affective when they are conveyed in a nonemotional manner states on target evaluations. (e.g., Hendrickx, Vlek, & Oppewal, 1989; Loewenstein, Weber, Hsee, & Welch, 2001). In a Properties of Integral Emotional Responses as different setting, Ratner and Herbst (2005) observed Proxies for Value that, after investment with a broker with a strong track record produced a disappointing outcome, Speed and Processing Efficiency. Judgments individuals who focused on their affective responses and decisions based on integral emotional responses to the outcome tended to “overreact” and switch their investments to a broker with an inferior track record. Emotional responses to a single outcome appeared to 4 Another explanation could be that chronic override the presumably more reliable track-record of emotional states have lesser effects on risk-taking than the brokers. momentary states.

Review of General Psychology, Vol. 11 (2), 155-158 7 Emotion and Rationality 8

Several factors can explain why evaluative undermine the apparent ecological rationality of this and behavioral responses based on integral affect function. tend to be more polarized. First, affective responses Myopia. Evaluations and decisions based on to everyday objects tend to be more extreme than integral affect tend to be more myopic in that reason-based assessments of the same objects, even immediate affective rewards and punishments are when the information about the object is held weighted much more heavily than delayed affective constant (M. T. Pham, Cohen, Pracejus, & Hughes, consequences (Loewenstein, 1996). This property is 2001). Second, initial affective responses to an object very salient in self-control situations where people seem to trigger a confirmatory search for information have to trade-off the immediate hedonic that supports the initial feelings (M. T. Pham, Cohen, consequences of an option against its long-term Pracejus, & Hughes, 2001; Yeung & Wyer, 2004). consequences. According to Loewenstein (1996), the This confirmatory search increases the subjective myopia of affect-based judgments and decisions is coherence of judgments based on affect (M. T. Pham, due to the differential accessibility of current and 2004), increasing polarization. In addition, as delayed affective responses. Whereas the experience discussed further, affective responses seem to be of immediate integral affect has strong drive relatively insensitive to probability and quantity, properties, it is much more difficult to picture a which would otherwise mitigate interpretations of future affective experience. Consistent with this these responses (Hsee & Rottenstreich, 2004; proposition, recent brain imaging studies indicate that Rottenstreich & Hsee, 2001). Finally, integral preferences for immediate rewards are associated affective responses have inherently strong drive with greater activation in parts of the limbic system properties (Frijda, 1988; Lazarus, 1991). associated with affect (McClure, Laibson, The greater polarization of affect-based Loewenstein, & Cohen, 2004). Affective rules of evaluative and behavioral responses may lead to valuation thus seem to be geared to the present (M. T. logical irrationalities. For instance, Johnson, Hershey, Pham, 2004), which will lead to material irrationality Meszaros, and Kunreuther (1993) observed that in domains where present and long-term hedonic people were willing to pay more for a flight insurance consequences are negatively correlated. policy covering “death due to any act of terrorism”— Concreteness and Scale Insensitivity. As a concrete, emotion-producing threat—than for an epitomized by the notion of expected-monetary insurance policy covering “death due to any value, a logically rational measure of value would reason”—a logically higher, but less concrete threat. take into account the magnitude of the value- Similarly, the “overreaction” of emotion-focused producing stimulus and the uncertainty that surrounds investors observed by Ratner and Herbst (2005) it. It appears, however, that when integral affective seems to be a case of logically irrational responses are used as proxies for value, these overweighing of single instances relative to more responses are not scaled properly for either reliable base-rates. Nevertheless, it would seem magnitude or probability. Hsee and Rottenstreich premature to draw firm conclusions about the general (2004) observed that judgments and decisions based rationality or irrationality of affect-based response on integral affective responses tend to be sensitive to polarizations. For example, in the Ratner and Herbst the presence or absence of affect-producing stimuli (2005) studies, it is not clear that it was irrational for but relatively insensitive to variations in the the emotional investors to place more weight on the magnitude of these stimuli. In one study, respondents broker’s recent performance than on the broker’s were asked how much they would be willing to stated track record, especially in a domain where past donate to save either one or four pandas. When the records are notoriously unpredictive of the future number of pandas saved was represented in an performance. It is also not clear that the decrease in abstract fashion, donations were much higher in the beef consumption following articles emotional titled four-panda condition than in the one-panda condition, “Mad Cow Disease” was materially, ecologically, or as would logically be expected. However, when the even logically irrational. Rather, it seems that number of pandas saved was represented in an response polarization is a natural by-product of an affectively-rich fashion, donations were not different important function of affective and emotional in the four- and one-panda conditions. These results responses, which is to motivate behavior and redirect echo other findings showing that when assessing the action if necessary. That this function occasionally value of programs designed to save a large number of produces “overreactions” does not necessarily human lives—an emotionally charged judgment— people exhibit substantial insensitivity to the absolute

Review of General Psychology, Vol. 11 (2), 155-158 8 Emotion and Rationality 9 number of lives saved (Fetherstonhaugh, Slovic, probabilities of winning in big lotteries, a prospect Johnson, & Friedrich, 1997). According to Hsee and rich in affect, into probabilities that are infinitesimal. Rottenstreich (2004), this phenomenon arises because Similarly, most people would be willing to pay large integral affect-based evaluations are often based on insurance or security premiums to convert minute mental images (see also M. T. Pham, 1998). These probabilities of catastrophic events, prospects also images tend to involve discrete prototypical rich in affect, into zero probabilities. In affective representations of the target but not continuous valuations people thus appear to be sensitive to quantitative information (see also Kahneman, Ritov, possibility—that is, deviations from certainty—rather & Schkade, 1999). than actual probability (see also Slovic, Finucane, Similarly, evaluations and decisions based Peters, & MacGregor, 2002). on integral affective responses tend to be insensitive Overall, the above findings suggest that to probabilities, except for the presence or absence of valuations based on integral affect tend to be uncertainty (Loewenstein, Weber, Hsee, & Welch, insensitive to scale, whether scale refers to the 2001; Monat, Averill, & Lazarus, 1972; Rottenstreich quantitative magnitude of the stimulus or to the & Hsee, 2001; Sunstein, 2003). For example, probability that surrounds it. This scale insensitivity Rottenstreich and Hsee (2001) observed that, violates logical rationality. One interpretation is that consistent with economic theory, people were willing integral affect is a mostly categorical means of to pay much more to avoid a high probability of assessing value. Integral affective responses arise losing $20 than to avoid a low probability of losing from objects being categorized in terms of their $20. However, people were not willing to pay much significance for well-being (Lazarus, 1991). This more to avoid a high probability of receiving an categorization appears to obey a principle of electric shock—a prospect rich in negative affect— concreteness. For objects to be categorized as than to avoid a low probability of receiving the same emotionally significant—that is, for them to elicit shock (see also Sunstein, 2003). According to integral affective responses—they need to be Loewenstein and colleagues (2001), this phenomenon represented concretely. For example, people will be again arises because affective decisions under happier if they know for sure whether they have won uncertainty rely on discrete images of the options that a dinner or a CD than if they are uncertain about do not incorporate probabilities. This is consistent which of these two prizes they have won (Vandijk & with the finding that awareness of the timing of Zeelenberg, 2006). In this representation, the identity imminent threat produces the same level of stress and of the object is more critical than its distribution. In physiological arousal whether the threat has a 5%, the resultant affective accounting of value, saving 50%, or 100% probability of occurrence (Monat, people’s lives or avoiding electrical shocks is Averill, & Lazarus, 1972). This is also consistent important, regardless of the number of lives saved or with the finding that, when given a chance to draw a the actual probability of shock. winning red bean either from a small bowl containing Reference-Dependence. Compared to those a single red bean and 9 white beans or from a larger based on more descriptive inputs, assessments of bowl containing between 5 and 9 red beans and 91 to value that are based on integral affective responses 95 white beans, many people’s “gut feeling” is to tend to be more relativistic or reference-dependent. draw from the larger bowl, even though the That is, affective valuations are often not based on probability of winning is greater when drawing from the focal object or outcome in isolation, but in the smaller one (Denes-Raj & Epstein, 1994; relation to other objects or outcomes. For example, Kirkpatrick & Epstein, 1992). This finding again emotional responses to the outcome of a gamble are suggests that affective valuations of chances are driven not only by the monetary value of the actual driven by concrete representations of exemplars (realized) outcome, but also by how this outcome (seeing multiple red beans) rather than more abstract compares relative to unrealized outcomes (Mellers, notions of probability (the distribution of beans in a Schwartz, Ho, & Ritov, 1997). Winning $10 in a random draw process). gamble will elicit greater pleasure if the other Note, however, that affect-based decisions possible outcome is losing $5 than if it is losing only and evaluations are very sensitive to one range of $1. Similarly, losing $5 in a gamble will elicit greater probability: deviations from absolute certainty, from displeasure if the other possible outcome is winning impossibility to small probability and vise versa $10 than if it is winning only $2. This finding is (Brandstatter, Kuhberger, & Schneider, 2002). For consistent with a large body of research showing that example, people grossly overpay to turn zero emotional responses to outcomes are very sensitive to

Review of General Psychology, Vol. 11 (2), 155-158 9 Emotion and Rationality 10 spontaneous comparisons with outcome employees better. This relativism may be also counterfactuals (e.g., Kahneman & Miller, 1986; beneficial from an ecological standpoint, as discussed Landman, 1987; Medvec, Madey, & Gilovich, 1995). further in the discussion of the role of emotion in More direct evidence comes from a study by Tversky social and economic interactions. In addition, the and Griffin (1991) in which participants were asked counterfactual outcome comparisons that underlie to evaluate two hypothetical jobs: one company emotional experiences of regret, rejoicing, offered a higher salary, but offered other colleagues disappointment, or elation may help people learn even more money; the other offered a lower salary, from their failures and successes (Roese, 1997), but offered other colleagues even less money. When whereas social comparisons with other’s welfare may asked to predict which of the two jobs they would by support norms of justice and equity that seem broadly happier at—that is, when asked to make an affective desirable at a societal level. One possible evaluation—most participants selected the lower- interpretation of this relativism is that the affective paying job. When asked to make a choice between system of valuation is mostly ordinal, as opposed to the two jobs—a presumably less affective cardinal. That is, the affective system may be more evaluation—most participants chose the higher- concerned with the relative desirability ordering of paying job. Therefore, participants asked to make a alternative states of the world or alternative courses “cold” decision seem to focus on the objective of actions than with their absolute desirability. personal payoff, whereas those asked to make a more Interpersonal and Intra-personal affective assessment seem to also take into account Consistency. A growing body of evidence suggests social comparisons with their colleagues. In a similar that judgments based on integral affective responses study, participants were asked to compare two have high interpersonal consistency. Contrary to the hypothetical jobs that were identical in terms of notion that “Beauty is in the eye of the beholder,” compensation and workload: company A offered a judgments of physical attractiveness, which are small office and gave another comparable employee largely affective, exhibit a high degree of consensus an equally small office, whereas company B offered a both within and across cultures (Langlois et al., larger office and gave another comparable employee 2000). Similarly, emotional responses to music an even larger office (Hsee, Zhang, Yu, & Xi, 2003). appear largely shared (Peretz, Gagnon, & Bouchard, When asked to make a cold choice, participants 1998). People also exhibit high consensus in how tended to choose the job with the objectively larger outraged they are by various types of wrongdoings office. However, when asked to assess which job (Kahneman, Schkade, & Sunstein, 1998) and how would make them feel happier, participants tended to upset they feel about various environmental problems select the one with the smaller office. (Kahneman, Ritov, Jacowitz, & Grant, 1993). People The greater relativism or reference- seem to agree more on how they feel toward dependence of affective assessments of value is everyday stimuli such as magazine pictures and TV logically irrational. Logically, utility should depend commercials than they do on their reason-based only on realized and personal outcomes, not on assessments of the same stimuli (M. T. Pham, Cohen, counterfactual comparisons with unrealized outcomes Pracejus, & Hughes, 2001). Therefore, contrary to or social comparisons with others’ welfare. Whether popular beliefs that feelings are highly subjective, a this relativism is also materially irrational depends on variety of findings suggest that judgments based on assumptions about people’s utility function. If we integral affect are quite consensual, sometimes even assume that this function incorporates only objective more so than judgments based on descriptive inputs.5 arguments (e.g., the magnitude of the person’s lottery gain, the absolute size of the person’s office), this 5 According to Pham and colleagues (2001), relativism would be materially irrational. However, if affect-based judgments will tend to be very consensual we assume that people’s long-term well-being also whenever the underlying integral affective responses are depends on the broader context in which personal triggered through hardwired programs involved in bio- outcomes are realized and experienced, this regulation or through emotional schemata acquired relativism may in fact be materially rational. After through socialization. Such affect-based judgments will all, a person working for a more modest salary in a generally be more consensual than reason-based smaller office at a company that treats its employees judgments that are constructed in a piecemeal fashion. with equity may be happier than a comparable However, affect-based judgments will be less employee working for a larger salary in a larger consensual when based on integral affective responses office at a company that treats other comparable arising through controlled appraisal processes. Such judgments will generally be less consensual than

Review of General Psychology, Vol. 11 (2), 155-158 10 Emotion and Rationality 11 There is also growing evidence that In an influential series of studies, patients judgments and decisions based on integral affect also with emotional deficits related to damages in the tend to be more consistent intra-personally. Indirect ventromedial prefrontal cortex area (VMPC) were evidence comes from the finding that individuals who found to perform more poorly than normal and verbalized their reasons for liking or disliking various presumably emotionally-functional participants on a posters before making a choice, and presumably task known as the Iowa Gambling Task (IGT) relied on these reasons, were subsequently less (Bechara, Damasio, Damasio, & Anderson, 1994; satisfied with their choice than individuals who were Bechara, Damasio, Tranel, & Damasio, 1997; not asked to verbalize their reasons and presumably Damasio, 1994). The IGT involves repeated drawings relied on their spontaneous affective responses to the from four decks of cards. Two decks have higher posters (Wilson, Lisle, Schooler, Hodges, & et al., nominal card value but lower expected value due to 1993). More direct evidence comes from the finding severe occasional penalties, and two decks have that, for high-involvement products (e.g., lower nominal card value but higher expected value camcorders, cell phones), integral affective responses due to lesser penalties. Compared to normal to the products are more predictive of long-term participants, VMPC patients were found to draw satisfaction with the products than utilitarian beliefs more from the riskier and less advantageous decks, about the product’s benefits (P. R. Darke, resulting in lower monetary performance (Bechara, Chattopadhyay, & Ashworth, 2006). Other research Damasio, Damasio, & Anderson, 1994). Over time, suggests that reliance on integral affective responses normal participants also exhibited heightened levels increases not only intra-personal consistency over of galvanic skin response whenever they were about time, but also intra-personal consistency over to choose from the risky decks, and this apparently choices. Lee and Ariely (2006) recently observed that even before they could consciously recognize the conditions that are known to increase the reliance on structure of the desks; VMPC patients did not exhibit affect in decision making (e.g., pictures, time such anticipatory activation (Bechara, Damasio, pressure, memory load, expression of own Tranel, & Damasio, 1997). According to the somatic preference) also increase the transitivity of choices marker hypothesis (Damasio, 1994), in normal between products. individuals, emotional responses evoked by objects In summary, compared to judgments and are stored with memory representations of these decisions based on descriptive inputs, judgments and objects as somatic markers of these objects’ value. decisions based on integral affective responses tend Subsequent encounters with these objects will trigger to exhibit higher consistency, both within and across anticipatory feelings that will steer the decision people. Intra-personal consistency is logically maker either toward or away from these objects desirable. It is also materially desirable if it reduces depending on the valence of the stored markers. the chance of post-decisional regret. In domains Among normal participants in the above studies, where a common criterion of value can be assumed, emotional responses to early penalties of the risky interpersonal consistency is also logically desirable decks were registered and subsequently steered because, even if interpersonal agreement does not participants away from these desks, apparently even guarantee accuracy, lack of agreement implies that at unconsciously. Among emotionally-deficient VMPC least one party is inaccurate (Kruglanski, 1989). patients, no such learning took place; as a result, they Consensus is also desirable on ecological grounds. continued to draw from the more tempting but less For example, if a society is to set standards of advantageous decks. According to Damasio and his punishment for various types of crimes, it would be colleagues (see Bechara, 2004; Damasio, 1994), desirable to have a common metric to assess each emotional deficits associated with prefrontal damages crime’s undesirability. The emotional outrage that impair performance not just in the IGT but in people have been found to share could provide such a decision making in general. For example, Bar-On, metric. Tranel, Denburg, and Bechara (2003) found that, compared to control patients with brain lesions Integral Affective Responses as Somatic Markers outside those associated with emotional responding, patients with VMPC damages score lower on various measures of emotional and social intelligence and various dimensions of social functioning such as post-lesion employment status, social status, and reason-based judgments based on shared stereotypes interpersonal relationships. and widely-accepted normative criteria.

Review of General Psychology, Vol. 11 (2), 155-158 11 Emotion and Rationality 12 Although these findings have been widely associated with higher expected returns. Emotionally- interpreted as demonstrating the importance of functional subjects tend to be overly apprehensive emotions for (materially) rational decision making, about taking risks, especially after incurring a loss more recent studies have challenged their original (Shiv, Loewenstein, Bechara, Damasio, & Damasio, interpretation. Findings by Maia and McClelland 2005). (2004) suggest that in the IGT task normal The overall evidence about the somatic participants are more conscious of decks’ structure marker hypothesis seems to warrant only a weaker than previously thought, challenging the notion that (and relatively mundane) version of this hypothesis. somatic markers could guide choice unconsciously. It is well established that integral affective responses Findings by Fellows and Farah (2005) suggest that to a target that are positive generally trigger approach the poor performance of VMPC patients in the tendencies, whereas those that are negative generally original IGT studies may be due to an inability to trigger avoidant tendencies,6 even if descriptions of reverse learned associations acquired in the early the targets and their cognitive assessments are held rounds of the game, not the inability to encode constant (e.g., Abelson, Kinder, Peters, & Fiske, somatic markers. In tasks that do not involve 1982). In other words, integral affective responses outcome feedback—one-shot choices among risky often serve as distinct proxies for value. What the lotteries, inter-temporal preferences, and behavior in Damasio studies, along with other studies (Ratner & wealth-sharing games—VMPC patients and normal Herbst, 2005), suggest is simply that integral-affect- subjects exhibit comparable levels of risk-seeking motivated approach and avoidance—that is, affective and impulsivity (Leland & Grafman, 2005). This behavioral regulation—is very sensitive to emotion- suggests that presumably emotionally-impaired producing outcome feedback. Moreover, the contrast VMPC patients are not inherently more risk-seeking between the original Damasio findings and more and impulsive; rather they differ in how they respond recent Shiv and colleagues findings shows that it is to and learn from outcome feedback. Moreover, not possible to draw firm conclusions from these studies that have used psychopathy as an alternative studies about the general material rationality or operationalization of emotional deficit have irrationality of integral affect as a proxy for value: it uncovered inconsistent effects of psychopathy on depends on the correlation between the emotional performance on the IGT (e.g., Blair & Cipolotti, responses to the target and its criterion value—a 2000; Losel & Schmucker, 2004; Mitchell, Colledge, correlation that is under the researcher’s control. Leonard, & Blair, 2002; Schmitt, Brinkley, & Newman, 1999). The lack of parallelism between Integral Affective Responses in Social and results obtained with VMPC patients and with Economic Interactions psychopaths raises the possibility that an emotional deficit may not be the only factor at work in the Although most of the empirical evidence on original Damasio findings. Finally, even if the emotion and rationality pertains to individuals original VMPC/IGT findings were correct, the considered in isolation, it is important to keep in superior economic performance of emotionally- mind that humans are also part of social groups such functional participants does not, by itself, establish as families, organizations, communities, markets, and the superiority of emotion-based decision-making. societies. As highlighted by the notion of ecological One should note that in the IGT there is a negative rationality, the desirability of emotions should also be correlation between the riskiness of the decks and assessed in terms of how they influence an their long-term expected monetary value. Therefore, individual’s ability to function as a member of a negative emotional responses to the risky decks’ group and how they affect the group’s overall welfare penalties are good predictors of the decks’ actual (see also Clore, 2005). undesirability. In other domains, however, the Emotional responses appear to play an correlation between risks and expected returns is in important and mostly positive role in the regulation fact positive. In such situations, emotional of social and moral behavior (Eisenberg, 2000). apprehension toward taking risks may be materially Emotional responses are a necessary component of detrimental. Consistent with this reasoning, empathic responses, an established determinant of emotionally-deficient VMPC patients (who do poorly pro-social behavior (Eisenberg et al., 1994; Eisenberg in the original Damasio task) were found to performe better than emotionally-functional control subjects in 6 a repeated investment task where investment risk was One exception is anger, which generally triggers a drive to confront the anger-producing object.

Review of General Psychology, Vol. 11 (2), 155-158 12 Emotion and Rationality 13 & Miller, 1987). More importantly, emotions are Pham, Philippot, & Rime, 2000). This lower general very sensitive to the fulfillment or violation of social emotionality may explain psychopaths’ characteristic and moral norms such as honesty, reciprocity, and lack of guilt, remorse, and empathy (Blair, 1995), and loyalty. Guilt and shame, for example, typically arise therefore their common pattern of antisocial and from the perception of having transgressed such immoral behavior (Blair, 1997; Blair & Cipolotti, norms, while pride often arises from the perception of 2000). Therefore, while emotional responses such as having fulfilled or exceeded them. Anger and guilt, shame, indignation, or empathy entail psychic indignation often arise from observing others costs that may seem materially irrational, these transgressing social and moral norms, whereas responses appear to fulfill an important ecological gratitude and admiration are generally reserved for function: that of promoting socially and morally those who fulfill or exceed them (Ortony, Clore, & desirable behavior, both by the person experiencing Collins, 1988). There is growing evidence that the emotion and by others eliciting the emotion. emotional responses are not just consequences of The ecological rationality of these responses social and moral appraisals but also inform these transpires as well in experimental studies of appraisals. For example, people tend to make more economic interactions in which participants play severe moral judgments and become less tolerant of “games” where they interact with one another along a moral violations if the level of repulsion that they specific set of rules to maximize some economic experience when making moral judgments is payoff (see Goette & Huffman, this issue, for a artificially increased through an incidental affect review). For example, in the ultimatum game, two manipulation (Schnall, Haidt, & Clore, 2006; players have to split a given amount of money (e.g., Trafimow, Bromgard, Finlay, & Ketelaar, 2005) or $20). One player, the proposer, makes an offer (e.g., through hypnotic conditioning (Wheatley & Haidt, $5 for you/$15 for me), which the other player, the 2005). Conversely, they become more tolerant of receiver, either accepts or rejects. If the offer is moral violations if they are led to misattribute part of accepted, the money is split accordingly; if it is their integral affective responses to these violations to rejected, neither player receives anything. According external sources (Trafimow, Bromgard, Finlay, & to economic theory, a materially rational receiver Ketelaar, 2005). These findings suggest that people would accept any offer greater than zero because this infer the severity of moral violations in part from offer would always be more attractive than rejecting their spontaneous affective responses to these and receiving nothing. Anticipating this, a materially violations (see also Kahneman, Schkade, & Sunstein, (and logically) rational proposer would make the 1998). Spontaneous emotional responses may in fact smallest offer possible (e.g., $0.05 for you/$19.95 for be the primary means by which many moral me), knowing that it would be accepted. The dilemmas are evaluated, including some that are very logically and materially rational “equilibrium” of this difficult to solve using pure logic (Haidt, 2001). game is thus a split in which the proposer keeps most Neuroimaging studies indicate that parts of the brain of the money and the receiver receives a positive typically associated with emotions are uniquely residual. Numerous studies have shown that this engaged by certain moral dilemmas and that it is this prediction is rarely fulfilled (Camerer, 2003; Roth, emotional engagement that drives the resolution of 1995). Offers of less than 20% of the total amount the dilemma (J. Greene & Haidt, 2002; J. Greene, typically have a 50% chance of being rejected by the Sommerville, Nystrom, Darley, & Cohen, 2001; Moll receiver. Many receivers would therefore sacrifice et al., 2002; Moll et al., 2005). their material self-interest rather than accept an offer The importance of emotions in social and they perceive unfair. These materially irrational moral regulation is also evident from studies of rejections appear to be driven primarily by emotional psychopathic populations. Clinical psychopaths responses of anger following unfair offers (Pillutla & typically exhibit antisocial personality disorders such Murnighan, 1996; Sanfey, Rilling, Aronson, as consistent disregard for social norms, pervasive Nystrom, & Cohen, 2003). Interestingly, however, violations of the rights of others, and a tendency for many proposers seem to anticipate these emotional aggression and violence (Hare, 1985). Numerous responses, and make offers that are more equitable studies suggest that psychopaths tend to have lower and therefore more likely to be accepted. The modal baseline levels of emotional activity and weaker empirical offer in ultimatum games is 40-50%, which physiological responses to emotional stimulation, is generally accepted (Camerer, 2003). Therefore, a especially with respect to negative stimuli (Blair & materially irrational emotional response—anger at an Cipolotti, 2000; Lorber, 2004; Patrick, 1994; T. H. unfair, tough positive offer—and its anticipation lead

Review of General Psychology, Vol. 11 (2), 155-158 13 Emotion and Rationality 14 to an ecologically desirable solution, one that leaves responses to the violation or fulfillment of norms both players better off than the dead-end usually such as fairness, equity, and trust, and anticipations produced with materially rational strategies. This of these responses by the various players, seem to equilibrium is similar to the type of social and moral promote ecologically desirable cooperation and equilibrium discussed above. Transgressions of social norm-consistent behavior. Some theorists suggest norms (here, fairness) usually elicit angry responses that it is precisely because they make players among others and activate unpleasant moral emotions override their material self-interest that emotional such as guilt or shame within the self. Anticipation of responses and their anticipations produce these these emotions acts a strong deterrent against norm socially desirable outcomes (Frank, 1988; Howard, violations, leading to more socially sustainable 1993). equilibriums. Similar findings emerge with other games. The Affective System of Behavioral Regulation For example, in public-good games, players have to and Judgment: Empirical Generalizations and decide how much to contribute to a collective good. Theoretical Conjectures The more people contribute, the better off the entire community is. However, non-contributing members This review suggests several empirical cannot be easily excluded from the benefits of the generalizations and associated theoretical conjectures public good. In these games, the materially rational about the affective system of behavioral regulation course of action is to “free-ride:” not contribute and judgment. anything and enjoy the common good supported by Negative emotional states of high intensity others. Of course, if every player behaves this way, generally interfere with people’s reasoning abilities. the system collapses and the entire community This is true of both high activation states such as suffers. Research suggests that a potent mechanism anxiety and low activation states such as intense for ensuring cooperation in public good games is to depression. The only reliable reasoning benefit of allow players to punish free-riders (Fehr & intense emotional states seems to be a greater Fischbacher, 2002). Public-good experiments show selective reliance on diagnostic information under that a majority of players are willing to incur personal states of high activation. These findings seem to financial costs to punish free-riders even in settings reflect an affective system of behavioral regulation, where players meet only once and where, therefore, which under high activation mobilizes responses there is no material incentive to punish—a based on quick assessments of diagnostic features of phenomenon called altruistic punishment (Fehr & the situation rather than more careful considerations Gachter, 2002). Again, it appears that the main driver of potential consequences of alternative courses of of these altruistic—and materially irrational— action. This conjecture is consistent with other punishments is the anger that punishers feel toward findings showing that judgments and decisions based the free-riders (de Quervain et al., 2004; Fehr & on integral affective responses are less resource- Gachter, 2002). Interestingly, when such punishment demanding and reached faster than those based on opportunities are present, cooperation increases over reason-based assessments. In contrast, the effects of time, whereas when they are absent, cooperation intensely pleasant emotional states (e.g., intense joy decreases over time. Therefore, as with the ultimatum or pride) are less well understood. If the proposed games, in public-good games, emotional responses to resource-mobilization conjecture is correct, intense norm-violating behavior again produce norm- positive emotional states might have less influence on enforcement behaviors that may appear materially reasoning processes because situations conducive of irrational in the short term but lead to more such states typically require less behavioral ecologically rational outcomes in the long run. adjustment. Emotional responses and their anticipation Milder incidental states of sadness generally appear to have similar effects in various other games promote a more systematic, data-driven, and such as the dictator game (a variant of the ultimatum analytical form of reasoning, whereas positive mood game), the classic prisoner’s dilemma game, and states generally promote a less systematic, more top- various “trust” games that involve costly investments down, but more flexible and creative form of in a partnership that may or may not be honored by reasoning. Whereas the effects of intense emotional the other party (Camerer, 2003; Dawes, McTavish, & states on reasoning may reflect the requirements of Shaklee, 1977; Fehr & Fischbacher, 2002; Lehmann, fast response mobilization, the effects of milder 2001). In many economic interactions, emotional sadness and positive moods seem to arise from a

Review of General Psychology, Vol. 11 (2), 155-158 14 Emotion and Rationality 15 signaling function of these milder states (N. Schwarz, alleviate an intense sleep deprivation, an emotionally 2002). From an evolutionary standpoint, positive upset person would likely consider a wider range of mood states may have served as a signal to engage in options in order to feel better. This suggests that the more contemplative thoughts and explorative affective system of behavioral regulation operates at behaviors. This conjecture is consistent with the a more abstract level than the various physiological finding that positive mood encourages variety- drive systems. seeking (Kahn & Isen, 1993) and the positive- While the evidence does not support a psychology hypothesis that a primary function of radical form of the somatic marker hypothesis, there positive affect is to broaden the thought-action is considerable support across disciplines for a more repertoire in order to build future resources benign version of this hypothesis, which is that (a) (Fredrickson, 1998). In contrast, because sadness integral emotional responses to objects are often typically highlights a discrepancy between a desired interpreted as signals of these objects’ value, (b) state and a current state (Higgins, 1987), states of these integral responses are very sensitive to recent sadness may signal a need to reassess the situation experiences with the target object. Besides being and analyze environmental inputs more carefully, typically faster and resource-efficient, affective which may explain occasional findings of depressive assessments of value differ from typical reason-based realism among individuals whose depression is not assessments in important respects. Affective too severe on tasks that are interpretable in terms of assessments of value tend to be more extreme and situational reassessment. Tasks that cannot be polarized and more sensitive to recent concrete interpreted in those terms may not be amenable to outcomes. This phenomenon may again reflect the depressive realism. response-mobilization function of emotions, which Incidental emotional states are often are meant to motivate and redirect behavior if misattributed to attentional objects and tend to distort necessary. While occasionally logically irrational, the beliefs about these objects in an assimilative fashion, greater polarization of affect-based valuations may especially if the emotional states are perceived to be thus have been evolutionary adaptive. representative of the objects. While logically Valuations based on integral affect tend to be irrational, this phenomenon may be rooted in an myopic, emphasizing immediate hedonic ecologically rational property of emotions. If an consequences (positive or negative) over future original function of emotion was to promote fast consequences. The affective system of valuation and responses to the environment, it would be behavioral regulation seems to be a system of the ecologically efficient to assume by default that one’s present. This anchoring in the present may have been emotional experiences are genuine responses to the ecologically rational in a world where the current focal objects, especially if these feeling experiences value of objects was generally positively correlated appear representative of these objects. The with their long-term value. However, in today’s assimilation of beliefs toward the content of emotions world, this anchoring in the present would produce may have had the purpose of promoting faster materially irrational valuations in domains where the behavioral responses by promoting intra-psychic correlation between short-term hedonic value and consistency (M. T. Pham, 2004). This is consistent long-term criterion value is negative. One of these with findings showing that integral affective domains is the investment domain, where the fear of responses to target objects also steer thoughts in the losses may inhibit investments in risky prospects direction of initial affective responses. such as stocks, even though stocks generally have Negative emotional states that are intense higher expected returns—a phenomenon known as generally disrupt self-control, but do not necessarily myopic loss aversion (Benartzi & Thaler, 1995). Still, increase risk seeking. This phenomenon again seems one should be careful not to over-generalize this case to reflect a response mobilization function of intense of material irrationality, for it is restricted to those negative emotional states. Like drive states such as domains where the correlation between short-term hunger, tiredness and sexual arousal, intense negative hedonic value and long-term criterion value is emotional states seem to direct behavior toward goal negative. One should bear in mind that in many objects that appear capable of alleviating these states. domains the correlation is likely positive. However, the goal objects being pursued under The above caveat applies more broadly to intense negative emotional states appear to be less alleged demonstrations of the material rationality or specific than those pursued under typical drive states. irrationality of integral affective responses as proxies While a person would likely forego food or sex to for value. For example, some studies suggest that

Review of General Psychology, Vol. 11 (2), 155-158 15 Emotion and Rationality 16 VMPC-related emotional deficits decrease primarily through object categorization, evolutionary performance (and economic welfare) in the Iowa pressures would have demanded that these Gambling Task while other findings suggest that the categorizations be consistent both within and across same deficits enhance performance in other individuals. For example, an ecological benefit of investment tasks. In any particular study, appearances consistent of affective valuation across individuals of rationality or irrationality will be driven by the would be improved social coordination. A society correlation between the emotional responses to the would not be stable if object-emotion mappings (e.g., target (conveyed for instance through outcome what is considered shameful) varied widely among its feedback) and its criterion value. One should not members. However, the ecologically-rational forget that this correlation is a parameter under the consistency of affective valuations comes at a cost. researcher’s control. A more pertinent question, As is evident in pathological phobias, affect- therefore, is whether integral affective responses are producing categorizations are difficult to unlearn, good predictors of value in a broader ecological which will produce instances of logical or material sense. As discussed by Pham (2004) using irrationality. Brunswick’s (1952) terminology, this would depend Valuations based on integral affect are also on the relative magnitude of two correlation more relativistic and reference-dependent. They are coefficients in the representative universe of objects very sensitive, for instance, to spontaneous to be evaluated: (a) the correlation between the comparisons with outcome counterfactuals and to integral feelings elicited by this universe of objects social comparisons. One conjecture is that the and these objects’ true criterion value (the ecological affective system of valuation is mostly ordinal as validity of the feelings) and (b) the correlation opposed to cardinal (i.e., interval scaled). This would between other available proxies of value and the be sensible from an evolutionary standpoint, if objects’ criterion value (the ecological validity of affective valuations were primarily meant to support alternative bases of evaluation). The ecological behavioral choices (do A, B, or C), which require merits of integral feelings as proxies of value would only ordinal judgments. The affective system may be additionally depend on the relative reliabilities of more concerned with the relative desirability ordering integral feelings and alternative inputs because, of alternative states of the world or alternative everything else equal, more reliable predictors make courses of actions than with their absolute better proxies. Unfortunately, empirical estimates of desirability. An ordinal affective system of valuation these correlations and reliabilities across broadly may also explain why people do not place much representative sets of targets have yet to be weight on the duration of hedonic experiences in documented. retrospective valuations these experiences Valuations based on integral affect also tend (Fredrickson & Kahneman, 1993). to be insensitive to scale, whether scale refers to the Finally, a major function of emotion appears quantitative magnitude of the stimulus or the to be the promotion of socially and morally desirable probability that surrounds it. Although it seems to behavior and the deterrence of undesirable behavior. violate logical rationality, this scale insensitivity may Emotions are a necessary component of empathic indicate that the affective system assigns value responses, which are important drivers of pro-social primarily through processes of categorization: target behavior. Emotions are also very sensitive to the objects or events are mapped onto existing categories fulfillment or violation of social and moral norms, or schemas and, depending on which category or and therefore an essential component of social and schema is activated, a particular value-laden moral self-regulation. Finally, in social and economic emotional response is triggered. In the affective interactions, emotions promote the fulfillment and system, the value of objects is thus determined by enforcement of social and moral obligations by their identity, not by their distribution. This overriding the players’ material self-interests. In this conjecture would explain why affective valuations sense, the emotional person is not irrational, but are particularly sensitive to concrete representations ecologically rational. Therefore, while the of exemplars. physiological drive systems (fatigue, pain, hunger, Affective valuations also tend to be more sexual arousal, etc.) may be inherently selfish consistent, both interpersonally and intra-personally. (Loewenstein, 1996), the affective system (love, This consistency may reflect a general reliability and pride, anger, guilt, shame, etc.) may be distinctively stability of integral feeling responses as a basis for social and moral. judgment. If the affective system assigns value

Review of General Psychology, Vol. 11 (2), 155-158 16 Emotion and Rationality 17 References

Abelson, R. P., Kinder, D. R., Peters, M. D., & Fiske, Bechara, A. (2004). The role of emotion in decision- S. T. (1982). Affective and Semantic making: Evidence from neurological patients Components in Political Person Perception. with orbitofrontal damage. Brain and Journal of Personality and Social Cognition, 55(1), 30-40. Psychology, 42(4), 619-630. Bechara, A., Damasio, A. R., Damasio, H., & Albarracin, D., & Kumkale, G. T. (2003). Affect as Anderson, S. W. (1994). Insensitivity to information in persuasion: A model of affect Future Consequences Following Damage to identification and discounting. Journal of Human Prefrontal Cortex. Cognition, 50(1- Personality and Social Psychology, 84(3), 3), 7-15. 453-469. Bechara, A., Damasio, H., Tranel, D., & Damasio, A. Allan, L. G., Siegel, S., & Hannah, S. ((in press)). R. (1997). Deciding advantageously before The Sad Truth About Depressive Realism. knowing the advantageous strategy. Science, Quarterly Journal of Experimental 275(5304), 1293-1295. Psychology. Benartzi, S., & Thaler, R. H. (1995). Myopic Loss Alloy, L. B., & Abramson, L. Y. (1979). Judgment of Aversion and the Equity Premium Puzzle. Contingency in Depressed and Nondepressed Quarterly Journal of Economics, 110(1), 73- Students - Sadder but Wiser. Journal of 92. Experimental Psychology-General, 108(4), Benassi, V. A., & Mahler, H. I. M. (1985). 441-485. Contingency Judgments by Depressed Alloy, L. B., & Abramson, L. Y. (1988). Depressive College-Students - Sadder but Not Always Realism: Four Theoretical Perspectives. In Wiser. Journal of Personality and Social L. B. Alloy (Ed.), Cognitive Processes in Psychology, 49(5), 1323-1329. Depression. New York: The Guilford Press. Blair, R. J. R. (1995). A Cognitive Developmental- Alloy, L. B., & Ahrens, A. H. (1987). Depression and Approach to Morality - Investigating the Pessimism for the Future - Biased Use of Psychopath. Cognition, 57(1), 1-29. Statistically Relevant Information in Blair, R. J. R. (1997). Moral reasoning and the child Predictions for Self Versus Others. Journal with psychopathic tendencies. Personality of Personality and Social Psychology, 52(2), and Individual Differences, 22(5), 731-739. 366-378. Blair, R. J. R., & Cipolotti, L. (2000). Impaired social Ariely, D., & Loewenstein, G. (2005). The heat of the response reversal - A case of 'acquired moment: the effect of sexual arousal on sociopathy'. Brain, 123, 1122-1141. sexual decision making. Journal of Bless, H., Bohner, G., Schwarz, N., & Strack, F. Behavioral Decision Making, 9999(9999), (1990). Mood and Persuasion - a Cognitive n/a. Response Analysis. Personality and Social Avnet, T., & Pham, M. T. (2004). Should I trust my Psychology Bulletin, 16(2), 331-345. feelings or not? The metacognition of affect- Bless, H., Mackie, D. M., & Schwarz, N. (1992). as-information.Unpublished manuscript, Mood Effects on Attitude Judgments - Columbia University. Independent Effects of Mood before and Bacon, S. J. (1974). Arousal and Range of Cue after Message Elaboration. Journal of Utilization. Journal of Experimental Personality and Social Psychology, 63(4), Psychology, 102(1), 81-87. 585-595. Bar-On, R., Tranel, D., Denburg, N. L., & Bechara, Bless, H., Schwarz, N., Clore, G. L., Golisano, V., & A. (2003). Exploring the neurological Rabe, C. (1996). Mood and the use of substrate of emotional and social scripts: Does a happy mood really lead to intelligence. Brain, 126, 1790-1800. mindlessness? Journal of Personality and Batra, R., & Stayman, D. M. (1990). The Role of Social Psychology, 71(4), 665-679. Mood in Advertising Effectiveness. Journal Bodenhausen, G. V. (1993). Emotions, Arousal, and of Consumer Research, 17(2), 203-214. Stereotype-based Discrimination: A Heuristic Model of Affect and Stereotyping. In D.M. Mackie & D. L. Hamilton, Affect,

Review of General Psychology, Vol. 11 (2), 155-158 17 Emotion and Rationality 18 cognition, and stereotyping (pp. 13-37). San discrete affect in consumer choice. Journal Diego, CA: Academic Press. of Consumer Research(December). Bodenhausen, G. V., Kramer, G. P., & Suesser, K. Darke, S. (1988a). Anxiety and Working Memory (1994). Happiness and stereotypic thinking Capacity. Cognition and Emotion, 2(2), 145- in social judgment. Journal of Personality & 154. Social Psychology, 66(4), 621-632. Darke, S. (1988b). Effects of Anxiety on Inferential Bodenhausen, G. V., Sheppard, L. A., & Kramer, G. Reasoning Task-Performance. Journal of P. (1994). Negative Affect and Social Personality and Social Psychology, 55(3), Judgment - the Differential Impact of Anger 499-505. and Sadness. European Journal of Social Dawes, R. M., McTavish, J., & Shaklee, H. (1977). Psychology, 24(1), 45-62. Behavior, Communication, and Assumptions Bouffard, J. A. (2002). The influence of emotion on About Other Peoples Behavior in a rational decision making in sexual Commons Dilemma Situation. Journal of aggression. Journal of Criminal Justice, Personality and Social Psychology, 35(1), 1- 30(2), 121-134. 11. Bower, G. H. (1981). Mood and Memory. American De Houwer, J., Thomas, S., & Baeyens, F. (2001). Psychologist, 36(2), 129-148. Associative learning of likes and dislikes: A Brandstatter, E., Kuhberger, A., & Schneider, F. review of 25 years of research on human (2002). A cognitive-emotional account of the evaluative conditioning. Psychological shape of the probability weighting function. Bulletin, 127(6), 853-869. Journal of Behavioral Decision Making, de Quervain, D. J. F., Fischbacher, U., Treyer, V., 15(2), 79-100. Schelthammer, M., Schnyder, U., Buck, A., Brunswick, E. (1952). The Conceptual Framework of et al. (2004). The neural basis of altruistic Psychology. Chicago: University of Chicago punishment. Science, 305(5688), 1254-1258. Press. Denes-Raj, V., & Epstein, S. (1994). Conflict Camerer, C. F. (2003). Behavioral Game Theory: between Intuitive and Rational Processing - Experiments on Strategic Interaction. When People Behave against Their Better Princeton: Princeton University Press. Judgment. Journal of Personality and Social Clore, G. L. (2005). For Love or Money: Some Psychology, 66(5), 819-829. Emotional Foundation of Rationality. DeSteno, D., Petty, R. E., Wegener, D. T., & Rucker, Chicago Kent Law Review, 80. D. D. (2000). Beyond valence in the Cohen, J. B., Pham, M. T., & Andrade, E. B. (in perception of likelihood: The role of emotion press). The Role of Affect In Consumer specificity. Journal of Personality and Social Judgment and Decision Making. In C. P. Psychology, 78(3), 397-416. Haugtvedt, P. M. Herr & F. R. Kardes Dobson, K., & Franche, R.-L. (1989). A Conceptual (Eds.), Handbook of Consumer Psychology. and Empirical Review of the Depressive Mahwah, NJ: Lawrence Erlbaum. Realism Hypothesis. Canadian Journal of Conway, M., & Giannopoulos, C. (1993). Dysphoria Behavioural Science-Revue Canadienne Des and Decision-Making - Limited Information Sciences Du Comportement, 21(4), 418-433. Use for Evaluations of Multiattribute Dommermuth, W. P., & Millard, W. J. (1967). Targets. Journal of Personality and Social Consumption Coincidence in Product Psychology, 64(4), 613-623. Evaluation. Journal of Marketing Research, Crepaz, N., & Marks, G. (2001). Are negative 4(4), 388-390. affective states associated with HIV sexual Dunning, D., & Story, A. L. (1991). Depression, risk behaviors? A meta-analytic review. Realism, and the Overconfidence Effect - Health Psychology, 20(4), 291-299. Are the Sadder Wiser When Predicting Damasio, A. R. (1994). Descartes' Error: Emotion, Future Actions and Events. Journal of Reason, and the Human Brain. New York: Personality and Social Psychology, 61(4), Putnam. 521-532. Darke, P. R., Chattopadhyay, A., & Ashworth, L. Dutton, D. G., & Aron, A. P. (1974). Some Evidence (2006). Going with your "gut feeling": the for Heightened Sexual Attraction under importance and functional significance of Conditions of High Anxiety. Journal of

Review of General Psychology, Vol. 11 (2), 155-158 18 Emotion and Rationality 19 Personality and Social Psychology, 30(4), Behavior (pp. 100-119). Toronto: C.J. 510-517. Hogrefe. Eisenberg, N. (2000). Emotion, regulation, and moral Forgas, J. P. (1995). Mood and judgement: The development. Annual Review of Psychology, affect infusion model (AIM). Psychological 51, 665-697. Bulletin, 117(1), 39. Eisenberg, N., Fabes, R. A., Murphy, B., Karbon, M., Forgas, J. P. (1998). On being happy and mistaken: Maszk, P., Smith, M., et al. (1994). The Mood effects on the fundamental attribution Relations of Emotionality and Regulation to error. Journal of Personality and Social Dispositional and Situational Empathy- Psychology, 75(2), 318-331. Related Responding. Journal of Personality Foster, C. A., Witcher, B. S., Campbell, W. K., & and Social Psychology, 66(4), 776-797. Green, J. D. (1998). Arousal and attraction: Eisenberg, N., & Miller, P. A. (1987). The Relation Evidence for automatic and controlled of Empathy to Prosocial and Related processes. Journal of Personality and Social Behaviors. Psychological Bulletin, 101(1), Psychology, 74(1), 86-101. 91-119. Frank, R. H. (1988). Passions Within Reason: The Epstein, S. (1990). Cognitive-Experiential Self- Strategic Role of the Emotions. New York: Theory. In L. A. Pervin (Ed.), Handbook of W.W. Norton & Company. Personality: Theory and Research (pp. 165- Fredrickson, B. L. (1998). What Good Are Positive 192). New York / London: The Guilford Emotions? Journal of General Psychology, Press. 2(3), 300-319. Etcoff, N. (1999). Survival of the Prettiest. New Fredrickson, B. L., & Kahneman, D. (1993). York: Doubleday. Duration Neglect in Retrospective Faber, R. J., & Christenson, G. A. (1996). In the Evaluations of Affective Episodes. Journal mood to buy: Differences in the mood states of Personality and Social Psychology, 65(1), experienced by compulsive buyers and other 45-55. consumers. Psychology & Marketing, 13(8), Frijda, N. H. (1988). The Laws of Emotion. 803-819. American Psychologist, 43(5), 349-358. Fehr, E., & Fischbacher, U. (2002). Why social Fry, P. S. (1975). Affect and Resistance to preferences matter - The impact of non- Temptation. Developmental Psychology, selfish motives on competition, cooperation 11(4), 466-472. and incentives. Economic Journal, 112(478), Goldberg, M. E., & Gorn, G. J. (1987). Happy and C1-C33. Sad Tv Programs - How They Affect Fehr, E., & Gachter, S. (2002). Altruistic punishment Reactions to Commercials. Journal of in humans. Nature, 415(6868), 137-140. Consumer Research, 14(3), 387-403. Fellows, L. K., & Farah, M. J. (2005). Different Gorn, G. J., Goldberg, M. E., & Basu, K. (1993). Underlying Impairments in Decision-making Mood, Awareness and Product Evaluation. Following Ventromedial and Dorsolateral Journal of Consumer Psychology, 2(3), 237- Frontal Lobe Damage in Humans. Cerebral 256. Cortex, 15(1), 58-63. Gorn, G. J., Pham, M. T., & Sin, L. Y. (2001). When Fessler, D. M. T., Pillsworth, E. G., & Flamson, T. J. arousal influences ad evaluation and valence (2004). Angry men and disgusted women: does not (and vice versa). Journal of An evolutionary approach to the influence of Consumer Psychology, 11(1), 43-55. emotions on risk taking. Organizational Gotlib, I. H., McLachlan, A. L., & Katz, A. N. Behavior and Human Decision Processes, (1988). Biases in Visual Attention in 95(1), 107-123. Depressed and Nondepressed Individuals. Fetherstonhaugh, D., Slovic, P., Johnson, S. M., & Cognition and Emotion, 2(3), 185-200. Friedrich, J. (1997). Insensitivity to the value Greene, J., & Haidt, J. (2002). How (and where) does of human life: A study of psychophysical moral judgment work? Trends in Cognitive numbing. Journal of Risk and Uncertainty, Sciences, 6(12), 517-523. 14(3), 283-300. Greene, J., Sommerville, R. B., Nystrom, L. E., Fiedler, K. (1988). Emotional Mood, Cognitive Style, Darley, J. M., & Cohen, J. D. (2001). An and Behavior Regulation. In K. Fiedler & J. fMRI investigation of emotional engagement Forgas (Eds.), Affect, Cognition and Social

Review of General Psychology, Vol. 11 (2), 155-158 19 Emotion and Rationality 20 in moral judgment. Science, 293(5537), Experimental Psychology-General, 133(1), 2105-2108. 23-30. Greene, T. R., & Noice, H. (1988). Influence of Hsee, C. K., Zhang, J., Yu, F., & Xi, Y. (2003). Lay Positive Affect Upon Creative-Thinking and rationalism and inconsistency between Problem-Solving in Children. Psychological predicted experience and decision. Journal Reports, 63(3), 895-898. of Behavioral Decision Making, 16(4), 257- Haidt, J. (2001). The emotional dog and its rational 272. tail: A social intuitionist approach to moral Humphreys, M. S., & Revelle, W. (1984). judgment. Psychological Review, 108(4), Personality, Motivation, and Performance - a 814-834. Theory of the Relationship between Hare, R. D. (1985). Comparison of Procedures for the Individual-Differences and Information- Assessment of Psychopathy. Journal of Processing. Psychological Review, 91(2), Consulting and Clinical Psychology, 53(1), 153-184. 7-16. Isen, A. M. (2001). An influence of positive affect on Hartlage, S., Alloy, L. B., Vazquez, C., & Dykman, decision making in complex situations: B. (1993). Automatic and Effortful Theoretical issues with practical Processing in Depression. Psychological implications. Journal of Consumer Bulletin, 113(2), 247-278. Psychology, 11(2), 75-85. Hendrickx, L., Vlek, C., & Oppewal, H. (1989). Isen, A. M., & Daubman, K. A. (1984). The Relative Importance of Scenario Information Influence of Affect on Categorization. and Frequency Information in the Judgment Journal of Personality and Social of Risk. Acta Psychologica, 72(1), 41-63. Psychology, 47(6), 1206-1217. Higgins, E. T. (1987). Self-Discrepancy - a Theory Isen, A. M., Daubman, K. A., & Nowicki, G. P. Relating Self and Affect. Psychological (1987). Positive Affect Facilitates Creative Review, 94(3), 319-340. Problem-Solving. Journal of Personality and Hirshleifer, D., & Shumway, T. (2003). Good day Social Psychology, 52(6), 1122-1131. sunshine: Stock returns and the weather. Isen, A. M., & Means, B. (1983). The Influence of Journal of Finance, 58(3), 1009-1032. Positive Affect on Decision-Making Hirt, E. R., Melton, R. J., McDonald, H. E., & Strategy. Social Cognition, 2(1), 18-31. Harackiewicz, J. M. (1996). Processing Isen, A. M., Niedenthal, P. M., & Cantor, N. (1992). goals, task interest, and the mood- An Influence of Positive Affect on Social performance relationship: A mediational Categorization. Motivation and Emotion, analysis. Journal of Personality and Social 16(1), 65-78. Psychology, 71(2), 245-261. Isen, A. M., & Shalker, T. E. (1982). The Effect of Hockey, G. R. J., Maule, A. J., Clough, P. J., & Feeling State on Evaluation of Positive, Bdzola, L. (2000). Effects of negative mood Neutral, and Negative Stimuli - When You states on risk in everyday decision making. Accentuate the Positive, Do You Eliminate Cognition & Emotion, 14(6), 823-855. the Negative. Social Psychology Quarterly, Hockey, R. (1973). Changes in Information-Selection 45(1), 58-63. Patterns in Multisource Monitoring as a Isen, A. M., Shalker, T. E., Clark, M., & Karp, L. Function of Induced Arousal Shifts. Journal (1978). Affect, Accessibility of Material in of Experimental Psychology, 101(1), 35-42. Memory, and Behavior - Cognitive Loop. Howard, N. (1993). The Role of Emotions in Journal of Personality and Social Multiorganizational Decision-Making. Psychology, 36(1), 1-12. Journal of the Operational Research Society, Johnson, E. J., Hershey, J., Meszaros, J., & 44(6), 613-623. Kunreuther, H. (1993). Framing, Probability Hsee, C. K., & Kunreuther, H. C. (2000). The Distortions, and Insurance Decisions. affection effect in insurance decisions. Journal of Risk and Uncertainty, 7(1), 35-51. Journal of Risk and Uncertainty, 20(2), 141- Johnson, E. J., & Tversky, A. (1983). Affect, 159. Generalization, and the Perception of Risk. Hsee, C. K., & Rottenstreich, Y. (2004). Music, Journal of Personality and Social pandas, and muggers: On the affective Psychology, 45(1), 20-31. psychology of value. Journal of

Review of General Psychology, Vol. 11 (2), 155-158 20 Emotion and Rationality 21 Kahn, B., & Isen, A. M. (1993). The Influence of Landman, J. (1987). Regret and elation following Positive Affect on Variety Seeking among action and inaction: Affective responses to Safe, Enjoyable Products. Journal of positive versus negative outcomes. Consumer Research, 20(2), 257-270. Personality and Social Psychology Bulletin, Kahneman, D. (1994). New Challenges to the 13(4), 524-536. Rationality Assumption. Journal of Langlois, J. H., Kalakanis, L., Rubenstein, A. J., Institutional and Theoretical Economics- Larson, A., Hallam, M., & Smoot, M. Zeitschrift Fur Die Gesamte (2000). Maxims or myths of beauty? A meta- Staatswissenschaft, 150(1), 18-36. analytic and theoretical review. Kahneman, D., & Miller, D. T. (1986). Norm Theory Psychological Bulletin, 126(3), 390-423. - Comparing Reality to Its Alternatives. Lazarus, R. S. (1991). Emotion and adaptation. New Psychological Review, 93(2), 136-153. York: Oxford University Press. Kahneman, D., Ritov, I., Jacowitz, K. E., & Grant, P. LeDoux, J. E. (1996). The Emotional Brain: The (1993). Stated Willingness-to-Pay for Mysterious Underpinnings of Emotional Public-Goods - a Psychological Perspective. Life. New York: Simon & Schuster. Psychological Science, 4(5), 310-315. Lee, L., & Ariely, D. (2006). In Search of Homo Kahneman, D., Ritov, I., & Schkade, D. (1999). Economicus: Preference Consistency, Economic preferences or attitude Emotions, and Cognition: Columbia expressions? An analysis of dollar responses University. to public issues. Journal of Risk and Lehmann, D. R. (2001). The impact of altruism and Uncertainty, 19(1-3), 203-235. envy on competitive behavior and Kahneman, D., Schkade, D., & Sunstein, C. R. satisfaction. International Journal of (1998). Shared outrage and erratic awards: Research in Marketing, 18(1-2), 5-17. The psychology of punitive damages. Leith, K. P., & Baumeister, R. F. (1996). Why do bad Journal of Risk and Uncertainty, 16(1), 49- moods increase self-defeating behavior? 86. Emotion, risk taking, and self-regulation. Kamstra, M. J., Kramer, L. A., & Levi, M. D. (2003). Journal of Personality and Social Winter blues: A SAD stock market cycle. Psychology, 71(6), 1250-1267. American Economic Review, 93(1), 324-343. Leland, J. W., & Grafman, J. (2005). Experimental Keinan, G. (1987). Decision-Making under Stress - tests of the Somatic Marker hypothesis. Scanning of Alternatives under Controllable Games and Economic Behavior, 52(2), 386- and Uncontrollable Threats. Journal of 409. Personality and Social Psychology, 52(3), Leon, M. R., & Revelle, W. (1985). Effects of 639-644. Anxiety on Analogical Reasoning - a Test of Keller, P. A., Lipkus, I. M., & Rimer, B. K. (2002). 3 Theoretical-Models. Journal of Personality Depressive realism and health risk accuracy: and Social Psychology, 49(5), 1302-1315. The negative consequences of positive Lerner, J. S., & Keltner, D. (2001). Fear, anger, and mood. Journal of Consumer Research, 29(1), risk. Journal of Personality & Social 57-69. Psychology, 81(1), 146-159. Kirkpatrick, L. A., & Epstein, S. (1992). Cognitive Lerner, J. S., Small, D. A., & Loewenstein, G. Experiential Self-Theory and Subjective- (2004). Heart strings and purse strings - Probability - Further Evidence for 2 Carryover effects of emotions on economic Conceptual Systems. Journal of Personality decisions. Psychological Science, 15(5), and Social Psychology, 63(4), 534-544. 337-341. Knapp, A., & Clark, M. S. (1991). Some Detrimental Lewinsohn, P. M., Mischel, W., Chaplin, W., & Effects of Negative Mood on Individuals Barton, R. (1980). Social Competence and Ability to Solve Resource Dilemmas. Depression - the Role of Illusory Self- Personality and Social Psychology Bulletin, Perceptions. Journal of Abnormal 17(6), 678-688. Psychology, 89(2), 203-212. Kruglanski, A. W. (1989). The psychology of being Loewenstein, G. (1996). Out of control: Visceral "right": The problem of accuracy in social influences on behavior. Organizational perception and cognition. Psychological Behavior and Human Decision Processes, Bulletin, 106(3), 395-409. 65(3), 272-292.

Review of General Psychology, Vol. 11 (2), 155-158 21 Emotion and Rationality 22 Loewenstein, G., Weber, E. U., Hsee, C. K., & reactions to the outcomes of risky options. Welch, N. (2001). Risk as feelings. Psychological Science, 8(6), 423-429. Psychological Bulletin, 127(2), 267-286. Miniard, P. W., Bhatla, S., & Sirdeshmukh, D. Lorber, M. F. (2004). Psychophysiology of (1992). Mood as a Determinant of aggression, psychopathy, and conduct Postconsumption Product Evaluations. problems: A meta-analysis. Psychological Journal of Consumer Psychology, 1(2), 173- Bulletin, 130(4), 531-552. 195. Losel, F., & Schmucker, M. (2004). Psychopathy, Mitchell, D. G. V., Colledge, E., Leonard, A., & risk taking, and attention: A differentiated Blair, R. J. R. (2002). Risky decisions and test of the somatic marker hypothesis. response reversal: is there evidence of Journal of Abnormal Psychology, 113(4), orbitofrontal cortex dysfunction in 522-529. psychopathic individuals? Mackie, D. M., & Worth, L. T. (1989). Processing Neuropsychologia, 40(12), 2013-2022. Deficits and the Mediation of Positive Affect Moll, J., de Oliveira-Souza, R., Eslinger, P. J., in Persuasion. Journal of Personality and Bramati, I. E., Mourao-Miranda, J., Social Psychology, 57(1), 27-40. Andreiuolo, P. A., et al. (2002). The neural Maia, T. V., & McClelland, J. L. (2004). A correlates of moral sensitivity: A functional reexamination of the evidence for the magnetic resonance imaging investigation of somatic marker hypothesis: What basic and moral emotions. Journal of participants really know in the Iowa Neuroscience, 22(7), 2730-2736. gambling task. Proceedings of the National Moll, J., de Oliveira-Souza, R., Moll, F. T., Ignacio, Academy of Sciences of the United States of F. A., Bramati, I. E., Caparelli-Daquer, E. America, 101(45), 16075-16080. M., et al. (2005). The moral affiliations of Mano, H. (1992). Judgments under Distress - disgust - A functional MRI study. Cognitive Assessing the Role of Unpleasantness and and Behavioral Neurology, 18(1), 68-78. Arousal in Judgment Formation. Monat, A., Averill, J. R., & Lazarus, R. S. (1972). Organizational Behavior and Human Anticipatory Stress and Coping Reactions Decision Processes, 52(2), 216-245. under Various Conditions of Uncertainty. Mano, H. (1994). Risk-Taking, Framing Effects, and Journal of Personality and Social Affect. Organizational Behavior and Human Psychology, 24(2), 237-253. Decision Processes, 57(1), 38-58. Msetfi, R. M., Murphy, R. A., Simpson, J., & Martin, D. J., Abramson, L. Y., & Alloy, L. B. Kornbrot, D. E. (2005). Depressive realism (1984). Illusion of Control for Self and and outcome density bias in contingency Others in Depressed and Nondepressed judgments: The effect of the context and College-Students. Journal of Personality and intertrial interval. Journal of Experimental Social Psychology, 46(1), 125-136. Psychology-General, 134(1), 10-22. Mattes, J., & Cantor, J. (1982). Enhancing Responses Mueller, J. H. (1977). Test Anxiety, Input Modality, to Television Advertisements via the and Levels of Organization in Free-Recall. Transfer of Residual Arousal from Prior Bulletin of the Psychonomic Society, 9(1), Programming. Journal of Broadcasting, 67-69. 26(4), 553-566. Mueller, J. H. (1978). Effects of Individual- Mayer, J. D., Gaschke, Y. N., Braverman, D. L., & Differences in Test Anxiety and Type of Evans, T. W. (1992). Mood-Congruent Orienting Task on Levels of Organization in Judgment Is a General Effect. Journal of Free-Recall. Journal of Research in Personality and Social Psychology, 63(1), Personality, 12(1), 100-116. 119-132. Neufeldt, V. (1991). Webster's New World Medvec, V. H., Madey, S. F., & Gilovich, T. (1995). Dictionary (3rd ed.). New York: Prentice When Less Is More - Counterfactual Hall. Thinking and Satisfaction among Olympic Nisbett, R., & Ross, L. (1980). Human Inference: Medalists. Journal of Personality and Social Strategies and Shortcomings of Social Psychology, 69(4), 603-610. Judgment. Englewood Cliffs, NJ: Prentice- Mellers, B. A., Schwartz, A., Ho, K., & Ritov, I. Hall. (1997). Decision affect theory: Emotional

Review of General Psychology, Vol. 11 (2), 155-158 22 Emotion and Rationality 23 Nowlis, S. M., & Shiv, B. (2005). The influence of Raghunathan, R., & Pham, M. T. (1999). All negative consumer distractions on the effectiveness of moods are not equal: Motivational influences food-sampling programs. Journal of of anxiety and sadness on decision making. Marketing Research, 42(2), 157-168. Organizational Behavior and Human Oaksford, M., Morris, F., Grainger, B., & Williams, Decision Processes, 79(1), 56-77. J. M. G. (1996). Mood, reasoning, and Raghunathan, R., Pham, M. T., & Corfman, K. P. central executive processes. Journal of (2006). Informational Properties of Anxiety Experimental Psychology-Learning Memory and Sadness, and Displaced Coping. Journal and Cognition, 22(2), 476-492. of Consumer Research, 32(4). Ortony, A., Clore, G. L., & Collins, A. (1988). The Ratner, R. K., & Herbst, K. C. (2005). When good Cognitive Structure of Emotions. Cambridge: decisions have bad outcomes: The impact of Cambridge University Press affect on switching behavior. Organizational Ottati, V. C., & Isbell, L. M. (1996). Effects of mood Behavior and Human Decision Processes, during exposure to target information on 96(1), 23-37. subsequently reported judgments: An on-line Roese, N. J. (1997). Counterfactual thinking. model of misattribution and correction. Psychological Bulletin, 121(1), 133-148. Journal of Personality and Social Roth, A. E. (1995). Introduction to Experimental Psychology, 71(1), 39-53. Economics. In J. H. Kagel & A. E. Roth Pacini, R., Muir, F., & Epstein, S. (1998). Depressive (Eds.), The Handbook of Experimental realism from the perspective of cognitive- Economics (pp. 3-109). Princeton: Princeton experiential self-theory. Journal of University Press. Personality and Social Psychology, 74(4), Rottenstreich, Y., & Hsee, C. K. (2001). Money, 1056-1068. kisses, and electric shocks: On the affective Patrick, C. J. (1994). Emotion and Psychopathy - psychology of risk. Psychological Science, Startling New Insights. Psychophysiology, 12(3), 185-190. 31(4), 319-330. Ruder, M., & Bless, H. (2003). Mood and the Peretz, I., Gagnon, L., & Bouchard, B. (1998). Music reliance on the ease of retrieval heuristic. and emotion: perceptual determinants, Journal of Personality and Social immediacy, and isolation after brain damage. Psychology, 85(1), 20-32. Cognition, 68(2), 111-141. Ruderman, A. J. (1986). Dietary Restraint - a Pham, M. T. (1996). Cue representation and selection Theoretical and Empirical Review. effects of arousal on persuasion. Journal of Psychological Bulletin, 99(2), 247-262. Consumer Research, 22(4), 373-387. Sanbonmatsu, D. M., & Kardes, F. R. (1988). The Pham, M. T. (1998). Representativeness, relevance, Effects of Physiological Arousal on and the use of feelings in decision making. Information-Processing and Persuasion. Journal of Consumer Research, 25(2), 144- Journal of Consumer Research, 15(3), 379- 159. 385. Pham, M. T. (2004). The Logic of Feeling. Journal of Sanfey, A. G., Rilling, J. K., Aronson, J. A., Consumer Psychology, 14(4). Nystrom, L. E., & Cohen, J. D. (2003). The Pham, M. T., Cohen, J. B., Pracejus, J. W., & neural basis of economic decision-making in Hughes, G. D. (2001). Affect monitoring and the ultimatum game. Science, 300(5626), the primacy of feelings in judgment. Journal 1755-1758. of Consumer Research, 28(2), 167-188. Saunders, E. M. J. (1993). Stock-Prices and Wall- Pham, T. H., Philippot, P., & Rime, B. (2000). Street Weather. American Economic Review, Subjective and autonomic responses to 83(5), 1337-1345. emotion induction in psychopaths. Schmitt, W. A., Brinkley, C. A., & Newman, J. P. Encéphale, 26(1), 45-51. (1999). Testing Damasio's somatic marker Pillutla, M. M., & Murnighan, J. K. (1996). hypothesis with psychopathic individuals: Unfairness, anger, and spite: Emotional Risk takers or risk averse? Journal of rejections of ultimatum offers. Abnormal Psychology, 108(3), 538-543. Organizational Behavior and Human Schnall, S., Haidt, J., & Clore, G. L. (2006). Disgust Decision Processes, 68(3), 208-224. as Embodied Moral Judgment.Unpublished manuscript, University of Virginia.

Review of General Psychology, Vol. 11 (2), 155-158 23 Emotion and Rationality 24 Schwarz, J. C., & Pollack, P. R. (1977). Affect and Logic and Strategy in an Abstract Reasoning Delay of Gratification. Journal of Research Task. Archives of General Psychiatry, 40(7), in Personality, 11(2), 147-164. 775-780. Schwarz, N. (1990). Feelings as information: Sinaceur, M., Heath, C., & Cole, S. (2005). Informational and motivational functions of Emotional and deliberative reactions to a affective states. In R. M. Sorrentino & E. T. public crisis - Mad Cow disease in France. Higgins (Eds.), Handbook of motivation and Psychological Science, 16(3), 247-254. cognition (Vol. 2, pp. 521-561). New York: Sinclair, R. C. (1988). Mood, Categorization Breadth, Guilford Press, c1986-. and Performance-Appraisal - the Effects of Schwarz, N. (2002). Situated cognition and the Order of Information Acquisition and wisdom of feelings: Cognitive tuning. In L. Affective State on Halo, Accuracy, F. Barrett & P. Salovey (Eds.), The wisdom Information-Retrieval, and Evaluations. in feelings: Psychological processes in Organizational Behavior and Human emotional intelligence (pp. 144-166). New Decision Processes, 42(1), 22-46. York: The Guilford Pres. Sinclair, R. C., & Mark, M. M. (1995). The Effects of Schwarz, N., & Clore, G. L. (1983). Mood, Mood State on Judgmental Accuracy - Misattribution, and Judgments of Well- Processing Strategy as a Mechanism. Being - Informative and Directive Functions Cognition & Emotion, 9(5), 417-438. of Affective States. Journal of Personality Sinclair, R. C., Mark, M. M., & Clore, G. L. (1994). and Social Psychology, 45(3), 513-523. Mood-Related Persuasion Depends on Schwarz, N., & Clore, G. L. (1996). Feelings and (Mis)Attributions. Social Cognition, 12(4), Phenomenal Experiences. In E. T. Higgins & 309-326. A. W. Kruglanski (Eds.), Social Psychology: Slovic, P., Finucane, M., Peters, E., & MacGregor, D. Handbook of Basic Principles (pp. 433-465). G. (2002). The Affect Heuristic. In T. New York: The Guilford Press. Gilovich, D. Griffin & D. Kahneman (Eds.), Seeman, G., & Schwarz, J. C. (1974). Affective State Heuristics and biases: The psychology of and Preference for Immediate Versus intuitive judgment (pp. 397-420). New York: Delayed Reward. Journal of Research in Cambridge University Press. Personality, 7(4), 384-394. Strack, F. (1992). The Different Routes to Social Sen, A. K. (1990). Rational Behaviour. In J. Eatwell, Judgements: Experiential versus M. M. & N. P (Eds.), The New Palgrave: Informational Strategies. In L. L. M. A. Utility and Probability (pp. 198-216). New Tesser (Ed.), The Construction of Social York-London: W.W. Norton & Company. Judgements (pp. 249-275). Hillsdale, N.J.: Shiffman, S., & Waters, A. J. (2004). Negative affect Lawrence Erlbawm Associates. and smoking lapses: A prospective analysis. Strongman, K. T. (1987). The Psychology of Emotion Journal of Consulting and Clinical (3rd ed.). Chichester: John Wiley & Sons. Psychology, 72(2), 192-201. Sunstein, C. R. (2003). Terrorism and probability Shiv, B., & Fedorikhin, A. (1999). Heart and mind in neglect. Journal of Risk and Uncertainty, conflict: The interplay of affect and 26(2-3), 121-136. cognition in consumer decision making. Tice, D. M., Bratslavsky, E., & Baumeister, R. F. Journal of Consumer Research, 26(3), 278- (2001). Emotional distress regulation takes 292. precedence over impulse control: If you feel Shiv, B., Loewenstein, G., Bechara, A., Damasio, H., bad, do it! Journal of Personality and Social & Damasio, A. R. (2005). Investment Psychology, 80(1), 53-67. behavior and the negative side of emotion. Tiedens, L. Z., & Linton, S. (2001). Judgment under Psychological Science, 16(6), 435-439. emotional certainty and uncertainty: The Siemer, M., & Reisenzein, R. (1998). Effects of effects of specific emotions on information mood on evaluative judgements: Influence of processing. Journal of Personality and reduced processing capacity and mood Social Psychology, 81(6), 973-988. salience. Cognition & Emotion, 12(6), 783- Trafimow, D., Bromgard, I. K., Finlay, K. A., & 805. Ketelaar, T. (2005). The role of affect in Silberman, E. K., Weingartner, H., & Post, R. M. determining the attributional weight of (1983). Thinking Disorder in Depression -

Review of General Psychology, Vol. 11 (2), 155-158 24 Emotion and Rationality 25 immoral behaviors. Personality and Social Psychology Bulletin, 31(7), 935-948. Tversky, A., & Griffin, D. (1991). Endowment and Contrast in Judgments of Well-Being. In F. Strack, M. Argyle & N. Schwarz (Eds.), Subjective Well-Being: An Interdisciplinary Perspective (pp. 101-118). Oxford: Pergamon Press. Vandijk, E., & Zeelenberg, M. (2006). The Dampening Effect of Uncertainty on Positive and Negative Emotions. Journal of Behavioral Decision Making, 19, 171-176. Verplanken, B., Hofstee, G., & Janssen, H. J. W. (1998). Accessibility of effective versus cognitive components of attitudes. European Journal of Social Psychology, 28(1), 23-35. Wheatley, T., & Haidt, J. (2005). Hypnotic disgust makes moral judgments more severe. Psychological Science, 16(10), 780-784. White, G. L., Fishbein, S., & Rutsein, J. (1981). Passionate love and the misattribution of arousal. Journal of Personality & Social Psychology, 41(1), 56-62. Wilson, T. D., Lisle, D. J., Schooler, J. W., Hodges, S. D., & et al. (1993). Introspecting about reasons can reduce post-choice satisfaction. Personality & Social Psychology Bulletin, 19(3), 331-339. Worth, L. T., & Mackie, D. M. (1987). Cognitive Mediation of Positive Affect in Persuasion. Social Cognition, 5(1), 76-94. Wright, W. F., & Bower, G. H. (1992). Mood Effects on Subjective-Probability Assessment. Organizational Behavior and Human Decision Processes, 52(2), 276-291. Yeung, C. W. M., & Wyer, R. S. (2004). Affect, appraisal, and consumer judgment. Journal of Consumer Research, 31(2), 412-424. Zajonc, R. B. (1980). Feeling and Thinking - Preferences Need No Inferences. American Psychologist, 35(2), 151-175. Zillmann, D. (1971). Excitation Transfer in Communication-Mediated Aggressive Behavior. Journal of Experimental Social Psychology, 7(4), 419-&.

Review of General Psychology, Vol. 11 (2), 155-158 25