P1: GFZ 0521824176c13.xml CB798B/Holyoak 0521824176 October 31, 2004 6:5

CHAPTER 13 Motivated Thinking

Daniel C. Molden E. Tory Higgins

At one time or another, every one of us has judgment. More recently however, there has engaged in “wishful thinking,” or “let our been a sharp increase in attempts to achieve hearts influence our heads.” That is, every a marriage between these two viewpoints in one of us has felt the effects of our motiva- a wide variety of research areas. The pri- tions on our thought processes. Given this mary objective of this chapter is to review common everyday experience, it is not sur- these attempts and to demonstrate how it prising that an essential part of early psy- is not only possible, but also desirable, to chological research was the idea that drives, reintroduce motivational approaches to the needs, desires, motives, and goals can pro- study of basic thought processes. We begin foundly influence judgment and reasoning. by providing some historical background on More surprising is that motivational vari- such approaches. ables play only a small role in current the- ories of reasoning. Why might this be? One possible explanation is that since the cognitive revolution in the 1960s and 1970s, A Brief History of Motivated Thinking researchers studying motivational and cogni- tive processes have been speaking somewhat Motivational perspectives on thought and different languages. That is, there has been reasoning originated most prominently with a general failure to connect traditional moti- Freud’s (1905) clinical theorizing on the vational concepts, such as drives or motives, psychodynamic conflicts created by uncon- to information processing concepts, such as scious drives and urges. These perspectives expectancies or spreading activation, which quickly spread to other areas of psychology. form the foundation for nearly all contem- Early pioneers of experimental social psy- porary research on thinking and reasoning. chology gave primary emphasis to motiva- Foraperiod of time, this led not only to mis- tional variables such as drives, goals, and as- understanding, but also to conflict between pirations (e.g., Allport, 1920; Lewin, 1935). motivational and cognitive perspectives on The study of personality came to involve the 295 P1: GFZ 0521824176c13.xml CB798B/Holyoak 0521824176 October 31, 2004 6:5

296 the cambridge handbook of thinking and reasoning

identification and classification of different This chapter provides an overview of this types of needs and motives (e.g., Murray, “second generation” of research on moti- 1938). Even research on sensory and percep- vated thinking and discusses some of the tual processes was influenced by a motiva- larger principles that have emerged from the tional approach with the emergence of the study of the motivation/cognition interface. “New Look” school (e.g., McGinnies, 1949). We consider two general classes of motiva- After this early period of growth and tional influences; the first involves people’s expansion, however, research and theory desires for reaching certain types of outcomes on motivated thinking became quite con- in their judgments, and the second involves troversial. With the ascendance of cogni- people’s desires to use certain types of strate- tive perspectives on thinking and reason- gies while forming their judgments. In so do- ing in the 1960s and 1970s, many supposed ing, we adopt a rather broad focus and dis- instances of motivated reasoning were re- cuss several different varieties of motivated cast as merely a product of imperfect infor- thinking. Given space constraints, this broad mation processing by imperfect perceivers focus necessitates being selective in the phe- (compare Bruner, 1957, with McGinnies, nomena to be described. We have chosen 1949;Festinger, 1957, with Bem, 1967; those programs of research that we believe Bradley, 1978, with Nisbett & Ross, 1980). are representative of the larger literature and The various “motivation versus cognition” are especially relevant not only to the study debates that subsequently developed con- of reasoning, but also to other areas in cog- tinued off and on for years before they nitive psychology.1 After reviewing the sep- were declared not only unwinnable, but also arate influences of outcome- and strategy- counterproductive. An uneasy armistice was based motivations on thinking, we conclude declared (Tetlock & Levi, 1982) that ef- by suggesting potential directions for future fectively quieted the public conflict, but research, giving special attention to circum- did nothing to reconcile the deep con- stances where multiple sources of motiva- ceptual differences that still remained be- tion might operate simultaneously. tween researchers favoring cognitive or motivational perspectives. Following this period of conflict, enthu- siasm for questions concerning motivational Outcome-Motivated Thinking influences on thinking was dampened in the 1970s and early 1980s. Beginning in the late The most prominent approach to motivated 1980s, however, there was a resurgence of reasoning, in both classic and contemporary interest in this area (for recent reviews and perspectives, has been to examine the influ- overviews, see, Dunning, 1999; Gollwitzer ence on people’s thought processes of their & Bargh, 1996; Higgins & Molden, 2003; needs, preferences, and goals to reach desired Kruglanski, 1996;Kunda, 1990; Sorrentino outcomes (or avoid undesired outcomes). & Higgins, 1986). One reason for this new Although the types of preferred outcomes life is that instead of revisiting debates about that have been studied are highly diverse, the workings of motivational versus cog- they can be divided into two general classes: nitive processes, researchers began to ex- directional outcomes and nondirectional out- amine the important interactions between comes (see Kruglanski, 1996;Kunda, 1990). these two processes. Thus, more recent Individuals who are motivated by directional investigations have focused on the iden- outcomes are interested in reaching specific tification of principles that describe the desired conclusions, such as impressions of interface between motivation and cogni- themselves as intelligent, caring, and worthy tion, and the implications of this interface people (e.g., Dunning, 1999; Pyszczynski & for thinking, reasoning, and judgment (see Greenberg, 1987), or positive beliefs about Kruglanski, 1996;Kunda, 1990; Higgins & others whom they find likeable or to whom Molden, 2003). they are especially close (e.g., Murray, 1999). P1: GFZ 0521824176c13.xml CB798B/Holyoak 0521824176 October 31, 2004 6:5

motivated thinking 297

In contrast, individuals who are motivated different cognitive processes, including attri- by nondirectional outcomes have more gen- bution, evaluation of evidence, information eral concerns, such as reaching the most search, recall and knowledge activation, and accurate conclusion possible (e.g., Fiske & the organization of concepts in memory. Neuberg, 1990)ormaking a clear and con- cise decision (e.g., Kruglanski & Webster, 1996), whatever this conclusion or decision effects on attribution may be. Some of the first evidence for the effects Whether outcome motivation is direc- on reasoning of motivations for positive self- tional or nondirectional, however, this moti- evaluation grew out of work on attribution vation has been conceptualized as affecting (see Kelley, 1973). Early attributional re- thought and reasoning in the same way: by search found that when people were ex- directing people’s cognitive processes (e.g., plaining their performance on tasks measur- their recall, information search, or attribu- ing important abilities, they tended to take tions) in ways that help to ensure they reach responsibility for their success (i.e., cite in- their desired conclusions. That is, individu- ternal and stable causes, such as “I’m good als’ preferences for certain outcomes are be- at this task.”) and to deny responsibility for lieved to often shape their thinking so as to their failure (i.e., cite external and unstable all but guarantee that they find a way to be- causes, such as “I was unlucky.”). Such find- lieve, decide, and justify whatever they like. ings were typically described as stemming In this chapter, we review several programs from desires for positive beliefs about the of research that have more closely exam- self (for a review, see Bradley, 1978). ined the specific mechanisms by which this The motivational nature of these find- can occur, first in relation to motivations for ings was questioned, however. Several re- directional outcomes and then in relation searchers (e.g., Nisbett & Ross, 1980)ar- to motivations for nondirectional outcomes. gued that although one’s attributions may Following this, we discuss several limitations sometimes be biased, this does not neces- of the effects of outcome motivation on rea- sarily imply that motivational forces are at soning and identify circumstances in which work (e.g., previous expectancies for success these motivations are most likely to have could lead people to label an unexpected an impact. failure as unusual or unlucky). Yet, subse- quent research has found that, although peo- ple’s expectancies do play a role in these Influences of Directional attributional effects, there is substantial ev- Outcome Motivation idence that motivation plays an important Overall, the kinds of phenomena that have role as well (see Kunda, 1990; Pyszczynski been studied most extensively in research on & Greenberg, 1987). motivated thinking involve directional out- One type of evidence for the role of come preferences (i.e., individuals’ desires to motivation in self-serving attributions is reach specific conclusions about themselves that, independent of expectancies from prior and others; for reviews, see Dunning, 1999; success or failure, the more personally im- Kunda, 1990; Murray, 1999; Pyszczynski & portant a success is in any given situation, Greenberg, 1987). Although a variety of out- the stronger is the tendency to claim respon- comes have been investigated, people’s well- sibility for this success but to deny responsi- documented preference for viewing them- bility for failure (Miller, 1976). Another type selves, and those close to them, in a generally of evidence is that people’s attributions be- positive manner (see Baumeister, 1998) has, come increasingly self-serving when success by far, received the most attention. This out- or failure feedback is experienced as highly come is the primary focus here.2 In the next arousing. For instance, Gollwitzer, Earle, and sections, we review several effects of desires Stephan (1982) had participants first com- for positive self-evaluation involving many plete an intelligence test, then vigorously P1: GFZ 0521824176c13.xml CB798B/Holyoak 0521824176 October 31, 2004 6:5

298 the cambridge handbook of thinking and reasoning

ride a stationary bicycle while the test was positive rate). Those who believed TAA being scored (increasing their arousal), and had positive health consequences, however, finally, receive feedback about succeeding judged the test to be highly valid regardless or failing on the test. Feedback was given of its reliability (see also Doosje, Spears, & 1 minute, 5 minutes, or 9 minutes after riding Koomen, 1995). the bicycle. Both those receiving feedback An example of the second type of in- after 9 minutes, who were no longer aroused, fluence can be found in a study by Kunda and those receiving feedback after 1 minute, (1987). Participants read a scientific article who were aroused but still associated this reporting that caffeine consumption was re- arousal with the exercise, showed only lated to serious health problems in women. small attributional differences following suc- Afterward, women (but not men) who were cess versus failure feedback. In contrast, heavy caffeine consumers reported that the those receiving feedback after 5 minutes, article was less convincing than women who were still aroused but no longer asso- who were light caffeine consumers. In a ciated this with the exercise, misattributed follow-up study where people read a simi- their arousal to the feedback concerning lar article that revealed caffeine caused only the test and showed a strong tendency to mild health problems, there was no rela- credit their ability for success and blame tion between their evaluation of the ar- bad luck for failure (see also Stephan & ticle and their caffeine consumption. Be- Gollwitzer, 1981). cause, in both studies, people’s reasoning was altered only when there was a signif- icant threat to the self, this demonstrates effects on evidence evaluation the motivational nature of these results (see Similar to these attribution effects, more re- also Beauregard & Dunning, 1998; Ditto cent research has found that motivations for et al., 1998). positive self-evaluations also influence the Similar effects of people’s desire to view wayinwhich people evaluate information themselves positively have also been demon- that either supports or contradicts these pos- strated in domains that do not directly in- itive self-evaluations. In general, individuals volve health consequences. For instance, tend to (1) give more credence to, and be people who encounter scientific research more optimistic about, the validity of infor- that appears to support their cherished at- mation that supports or confirms their stand- titudes describe this research as being bet- ing as kind, competent, and healthy people; ter conducted, and its conclusions as being and (2)bemore skeptical and cautious about more valid, than those who encounter the information that threatens this standing. same research, but believe it to be in conflict An example of the first type of influence with their cherished attitudes (e.g., Lord, can be found in a study by Ditto, Scepansky, Ross, & Lepper, 1979). In addition, people Munro, Apanovitch, and Lockhart (1998). have been shown to engage in considerable Individuals were “tested” for the presence of counterfactual thinking (i.e., mentally un- a fictitious enzyme in the body, TAA, and doing the present state of affairs by imag- everyone was told that they had tested pos- ining “if only ...”; see Roese, 1997) when itive. Half of the people were informed that evidence supporting predictions from a pre- this had positive health consequences, and ferred theory or worldview fails to materi- half were informed that this had negative alize. Such counterfactual thinking allows consequences. Those who believed TAA had them to generate ways in which they were negative health consequences were largely almost correct. However, when evidence is dismissive of the test when told it was consistent with their theories, these same slightly unreliable (i.e., had a 10% false- individuals do not engage in counterfactual positive rate) and judged the result to be thinking, which would force them to gener- only somewhat more valid when told the test ate ways in which they were almost wrong was highly reliable (i.e., had a .05% false- (Tetlock, 1998). P1: GFZ 0521824176c13.xml CB798B/Holyoak 0521824176 October 31, 2004 6:5

motivated thinking 299

effects on information search sires for positive views of themselves (and The motivational influences discussed thus certain well-liked others) have been also far center on the quality of people’s in- found to influence their use of stored knowl- formation processing during reasoning (e.g., edge in memory, such as the selective ac- biased attributions, more or less critical tivation of concepts and recall of events evaluations). However, desires for positive that support these views. This phenom- self-evaluations also affect the quantity of ena is exemplified in a series of studies by 1990 people’s information processing (Kruglan- Santioso, Kunda, and Fong ( ). Partici- ski, 1996). Specifically, such desires moti- pants in these studies read fictitious articles vate decreased processing and quick accep- revealing that either introverts or extroverts tance of favorable evidence, and increased tend to have more academic and professional processing and hesitant acceptance of unfa- success. Following this, individuals who be- vorable evidence. As one example, Ditto and lieved that introversion was linked to success colleagues (Ditto & Lopez, 1992; Ditto et al., were more likely to recall, and were faster 1998) demonstrated that, compared with to recall, autobiographical instances of intro- evaluating favorable evidence, when people verted behaviors than extroverted behaviors. evaluate unfavorable evidence they spend a The opposite pattern of results was found for greater amount of time examining this evi- individuals who believed that extroversion dence and spontaneously generate more al- was linked to success. ternate hypotheses about why it might be More recent work has demonstrated that, unreliable (see also Pyszczynski & Green- in addition to creating selective recall, direc- berg, 1987). Moreover, they have also shown tional outcome motivation can also lead to that individuals who are prevented from the reconstruction of previous memories. For 1997 putting this extra cognitive effort into the instance, McDonald and Hirt ( ) showed examination of unfavorable evidence (e.g., people a videotape of a fellow college stu- participants who are placed under cognitive dent who was portrayed as either likeable load ), return evaluations that are substan- or unlikable. They then provided some ad- tially less critical. ditional information about the target, in- Additional evidence of increased infor- cluding his midterm scores in several classes. mation processing of information that is in- Later, when the target’s scores on his final consistent with preferred conclusions comes exams were revealed, those who found the from Chaiken and colleagues (Giner-Sorolla target likeable remembered some of the tar- & Chaiken, 1997; Liberman & Chaiken, get’s midterm scores as lower than they ac- 1992). In one experiment, for example, peo- tually were in order to make the final scores ple read scientific reports claiming that there more consistent with improvement. In con- was either a strong link or a weak link trast, those who found the target unlikable between caffeine consumption and signi- remembered some of the midterm scores as ficant health risks, similar to the Kunda higher than they actually were in order to (1987) studies discussed earlier. As before, make the final scores more consistent with 1984 the group of women who were the most decline (see also Conway & Ross, ). threatened by this information were the least Finally, besides influencing explicit recall, convinced by the reports. In addition, the motivations to reach specific preferred con- study found that the most threatened group clusions also influence more implicit pro- of participants also expended the most ef- cesses, such as knowledge activation and fort to find flaws in the studies described and accessibility. In one demonstration of this 1999 identified the most weaknesses. (Sinclair & Kunda, ), individuals either received positive or negative feedback from a person who was a member of multiple effects on recall and knowledge activation social categories. One of these social cate- In addition to affecting the appraisal and gories (doctor) was associated with mostly encoding of new information, people’s de- positive stereotypes and another (African P1: GFZ 0521824176c13.xml CB798B/Holyoak 0521824176 October 31, 2004 6:5

300 the cambridge handbook of thinking and reasoning

American) was associated with mostly neg- rated only those traits that were central to ative stereotypes. Those who had received their own orientation (e.g., determined in the positive feedback from the other person former case versus dependable in the latter) were faster than baseline to identify doctor- as more prototypical of successful leaders related words and slower than baseline to (see also Kunda, 1987). In another study, in- identify African American-related words on dividuals rated their own characteristics as a lexical-decision task. Those who had re- more prototypical of positive qualities such ceived negative feedback showed a reverse as intelligence, but as less prototypical of pattern of activation (see also Spencer, Fein, negative qualities such as aloofness. Wolfe, Hodgson, & Dunn, 1998; for a re- Use of the third strategy can be seen versal of these effects when people are in another series of experiments by Dun- motivated by egalitarian rather than self- ning and his colleagues (Beauregard & Dun- serving outcomes, see Moskowitz, Goll- ning, 1998; Dunning & Cohen, 1992; see also witzer, Wasel, & Schaal, 1999). Alicke et al., 1997). Participants in these ex- periments were asked to judge the abilities effects on organization of concepts of others in several domains (e.g., math, ath- in memory letics). When participants themselves were Finally, beyond affecting the activation of highly skilled in the domain they were con- knowledge from memory, motivation for di- sidering or had just experienced a relevant rectional outcomes can also influence the personal success, they set higher perfor- wayinwhich people come to organize this mance standards for others. That is, to dis- knowledge. The most widely studied exam- tinguish their own superiority, they judged ple of this concerns how desires for positive others as less successful. However, when par- self-evaluation lead people to form stronger ticipants themselves were not highly skilled associations between their self-concepts and in the domain they were considering or had attributes that they feel are praiseworthy or just experienced a relevant personal failure, related to success. Three primary strategies they set lower performance standards for by which people accomplish this have been others. That is, to cast those outperform- identified: (1) altering one’s self-concept ing them as relatively high achievers, they to include attributes that are believed to judged them as more successful. bring about successful outcomes (e.g., Klein In sum, motivations for directional out- &Kunda, 1992;Kunda & Santioso, 1989); comes can affect basic cognitive processes (2) coming to view the attributes that one and influence thinking in several profound already possesses as essential for success- ways. These types of motivations affect not ful outcomes (Dunning, Leuenberger, & only how people search for, evaluate, and ex- Sherman, 1995; Dunning, Perie, & Story, plain information in the world around them, 1991;Kunda, 1987); and (3) redefining the but also how they activate, access, and orga- criteria that must be met before one can nize their knowledge about themselves and be considered successful or in possession others. The next section reviews research in- of particular positive and negative qualities dicating that motivations for nondirectional (Beauregard & Dunning, 1998; Dunning & outcomes can be equally important. Cohen, 1992; see also Alicke, LoSchiavo, Zerbst, & Zhang, 1997). Influences of Nondirectional The second two strategies are of particu- Outcome Motivation lar relevance to the issue of knowledge or- ganization. Use of the second strategy can Although less research exists concerning the clearly be seen in a program of research by cognitive effects of nondirectional outcome Dunning and his colleagues (Dunning et al., motivation, several varieties have been con- 1995; Dunning et al., 1991). In one study, sidered in some depth (e.g., Cacioppo, Petty, people who considered themselves either Feinstein, & Jarvis, 1996; Fiske & Neuberg, more goal-oriented or more people-oriented 1990; Kruglanski & Webster, 1996; Lerner & P1: GFZ 0521824176c13.xml CB798B/Holyoak 0521824176 October 31, 2004 6:5

motivated thinking 301

Tetlock, 1999). Among these, the two most there is also evidence to suggest that they prominent are desires for accuracy (Fiske & can also be influenced by accuracy and clo- Neuberg, 1990), and desires for clarity and sure motivations. conciseness, or closure (Kruglanski & Web- In one study, Tetlock (1985) had par- ster, 1996). Here, we consider the effects of ticipants read an essay either supporting these two motivations (which, as will be dis- or opposing affirmative action that had os- cussed, often have opposing effects on in- tensibly been written by someone from a formation processing) on many of the same previous experiment. They were then in- cognitive processes examined in the previ- formed that the author of the essay had been ous section. assigned to take this position by the exper- Before beginning, however, it should be imenter and asked to judge the extent to noted that both accuracy and closure mo- which the arguments presented in the essay tivation have been operationalized in multi- reflected the author’s own attitude. People ple ways. For example, motivations for accu- who were not provided with any additional racy have been studied in terms of wanting to motivations displayed the typical fixation on know as much as possible about a person on a single cause. These individuals reported whom one is going to be dependent (Neu- that the position taken in supportive es- berg & Fiske, 1987), feelings of accountabil- say could be explained by the positive atti- ity for one’s judgments (e.g., Tetlock, 1983), tude of the author toward affirmative action, a “fear of invalidity” (e.g., Kruglanski & Fre- whereas the position taken in the oppos- und, 1983), and simple desires to be as cor- ing essay could be explained by the nega- rect as possible (e.g., Neuberg, 1989). Mo- tive attitude of the author toward affirma- tivations for closure have been examined in tive action, despite knowing that both essays terms of feelings time pressure (Kruglanski & had been largely coerced by the experi- Freund, 1983), a desire to quickly complete menter. However, people who were moti- judgment tasks that are dull and unattrac- vated to make accurate judgments (by in- tive (Webster, 1993), and desires to escape forming them that they would later be noisy environments (Kruglanski, Webster, discussing the reasons for their impressions & Klem, 1993; see Kruglanski & Webster, with the experimenter) did consider the al- 1996). In the initial discussion presented, ternative cause represented by the experi- each of these varieties of accuracy or closure menter’s coercion. These individuals judged motivation are treated as equivalent; some the attitude of the author to be neutral, re- important differences among the effects of gardless of which essay they read. A study these various operationalizations, however, by Webster (1993) using a similar paradigm are considered at the end. showed that, in contrast, when participants’ motivation for closure was increased the typ- ical fixation on a single cause became even effects on attribution more pronounced. Thus, a need for accu- In addition to self-serving biases that oc- racy and a need for closure appear to have cur when people explain their own perfor- opposite effects on people’s considerations mance, as described previously, research on of alternate causes during attribution (see attribution has also identified more general Kruglanski & Freund, 1983; Kruglanski & biases. For example, there is the tendency for Webster, 1996). people to fixate on one particular cause for some action or event and then fail to ade- effects on evidence evaluation and quately consider alternative causes that are information search also possible (see Gilbert & Malone, 1995; As discussed earlier, research on directional see also Buehner & Cheng, Chap. 7; Kahne- outcome motivation has demonstrated that man & Frederick, Chap. 12). Although these people engage in increased evidence eval- attributional biases have been largely con- uations and prolonged information search sidered from a purely cognitive standpoint, when encountering evidence unfavorable P1: GFZ 0521824176c13.xml CB798B/Holyoak 0521824176 October 31, 2004 6:5

302 the cambridge handbook of thinking and reasoning

to their preferred self-views, and reduced to the participants in the study, supported evidence evaluation and information search a different verdict for each member of the when encountering evidence favorable to pair. Participants with high (versus low) clo- their preferred self-views. In contrast, ac- sure motivation attempted to bring about a curacy motivation produces prolonged in- quick end to the discussion. Moreover, when formation search, and closure motivation asked before the discussion, they expressed a produces reduced information search, re- strong preference for a partner who could be gardless of the circumstances. easily persuaded to their existing viewpoint, This consequence of accuracy motivation and once the discussion began, they stub- is evident in a study by Neuberg (1989), bornly attempted to convince their partner where people were asked to conduct a tele- to see things their way rather than consider- phone interview with a peer, but were given ing alternative arguments. unfavorable expectations concerning the in- terviewee. Those participants who were in- effects on evaluation complexity structed to “form the most accurate impres- In addition to affecting the length of peo- sions possible” of the other person spent ple’s analysis and evaluation of evidence, more time listening and provided more op- nondirectional outcome motivation can also portunities for the interviewee to elaborate influence the complexity of this analysis. his or her opinions. This in turn prevented Accuracy-motivated individuals form judg- their unfavorable expectations from creat- ments that show greater consideration of ing negative final impressions of the inter- conflicting opinions and evidence, whereas viewee, which is what occurred with those closure-motivated individuals form judg- participants who were not given any special ments that show less of this type of con- instructions for the interview. sideration. Tetlock and colleagues demon- Similar consequences of accuracy moti- strated these effects in experiments where vation are also seen in research by Chaiken participants will asked to write down their and colleagues (for reviews, see Chen & thoughts about topics such as affirmative ac- Chaiken, 1999; Eagly & Chaiken, 1993). tion, American foreign policy, and the causes For example, in one study by Maheswaran of certain historical events (for a review, see and Chaiken (1991), participants evaluated Lerner & Tetlock, 1999). Responses were a product based on a detailed review that then coded for their integrative complexity, described this product more favorably or which was defined in terms of the degree less favorably than similar products. Partic- to which multiple perspectives on an is- ipants who were high in accuracy motiva- sue were both identified and then integrated tion, because they believed their evaluations into a framework that includes complex con- would have important consequences, gener- nections between them. Findings with peo- ated more thoughts about the strengths and ple who were both novices and experts on weaknesses of the specific product-quality the issues they were analyzing (i.e., college arguments that were listed in the review students and professional historians, respec- than did those who were low in accuracy tively) indicated that those with increased motivation. This again attenuated any ef- accuracy motivation provided a more in- fects of people’s prior expectations on their tegratively complex analysis (e.g., Tetlock, final evaluations. 1983), whereas those with increased clo- The consequences of closure motiva- sure motivation provided a less integratively tion on evidence evaluation and information complex analysis (Tetlock, 1998). search has been shown in several studies by Kruglanski et al. (1993). People were paired with someone else for a discussion about effects on recall and knowledge activation the verdict of a mock trial. Before the dis- Whereas directional outcome motivation cussion, everyone received a summarized le- was seen earlier to have qualitative ef- gal analysis of the case which, unbeknownst fects on recall and knowledge activation, P1: GFZ 0521824176c13.xml CB798B/Holyoak 0521824176 October 31, 2004 6:5

motivated thinking 303

nondirectional outcome motivation has less, participants based their evaluations on largely quantitative effects. Once again, ac- whichever one of these concepts was most curacy motivation and closure motivation accessible to a greater extent when their have opposite influences. closure motivation was high but to a lesser In an investigation of accuracy moti- extent when their accuracy motivation was vation on recall during impression forma- high (Ford & Kruglanski, 1995; Thompson tion, Berscheid and colleagues found that et al., 1994). These effects have been found when people observed interviews involv- both when people are making online judg- ing individuals with whom they might later ments (Kruglanski & Freund, 1983; Schuette be paired, they paid more attention to &Fazio, 1995) and when they are recon- the interview and remembered more infor- sidering previously encountered information mation about the interviewees than when (Sanbonmatsu & Fazio, 1990; Thompson they did not expect any future interactions et al., 1994). (Berscheid, Graziano, Monson, & Dermer, Overall, then, motivations for nondirec- 1976; see also Srull, Lichtenstein, & Roth- tional outcomes can also affect basic cog- bart, 1985). However, in studies of closure nitive processes and profoundly influence motivation and impression formation, indi- thinking. Whereas motivations for direc- viduals with chronically high (versus low) tional outcomes were earlier shown to alter need for closure spent less time reading dif- how people activate, evaluate, and explain ferent pieces of behavioral information they information during reasoning, motivations were given about a target and later recalled for nondirectional outcomes (at least in fewer of these behaviors (Dijksterhuis, van terms of the accuracy and closure moti- Knippenberg, Kruglanski, & Schaper, 1996). vations reviewed here) instead alter how There is also evidence that people with high much activation, evaluation, or explanation, (versus low) accuracy motivation activate in fact, occurs. Furthermore, as the findings more pieces of individuating trait and behav- presented here illustrate, such quantitative ioral information when forming impressions differences in thought can often affect the of others (Kruglanski & Freund, 1983; Neu- outcomes of people’s judgments and deci- berg & Fiske, 1987), whereas people with sions just as much as the qualitative differ- high (versus low) need for closure display ences described previously.3 an increased tendency to rely solely on cat- egorical information during impression for- Limits to Outcome-Motivated Thinking mation (Dijksterhuis et al., 1996; Kruglanski &Freund, 1983; see also Moskowitz, 1993). Although, so far, people have been shown to Similar effects are found for the use have an impressive array of cognitive mech- of highly accessible knowledge structures anisms at their disposal when attempting or attitudes in judgment. In typical cir- to reach desired conclusions, limits do ex- cumstances, concepts or attitudes that have ist concerning when these mechanisms are been recently or frequently activated will applied. These limits are first described for lead people to assimilate their judgments directional outcome-motivated thinking and to this highly accessible information with- then for nondirectional outcome-motivated out considering any additional information thinking. (see Fazio, 1995; Higgins, 1996). Increased accuracy motivation can attenuate assimila- reality constraints on motivations for tion effects by increasing the activation of al- directional outcomes ternative interpretations, whereas increased Although there are often specific outcomes, closure motivation can exacerbate assimila- such as positive self-views, that people tion effects by decreasing the activation of al- have some preference for during judgment, ternative interpretations. For example, when most individuals still acknowledge there evaluating the behavior of a target person is some kind of “objective reality” about who was ambiguously adventurous or reck- whatever information they are considering. P1: GFZ 0521824176c13.xml CB798B/Holyoak 0521824176 October 31, 2004 6:5

304 the cambridge handbook of thinking and reasoning

That is, motivated thinking related to direc- punctuality or gullibility), people engage in tional outcomes operates within what Kunda less motivated reasoning (Dunning et al., (1990) has called reality constraints (see also 1989). Overall, these results suggest that Pyszczynski & Greenberg, 1987;cf. Kruglan- thinking and reasoning inspired by direc- ski, 1999). Therefore, although there is a tional outcomes does not so much lead peo- degree to which people adjust their defini- ple to ignore the sometimes disappointing tions of success, engage in selective recall, reality they face as it inspires them to exploit or seek to criticize unfavorable evidence, the uncertainties that exist in this reality to this does not make them entirely unrespon- their favor. sive to world around them, except perhaps in extreme circumstances (see Bachman & cognitive-resource constraints on Cannon, Chap. 21). accuracy motivation Indeed, evidence for this principle of re- Virtually all the effects of accuracy motiva- ality constraints has been repeatedly found tion reviewed here involve increases in the in the context of the research previously total amount of information processing that described. For example, a study using a people perform during judgment. Therefore, paradigm discussed earlier, where partici- in circumstances where one’s ability to en- pants first learned that introverts or extro- gage in this information processing is con- verts were generally more successful before strained, the effects of increased accuracy rating themselves on these traits, was per- motivation should be minimal (Fiske & Neu- formed using participants who had been pre- berg, 1990). One demonstration of this was selected as having high trait levels of either provided by Pendry and Macrae (1994). As introversion or extroversion (Santioso et al., described earlier, accuracy-motivated indi- 1990). Although beliefs that one trait was viduals who were forming an impression of more beneficial than the other increased ev- a target displayed an increased use of indi- eryone’s self-ratings concerning that trait, viduating trait and behavioral information demonstrating motivated reasoning, there when they possessed their full information was also a large effect of people’s chronic dis- processing resources (see Neuberg & Fiske, positions. Introverts’ ratings of themselves, 1987). However, accuracy-motivated indi- were always more introverted than extro- viduals, whose processing resources were verts’ ratings of themselves, no matter how depleted based their impression primarily beneficial the introverts believed the trait of on categorical information, similar to those extroversion to be. That is, regardless of how who had little accuracy motivation (see also desirable it would have been, introverts did Kruglanski & Freund, 1983). In addition, not suddenly believe themselves to be extro- Sanbonmatsu and Fazio (1990) showed that verts and vice versa. the influence of accuracy motivation in re- Another example of the influence of ducing people’s assimilation of their judg- reality constraints is that people’s think- ments to highly accessible attitudes disap- ing is guided by their preferred outcomes pears when people are placed under time to a much greater extent in situations of pressure, which prevents extended informa- uncertainty (e.g., Dunning, Meyerowitz, & tion processing. Holtzberg, 1989; Hsee, 1995). When there is more potential for constructing idiosyncratic does motivation for accuracy result in criteria for a certain judgment (e.g., judging accurate reasoning? whether one possess somewhat vague traits Another important consideration of the ef- such as sensibility or insecurity), then peo- fects of accuracy motivation on thinking and ple use this opportunity to select criteria that reasoning is that even when people high in allow them to reach their desired conclu- accuracy motivation are free to engage in sion. However, when there is less potential extended information processing, this does for this construction (e.g., judging whether not guarantee that they will arrive at more one possesses more precise traits such as accurate judgments. One obvious example P1: GFZ 0521824176c13.xml CB798B/Holyoak 0521824176 October 31, 2004 6:5

motivated thinking 305

of this situation is where evidence beyond of this motivation and the larger context in what is immediately and effortlessly avail- which it exists. able does not exist or has faded from mem- ory (see, e.g., Thompson et al., 1994). An- the influence of information availability on other is where people are affected by certain closure motivation biases that are outside their awareness, or Certain qualifications must also be noted in where people are aware of such biases, but the effects of closure motivation. All the unaware of what the proper strategy is to findings discussed so far have involved the correct them. In all these circumstances, al- tendency for people with increased closure though accuracy motivation might increase motivation to quickly assimilate their judg- information search, recall, and considera- ments to readily available or highly accessi- tion of multiple interpretations, it would not ble information, leading to an early “freez- be expected to eliminate judgment errors ing” of their information search. However, 1982 (Fischhoff, ), and might even increase in situations where little information is avail- 1995 them (Pelham & Neter, ;Tetlock & able, high closure motivation may inspire ef- 1989 Boettger, ). forts to find something clear and concise to “seize” upon and increase information search (see Kruglanski & Webster, 1996). For exam- distinctions among circumstances that lead 1993 to accuracy motivation ple, in the Kruglanski et al. ( ) studies described previously that involved partners As alluded to earlier, the different types discussing the verdict of a mock trial, people of accuracy motivation inductions reviewed with high closure motivation preferred eas- here are not always equivalent and can have ily persuadable partners and were unwilling markedly different effects. For example, al- to consider alternative arguments only when though having one’s outcomes dependent on they had enough information at their dis- another person can increase desires for accu- posal (i.e., a summarized legal analysis) to racy in diagnosing that person’s true charac- form a clear initial impression. When these 1987 ter (e.g., Neuberg & Fiske, ), in other same individuals were not provided with the cases such circumstances can produce a de- legal analysis and did not begin the discus- sire to see a person that one is going to be sion with a clear opinion, they expressed a depending on in the best possible light (e.g., desire to be paired with someone who was 1976 1992 Berscheid et al., ; Klein & Kunda, ; highly persuasive and shifted toward their 1996 see Kruglanski, ). As another exam- partner’s point of view. ple, although believing that one’s judgment has important consequences may motivate Conclusions on Outcome-Motivated an accurate consideration of all the relevant Thinking evidence, it could also motivate a more gen- eral need to increase elaborative thinking Recent research has uncovered many po- that is not necessarily focused on accuracy tential routes by which people’s desires for (see Footnote 3;Petty & Wegener, 1999). Fi- particular judgment outcomes can affect nally, although justifying one’s judgments to their thinking and reasoning. To summa- an audience can motivate accuracy when the rize, both directional outcome motivations, opinion of the audience is unknown, it can where people have a specific preferred con- also lead to more directional outcome mo- clusion they are trying to reach, and nondi- tivation, such as ingratiation toward this au- rectional outcome motivations, where peo- dience, when the opinion of the audience is ple’s preferred conclusions are more general, known (Tetlock, 1983; see Lerner & Tetlock, alter many basic cognitive processes during 1999). Therefore, when attempting to antic- reasoning. These include (1) the explana- ipate the effects of accuracy motivation on tion of events and behaviors; (2) the or- reasoning in a particular situation, it is im- ganization, recall, and activation of knowl- portant to consider both the current source edge in memory; and (3) the pursuit and P1: GFZ 0521824176c13.xml CB798B/Holyoak 0521824176 October 31, 2004 6:5

306 the cambridge handbook of thinking and reasoning

evaluation of evidence relevant to decision Regulatory focus theory distinguishes be- making. Outcome motivation effects involve tween two basic motivational orientations: both how such cognitive processes are initi- a promotion focus involving concerns with ated and directed, as well as how thoroughly advancement and approaching gains versus these processes are implemented. Moreover, avoiding non-gains, and a prevention focus in any given situation, the specific cognitive involving concerns with security and ap- processes influenced by outcome motivation proaching non-losses versus avoiding losses. are typically those that aid the gathering and Because it centers on the presence and ab- interpretation of information supporting the sence of positive outcomes, a promotion fo- favored outcome. In this self-fulfilling way, cus has been found to create preferences then, people’s outcome-motivated reason- for eager judgment strategies that empha- ing often successfully brings about their de- size advancement (or, to use signal detec- sired conclusions. tion terminology, finding hits) and ensure against overlooking something that might be important (or, to again use signal de- tection terminology, avoiding errors of omis- Strategy-Motivated Thinking sion). In contrast, because it centers on the presence and absence of negative outcomes, Although outcome-motivated thinking has a prevention focus has been found to engen- been the most widely studied form of mo- der preferences for vigilant judgment strate- tivated reasoning, other varieties of motiva- gies that emphasize protection (or making tional influences on cognition are also pos- correct rejections) and ensure against com- sible. One alternate perspective that has mitting to something that might be a mis- more recently emerged, and which com- take (or avoiding errors of commission; see plements an outcome-based view, proposes Higgins & Molden, 2003). Therefore, even that people are motivated not only with re- in circumstances where individuals are pur- spect to the outcomes of their judgments, suing the same outcome, they may show but also with respect to the manner in marked differences in their pursuit of this which they go about making these judg- outcome depending upon whether they are ments. That is, not only do people have currently promotion-focused or prevention- preferred conclusions, but they also have focused. The studies reviewed here are in- preferred strategies for reaching their conclu- tended to illustrate the effects of eager sions (Higgins & Molden, 2003;cf. Tyler or vigilant strategic motivation on several & Blader, 2000). Therefore, independent of types of thought processes similar to those whatever outcome holds the most interest found to be influenced by outcome moti- for them, people may be motivated to reach vation (for a larger overview, see Higgins & these outcomes using strategies that “feel Molden, 2003). right” in terms of, and allow them to sus- tain, their current motivational orientation effects on the consideration of (e.g., eagerly gathering evidence that might alternative hypotheses support a positive self-view or facilitate cog- Considering alternative hypotheses is a fun- nitive closure vs. vigilantly suppressing evi- damental component of many varieties of dence that could undermine a positive self- thinking (see Sloman & Lagnado, Chap. 5). view or threaten cognitive closure). How might eager versus vigilant strategic Several lines of research have examined preferences influence this process? In gen- how motivations for particular judgment eral, an eager strategy of considering alter- strategies can also influence people’s ba- natives would involve attempting to attain sic cognitive processes. In the vast majority hits and to ensure against errors of omission of these studies, strategic motivations were by generating and selecting any plausible measured and manipulated in terms of peo- hypotheses that could remotely be correct. ple’s regulatory focus (see Higgins, 1997). However, a vigilant strategy of considering P1: GFZ 0521824176c13.xml CB798B/Holyoak 0521824176 October 31, 2004 6:5

motivated thinking 307

alternatives would involve attempting to compared with those in a prevention focus make correct rejections and to ensure against (see Kelley, 1973). errors of commission by generating and se- Finally, additional research by Molden lecting only the most probable hypotheses and Higgins (2004) has more recently that seem likely to be correct. Therefore demonstrated similar effects for eager versus people in a promotion focus would be ex- vigilant strategic preferences on the genera- pected to consider a greater number of alter- tion and selection of alternatives during basic natives during thinking and reasoning than categorization processes. People were given people in a prevention focus. vague descriptions of a target person where This question was addressed in several it was not clear how to correctly categorize studies by Liberman, Molden, Idson, and him and a number of alternatives could all Higgins (2001). One important instance of have been possible. As before, participants considering alternatives occurs when people with either a chronic or experimentally in- form hypotheses about what they are per- duced promotion focus generated more pos- ceiving (see Tversky, Chap. 10). Therefore, sible categories for the target than those with Liberman et al. (2001) examined the effects either a chronic or experimentally induced of people’s strategic preferences on a task prevention focus. where people identified vague and distorted Overall, then, people’s eager versus vig- objects in a series of photographs. Across ilant strategic preferences play a significant several studies where a promotion or pre- role in their generation of alternatives during vention focus was both measured as an indi- a number of important thought processes. vidual differences variable and induced ex- Moreover, it is important to note that in all perimentally, results indicated that those in a the studies described in this section, every- promotion focus generated a greater number one was pursuing the exact same outcome of alternatives for the identity of the objects (identifying an object, explaining behaviors) than those in a prevention focus (see also and did not have motivations for any specific Crowe & Higgins, 1997). conclusion or end-state. Furthermore, mea- In addition to examining the effects of sures of people’s motivations for more gen- strategic preferences on generating alterna- eral outcomes such as accuracy and closure tive hypotheses for object perception, Liber- were also taken and these factors were sta- man et al. (2001) also investigated whether tistically removed from all analyses. There- similar effects occurred for social percep- fore, the observed effects of promotion or tion. Participants read a scenario describing prevention motivational orientations are dis- the helpful behavior of a target person and tinct from the outcome motivation effects were asked to evaluate several equally plau- reviewed earlier and can be attributed to the sible alternative explanations for this behav- influences of these orientations on people’s ior. Consistent with the results described strategic preferences. previously, participants in a promotion fo- cus again selected a greater number of al- ternative explanations than participants in effects on counterfactual thinking a prevention focus. Moreover, these effects Besides generating and evaluating hypothe- were also found to influence the general im- ses, another way in which people consider al- pressions people formed of the target. Af- ternatives during reasoning is in their use of ter selecting their reasons for the target’s counterfactuals. As briefly mentioned previ- helpful behavior, participants predicted how ously, counterfactual thinking involves men- helpfully he or she would behave in the fu- tally undoing the present state of affairs and ture. Those in a promotion focus, because imagining alternative realities “if only” dif- they were considering more interpretations ferent decisions had been made or actions of a target’s behavior, formed more equivo- been taken (Roese, 1997). Several differ- cal impressions and showed relatively little ent varieties of counterfactual thinking have generalization about the target’s behavior as been identified. One broad distinction that P1: GFZ 0521824176c13.xml CB798B/Holyoak 0521824176 October 31, 2004 6:5

308 the cambridge handbook of thinking and reasoning

has been made is between thoughts that how much regret they inspired (see Roese concern the reversal of a previous inaction et al., 1999). Therefore, the results can again (e.g., if only I had acted, things might have only be explained in terms of differences in gone better), or additive counterfactuals, and strategic motivation. thoughts that concern the reversal of a pre- vious action (e.g., if only I hadn’t acted, effects on fast versus accurate things wouldn’t be so bad), or subtractive information processing counterfactuals. A major question across many areas of psy- Because additive counterfactuals simu- chology has been when and why people late the correction of a past error of omis- choose to emphasize either speed or ac- sion, this type of thinking represents a more curacy in their thinking and decision mak- eager strategy of considering alternative real- ing (e.g., Josephs & Hahn, 1995; Zelaznik, ities. In contrast, because subtractive coun- Mone, McCabe, & Thaman, 1988). Forster,¨ terfactuals simulate the correction of a past Higgins, and Bianco (2003) more recently error of commission, this type of thinking investigated whether promotion preferences represents a more vigilant strategy of con- for strategic eagerness would result in faster sidering alternate realities. Therefore, a pro- information processing and a higher quan- motion focus should increase the generation tity of output in a search for possible hits, of additive counterfactuals and a prevention whereas prevention preferences for strategic focus should increase the generation of sub- vigilance would result in more accurate in- tractive counterfactuals. In line with this, formation processing and a higher quality of Roese, Hur, and Pennington (1999) found output in an effort to avoid mistakes. that, both when analyzing hypothetical ex- Participants were given a task involving amples and when describing particular in- four pictures taken from a children’s “con- stances of their own behavior, participants nect the dots” drawing book. For each pic- who considered promotion-related setbacks ture, the objective was to connect sequen- (i.e., nongains and missed opportunities for tially numbered dots within a given time advancement) offered a greater number of period in order to complete the outline of additive counterfactuals, whereas partici- an image. Participants’ speed on each pic- pants who considered prevention-related ture was assessed by the highest number setbacks (i.e., losses and missed opportuni- dot they reached by the end of the time ties to prevent mistakes) offered a greater period for that picture, and their accuracy number of subtractive counterfactuals. In on each picture was assessed by the num- the literature that exists on counterfactual ber of dots they skipped (i.e., that were thinking, it has been traditionally assumed not connected). Across two studies where that subtractive counterfactuals are more participants’ promotion or prevention focus common than additive counterfactuals and was both measured and experimentally in- that failures associated with action inspire duced, promotion-focused individuals were more regret than failures associated with in- faster and produced a higher quantity of re- action (Roese, 1997). However, the results sponses, whereas prevention-focused indi- of these studies demonstrate that, in some viduals were more accurate and produced cases, people’s strategic preferences can re- a higher quality of responses over the entire sult in additive counterfactuals being more task. Moreover, both of these tendencies in- common and perhaps being associated with creased in intensity as people moved closer greater regret (see also Camacho, Higgins, & to goal completion, resulting in stronger Lugar, 2003). effects of strategic preferences toward the It is important to note that care was taken end of a task than toward the beginning to make sure the outcomes that participants of a task (i.e., the “goal looms larger” ef- were considering in these studies did not dif- fect where motivation increases as one’s dis- fer across any important dimensions, such tance to the completion of a goal decreases; as how painful they were imagined to be or Lewin, 1935). This provides strong support P1: GFZ 0521824176c13.xml CB798B/Holyoak 0521824176 October 31, 2004 6:5

motivated thinking 309

that people’s motivations for different judg- tion that is more meaningful in the context ment strategies can alter their concerns with of eager strategic preferences, whereas in- different aspects of information processing dividuals who were chronically prevention- (e.g., speed vs. accuracy). focused showed a stronger sensitivity and re- call for loss-related information that is more meaningful in the context of vigilant strate- effects on knowledge activation and recall gic preferences. Analogous to the selective recall and ac- tivation of information from memory that Strategic Preferences and Regulatory Fit occurs in the presence of motivations for directional outcomes, another influence of Although the studies presented thus far strategic preferences on thinking is to in- have demonstrated how people’s motiva- crease sensitivities to, and recall of, informa- tional orientations can lead them to pre- tion that that is particularly relevant to these fer and choose certain judgment strategies, preferences. A study by Higgins, Roney, situations may exist in which they may be Crowe, and Hymes (1994) demonstrated more or less able to follow these preferences. this by having participants read an essay For example, some situations may gener- about the life of a hypothetical target person ally require greater use of eager strategies in which two different types of situations of pursuing gains or vigilant strategies of were encountered. In one type of situation, preventing mistakes, such as when supervi- the target used eager strategies that were ad- sors demand either innovative and creative vancement oriented (e.g., waking up early practices of all their employees in search in order to be on time for a favorite class), of advancement or cautious and responsi- whereas in the other type of situation, the ble practices in hope of preventing losses. target used vigilant strategies that were more What might be the consequences of making protection oriented (e.g., being careful not judgments and decisions in a way that ei- to sign up for a class whose schedule con- ther suits one’s current strategic preferences flicted with a desired activity). Individu- (i.e., promotion-focused individuals using als who had chronic promotion orientations eager strategies and prevention-focused in- showed a stronger sensitivity for information dividuals using vigilant strategies) or does related to advancement versus protection not suit one’s preferences (i.e., promotion- strategies, and later showed greater recall for focused individuals using vigilant strategies these episodes, whereas individuals who had and prevention-focused individuals using ea- chronic prevention orientations showed the ger strategies)? reverse effect. Higgins and colleagues have examined Another study by Higgins and Tykocin- this question and investigated how the regu- ski (1992), which again had people read an latory fit between one’s motivational orien- essay about the life of a hypothetical tar- tation and the means one uses during goal get person, extends these findings. In this pursuit affects thinking and reasoning (e.g., study, the target person experienced situa- Camacho et al., 2003;Freitas & Higgins tions that either involved the presence or 2002; Higgins, Idson, Freitas, Spiegel, & absence of gains (finding $20 on the street Molden, 2003). Although space limitations or missing a movie that he wanted to see, prohibit a more thorough review of this respectively) or the presence of absence of work here (see Higgins, 2000a; Higgins & losses (being stuck in a crowded subway Molden, 2003), the general findings have for an extended period of time or getting been that that the primary consequence of a day off from a particularly arduous class regulatory fit is to increase the perceived schedule, respectively). Similar to the previ- value of the goal one is pursuing. That is, reg- ous study, individuals who were chronically ulatory fit (as compared with nonfit) leads promotion-focused showed a stronger sen- people to “feel right” about their goal pur- sitivity and recall for gain-related informa- suit, which then leads them to (1) feel good P1: GFZ 0521824176c13.xml CB798B/Holyoak 0521824176 October 31, 2004 6:5

310 the cambridge handbook of thinking and reasoning

while pursuing these goals (i.e., what feels tions have important effects on their think- right feels good; see Freitas & Higgins, 2002); ing and reasoning above and beyond their (2) experience the outcomes they are striv- outcome motivations. ing for as having more value or worth (i.e., what feels right is good; see Higgins et al., 2003); and (3) believe the strategies they are using are inherently right (i.e., what feels General Conclusions and Future right is right; see Camacho et al., 2003). Directions Therefore, another avenue for future re- search on how people’s motivations to use The sheer number and diversity of the stud- certain judgment strategies during judgment ies reviewed here is a testament to the return can affect their thought processes is the fur- of motivational perspectives on cognition ther refinement and elaboration of the pro- to the vanguard of psychology. The rich- cess of regulatory fit. ness and consistency of the findings emerg- ing from these studies is also a testament to the utility of this perspective in the study Conclusions on Strategy-Motivated of thinking and reasoning. We optimistically Thinking forecast a further expansion of research in- formed by motivational perspectives and, in In sum, several emerging programs of re- conclusion, briefly outline two general di- search are beginning to demonstrate that, rections we believe should be priorities for beyond the effects on reasoning of people’s the future. desires for particular judgment outcomes, The first direction involves expanding there are additional effects on reasoning of current conceptualizations of the ways in people’s desires to use particular judgment which motivational and cognitive processes strategies. For example, preferences for ea- interact during judgment. Although there ger judgment strategies, shown by those with is still much to be learned from examining promotion concerns, versus preferences for the effects on thinking of people’s motiva- vigilant judgment strategies, shown by those tions for certain outcomes (either directional with prevention concerns, alter many basic or nondirectional), there may potentially cognitive processes during reasoning. These be other important sources of motivated include (1) the generation and testing of hy- thought as well. In this chapter, we reviewed potheses, (2) the use of counterfactual think- our own initial research on one of these ing, (3)anemphasis on fast versus accurate possible sources – people’s motivations for processing of information, and (4) knowl- employing preferred strategies during judg- edge activation and recall. Strategy motiva- ment. We expect that further study will lead tion effects include whether cognitive pro- to the development of additional perspec- cesses are implemented in order to advance tives on the interface of motivation and cog- the right decision and avoid errors of omis- nition that go beyond both motivated out- sion in judgment or to protect against the comes and motivated strategies. wrong decision and avoid errors of commis- The second direction involves moving sion in judgment. They also include whether past research that examines different vari- such implementation fits or does not fit one’s eties of motivated thinking in isolation from current motivational orientation. The imple- one another (i.e., studying situations where mentation of cognitive processes for either people are only motivated to achieve posi- of these strategic reasons or for regulatory tive self-views or only motivated to be ac- fit influences what pieces of information are curate). There is a need to consider how considered during judgment and how much multiple goals, desires, and motives inter- this information is valued in a final decision. act to influence the thought process, that In this way, then, people’s strategic motiva- is the effects of patterns of motivational P1: GFZ 0521824176c13.xml CB798B/Holyoak 0521824176 October 31, 2004 6:5

motivated thinking 311

forces. For instance, it has been noted for ing success but external causes following some time now that people possess many failure, showed strong generalizations fol- potential objectives when processing in- lowing success and almost no generaliza- formation (e.g., Chen & Chaiken, 1999). tion after failure. These results demonstrate Although it is certainly the case that, at the importance of considering the effects of times, objectives such as accuracy, ingra- multiple sources of motivated reasoning si- tiation, or self-enhancement may be pre- multaneously (see also Forster,¨ Higgins, & dominant (Kruglanski, 1999), it is also true Strack, 2000). that there are many instances in which sev- One final way in which investigating the eral of these objectives are pursued simul- cognitive effects of interacting motivational taneously. What happens when people not forces could be fruitfully expanded is by only want to be accurate, but also want to synthesizing work on how motivation influ- please others or boost their own self-esteem? ences reasoning with work on how affect in- Studies addressing these questions are just fluences reasoning (see Forgas, 2000; Mar- beginning to appear and early findings are tin & Clore, 2001). Great strides have been indicating that important interactions can made in determining the mechanisms by occur (Lundgren & Prislin, 1998; Nienhuis, which affective and emotional states can al- Manstead, & Spears, 2001; Ruscher, Fiske, & ter people’s judgments. Many of the changes Schnake, 2000). in the quality and quantity of information Similarly, although we have made a dis- processing found in this research bear a strik- tinction between outcome- and strategy- ing resemblance to the motivational effects motivated thinking and discussed their ef- reviewed here. For example, positive moods fects independently, there are situations have generally been found to support less where these two sources of motivation op- thorough and complex information process- erate in concert. One of these situations ing, similar to closure motivation, whereas has been the focus of recent studies by negative moods have generally been found to Molden and Higgins (2004). These studies support more thorough and complex infor- examined how preferences for eager ver- mation processing, similar to accuracy mo- sus vigilant decision strategies influence peo- tivation (for a review, see Schwarz & Clore, ple’s generation of alternative explanations 1996). This is not to say, however, that the for their own success and failure. In addi- effects reviewed here are actually just due tion to replicating both the previously dis- to changes in emotion, because many of the cussed self-serving pattern of attributions studies discussed carefully controlled for af- for performance (an outcome-motivated ef- fective influences and continued to find in- fect) and the selection of a greater number dependent effects. Therefore, it would be of alternative attributions by those prefer- fruitful to investigate how affective thinking ring eager strategies than vigilant strategies may give rise to motivational thinking (e.g., (a strategy-motivated effect), these studies Erber & Erber, 2000), and how motivational showed that self-serving and strategic moti- thinking may give rise to affective thinking vations interacted to determine the extent to (e.g., Higgins, 2000b), in order to develop a which people generalized their current ex- better understanding of how these two fac- periences to their future performance. Indi- tors are related and what their combined and viduals using eager strategies, because they separate consequences might be. tended to consider multiple attributions, in- To conclude, this chapter reviewed re- cluding both internal and external causes, search that displays the broad applicability showed only moderate generalization after of emerging motivational perspectives to the both success and failure. In contrast, indi- study of thinking and reasoning. Through viduals using vigilant strategies, because they this review, we attempted to convey the po- tended to consider only a few attributions, tential utility of these perspectives and to ad- including primarily internal causes follow- vocate a greater incorporation of principles P1: GFZ 0521824176c13.xml CB798B/Holyoak 0521824176 October 31, 2004 6:5

312 the cambridge handbook of thinking and reasoning

of outcome- and strategy-based motivation thinking are to be revealed, future investiga- in future research. The further refinement tions of outcome-motivated thinking in dif- and elaboration of these principles, we be- ferent cultures should take care to identify lieve, will benefit not only the study of think- which specific outcomes are culturally desirable ing but also cognitive science in general. in those contexts (e.g., proper fulfillment of one’s social duties to others, high social sta- tus relative to others; see, e.g., Endo, Heine, & Lehman, 2000). Acknowledgments 3. Another type of nondirectional outcome mo- tivation that has been the focus of considerable Preparation of this chapter was supported study is the need for cognition,orageneral desire 31 65772 01 for elaborative thinking and increased cogni- by NIMH grants F MH- - and 1996 39429 tive activity (Cacioppo et al., ). At times, MH . The authors want to thank the need for cognition has been considered Phoebe Ellsworth for her helpful comments equivalent to accuracy motivation (Chen & on an earlier draft. Chaiken, 1999). Consistent with this, research has shown that an increased need for cognition can affect thinking in the same way as height- ened accuracy motivation, reducing biases dur- Notes ing attribution (D’agostino & Fincher-Kiefer, 1992), increasing recall (Srull et al., 1985), less- 1. One area of study that is notably absent in ening assimilation to highly accessible attitudes this review concerns affective and emotional (Florack, Scarabis, & Bless, 2001), and increas- influences on reasoning. This important and ing information search (Verplanken, 1993; see extensive literature certainly enjoys a central Cacioppo et al., 1996). However, at times the place in the study of motivated thinking. How- effects of the need for cognition differ from ever, the topic of affect and cognition has re- those of accuracy motivation. Accuracy moti- cently been the subject of several entire hand- vation, because it inspires a thorough consid- books on its own (see Forgas, 2000; Martin eration of all available evidence, weakens the & Clore, 2001). Therefore, rather than at- tendency to base judgments on early superfi- tempt an extremely limited overview of this cial impressions (i.e., primacy effects; Kruglan- major topic alongside the other topics men- ski & Freund, 1983). In contrast, the need tioned previously, we instead refer the inter- for cognition, because it simply inspires cog- ested reader to these other sources. The larger nitive elaboration even if this involves only relation between research on emotional think- part of the available evidence, can lead to in- ing and the research described here is discussed creased rumination on one’s early superficial briefly below. impressions and strengthen primacy effects (see 2.Itisimportant to note that, although a wealth Petty & Wegener, 1999). Given these concep- of studies have demonstrated people’s broad tual and empirical distinctions, we have not in- and robust desires for positive self-evaluation, cluded research on need for cognition in our these studies have almost exclusively been per- larger review of the effects of accuracy moti- formed on members of Western, and gener- vation and consider it a separate form of nondi- ally more individualistic cultures (Baumeister, rectional outcome motivation (for a review 1998). In contrast, recent evidence collected of need for cognition effects, see Cacioppo from Eastern, and generally more collectivist et al., 1996). cultures, has demonstrated that, in these pop- ulations, such desires for self-evaluation are of- ten considerably less, and that some of the ef- fects described here are thereby weaker (see References Greenfield, Chap. 27). Yet, this should not be taken to mean that the general effects Alicke, M. D., LoSchiavo, F. M., Zerbst, J. I., & of outcome-motivated thinking are necessar- Zhang, S. (1997). The person who outperforms ily culture specific or only apply to West- me is a genius: Maintaining perceived compe- ern cultures. Instead, this indicates that, if tence in upward social comparison. Journal of general principles of this type of motivated Personality and Social Psychology, 73, 781–789. P1: GFZ 0521824176c13.xml CB798B/Holyoak 0521824176 October 31, 2004 6:5

motivated thinking 313

Allport, F. H. (1920). The influence of the group Dijksterhuis, A., van Knippenberg, A., Kruglan- upon association and thought. Journal of Exper- ski, A. W., & Schaper, C. (1996). Motivated imental Psychology, 3, 159–182. social cognition: Need for closure effects on Baumeister, R. F. (1998). The self. In D. Gilbert, memory and judgment. Journal of Experimen- S. Fiske, & G. Lindzey (Eds.), The handbook of tal Social Psychology, 32, 254–270. social psychology (4th ed., Vol. 1,pp.680–740). Ditto, P. H., & Lopez, D. F. (1992). Motivated New York: McGraw-Hill. skepticism: Use of differential decision criteria Beauregard, K. S., & Dunning, D. (1998). Turn- for preferred and non-preferred conclusions. ing up the contrast: Self-enhancement motives Journal of Personality and Social Psychology, 63, prompt egocentric contrast effects in social 569–584. judgment. Journal of Personality and Social Psy- Ditto, P. H., Scepansky, J. A., Munro, G. D., chology, 74, 606–621. Apanovitch, A. M., & Lockhart, L. K. Bem, D. J. (1967). Self-perception: An alter- (1998). Motivated sensitivity to preference- native interpretation of cognitive dissonance inconsistent information. Journal of Personality phenomena. Psychological Review, 74, 183– and Social Psychology, 75, 53–69. 200. Doosje, B., Spears, R., & Koomen, W. (1995). Berscheid, E., Graziano, W., Monson, T., & Der- When bad isn’t all bad: Strategic use of sample mer, M. (1976). Outcome dependency: At- information in generalization and stereotyping. tention, attribution, and attraction. Journal Journal of Personality and Social Psychology, 69, of Personality and Social Psychology, 34, 978– 642–655. 989. Dunning, D. (1999). A newer look: Motivated so- Bradley, G. W. (1978). Self-serving biases in the cial cognition and the schematic representation attribution process: A reexamination of the fact of social concepts. Psychological Inquiry, 10, or fiction question. Journal of Personality and 1–11. Social Psychology, 18, 68–88. Dunning, D., & Cohen, G. L. (1992). Egocentric Bruner, J. S. (1957). On perceptual readiness. Psy- definitions of traits and abilities in social judg- chological Review, 64, 123–152. ment. Journal of Personality and Social Psychol- ogy, , 341–355. Cacioppo, J. T., Petty, R. E., Feinstein, J. A., 63 &Jarvis, W. B. G. (1996). Dispositional dif- Dunning, D., Leuenberger, A., & Sherman, D. A. ferences in cognitive motivation: The life and (1995). A new look at motivated inference: Are times of individuals varying in need for cogni- self serving theories of success a product of mo- tion. Psychological Bulletin, 119, 197–253. tivational forces? Journal of Personality and So- 58 68 Camacho, C. J., Higgins, E. T.,& Lugar, L. (2003). cial Psychology, 69, – . Moral value transfer from regulatory fit: What Dunning, D., Meyerowitz, J. A., & Holtzberg, feels right is right and what feels wrong is A. D. (1989). Ambiguity and self-evaluation: wrong. Journal of Personality and Social Psychol- The role of idiosyncratic trait definitions in self- ogy, 84, 498–510. serving assessments of ability. Journal of Person- 1082 1090 Chen S., & Chaiken, S. (1999). The heuristic- ality and Social Psychology, 57, – . systematic model in its broader context. In S. Dunning, D. A., Perie, M., & Story, A. L. (1991). Chaiken & Y. Trope (Eds.), Dual-process theo- Self-serving prototypes of social categories. ries in social psychology (pp. 73–96). New York: Journal of Personality and Social Psychology, 61 , Guilford. 957–968. Conway, M., & Ross, M. (1984). Getting what Eagly, A. H., & Chaiken, S. (1993). The psy- you want by revising what you had. Journal of chology of attitudes. New York: Harcourt Brace Personality and Social Psychology, 47, 738–748. College. Crowe, E., & Higgins, E. T. (1997). Regulatory Endo, Y., Heine, S. J., & Lehman, D. R. focus and strategic inclinations: Promotion and (2000). Culture and positive illusions in close prevention in decision-making. Organizational relationships: How my relationships are better Behavior and Human Decision Processes, 69, than yours. Personality & Social Psychology Bul- 117–132. letin, 26, 1571–1586. D’agostino, P. R., & Fincher-Kiefer, R. (1992). Erber, M. W., & Erber, R. (2000). The role of mo- Need for cognition and the correspondence tivated social cognition in the regulation of af- bias. Social Cognition, 10, 151–163. fective states. In J. P. Forgas (Ed.), Handbook P1: GFZ 0521824176c13.xml CB798B/Holyoak 0521824176 October 31, 2004 6:5

314 the cambridge handbook of thinking and reasoning

of affect and social cognition (pp. 275–292). Giner-Sorolla, R., & Chaiken, S. (1997). Selective Mahwah, NJ: Erlbaum. use of heuristic and systematic processing un- Fazio, R. H. (1995). Attitudes as object- der defense motivation. Personality and Social evaluation associations: Determinants, conse- Psychology Bulletin, 23, 84–97. quences, and correlates of attitude accessibil- Gollwitzer, P. M., & Bargh, J. A. (Eds.). ity. In R. E. Petty & J. A. Krosnick (Eds.), (1996). The psychology of action: Linking cog- Attitude strength: Antecedents and consequences nition and motivation to behavior. New York: (pp. 247–282). Mahwah, NJ: Erlbaum. Guilford. Festinger, L. (1957). A theory of cognitive disso- Gollwitzer, P. M., Earle, W. B., & Stephan, W. G. nance. Stanford, CA: Stanford University Press. (1982). Affect as a determinant of egotism: Fischhoff, B. (1982). “Debiasing.” In D. Kahne- Residual excitation and performance attribu- man, P. Slovic, & A. Tversky (Eds.), Judgment tions. Journal of Personality and Social Psychol- under uncertainty (pp. 237–262). Cambridge, ogy, 43, 702–709. MA: Cambridge University Press. Higgins, E. T. (1996). Knowledge activation: Fiske, S. T., & Neuberg, S. L. (1990). A contin- Accessibility, applicability, and salience. In uum of impression formation, from category- E. T. Higgins & A. W. Kruglanski (Eds.), So- based to individuating processes: Influences of cial psychology: Handbook of basic principles information and motivation on attention and (pp. 133–168). New York: Guilford. interpretation. In M. P. Zanna (Ed.), Advances Higgins, E. T. (1997). Beyond pleasure and pain. in experimental social psychology (Vol. 23,pp.1– American Psychologist, 52, 1280–1300. 74 ). New York: Academic Press. Higgins, E. T. (2000a). Making a good decision: Florack, A., Scarabis, M., & Bless, H. (2001). Value from “fit.” American Psychologist, 55, When do associations matter? The use of auto- 1217–1230. matic associations toward ethnic groups in per- Higgins, E. T. (2000b). Promotion and preven- son judgments. Journal of Experimental Social tion experiences: Relating emotions to non- 518 524 Psychology, 37, – . emotional motivational states. In J. P. Forgas Ford, T. E., & Kruglanski, A. W. (1995). Effects (Ed.), Handbook of affect and social cognition of epistemic motivations on the use of acces- (pp. 186–211). Mahwah, NJ: Erlbaum. sible constructs in social judgment. Personality Higgins, E. T., Idson, L. C., Freitas, A. L., Spiegel, 950 962 & Social Psychology Bulletin, 21, – . S.,&Molden, D. C. (2003). Transfer of value Forgas, J. P. (Ed.). (2000). Handbook of affect and from fit. Journal of Personality and Social Psy- social cognition. Mahwah, NJ: Erlbaum. chology, 84, 1140–1153. Forster,¨ J., Higgins, E. T., & Bianco, A. T. (2003). Higgins, E. T., & Molden, D. C. (2003). How Speed/accuracy decisions in task performance: strategies for making judgments and decisions Built in trade-off of separate strategic con- affect cognition: Motivated cognition revisited. cerns. Organization Behavior and Human De- In G. V. Bodenhausen & A. J. Lambert (Eds.), cision Processes, 90, 148–164. Foundations of social cognition: A festschrift in Forster,¨ J., Higgins, E. T., & Strack, F. (2000). honor of Robert S Wyer, Jr. (pp. 211–236). Mah- When stereotype disconfirmation is a personal wah, NJ: Erlbaum. threat: How prejudice and prevention focus Higgins, E. T., Roney, C., Crowe, E., & Hymes, C. moderate incongruency effects. Social Cogni- (1994). Ideal versus ought predilections for ap- tion, 18, 178–197. proach and avoidance: Distinct self-regulatory Freitas, A. L., & Higgins, E. T. (2002). Enjoying systems. Journal of Personality and Social Psy- goal-directed action: The role of regulatory fit. chology, 66, 276–286. 1 6 Psychological Science, 13, – . Higgins, E. T., & Tykocinski, O. (1992). Self- Freud, S. (1905). Fragments of an analysis of a case discrepancies and biographical memory: Per- of hysteria. In J. Strachey (Ed.), “The Standard sonality and cognition at the level of psycholog- Edition of the Complete Psychological Works ical situations. Personality and Social Psychology of Sigmund Freud.” Standard edition, vol. 7. Bulletin, 18, 527–535. 1964 Macmillan: Hogarth, . Hsee, C. K. (1995). Elastic justification: How Gilbert, D. T., & Malone, P. S. (1995). The cor- tempting but task-irrelevant factors influence respondence bias. Psychological Bulletin, 117, decisions. Organizational Behavior and Human 21–38. Decision Processes, 62, 330–337. P1: GFZ 0521824176c13.xml CB798B/Holyoak 0521824176 October 31, 2004 6:5

motivated thinking 315

Josephs, R. A., & Hahn, E. D. (1995). Bias and ac- for attributional functions. Journal of Personal- curacy in estimates of task duration. Organiza- ity and Social Psychology, 80, 5–18. tional Behavior and Human Decision Processes, Lord, C. G., Ross, L., & Lepper, M. R. (1979). 61 , 202–213. Biased assimilation and attitude polarizations: Kelley, H. H. (1973). The process of causal attri- The effects of prior theories on subsequently bution. American Psychologist, 28, 107–128. considered evidence. Journal of Personality and Klein, W. M., & Kunda, Z. (1992). Motivated per- Social Psychology, 37, 2098–2109. son perception: Constructing justifications for Lundgren, S. R., & Prislin, R. (1998). Motivated desired beliefs. Journal of Experimental Social cognitive processing and attitude change. Per- Psychology, 28, 145–168. sonality and Social Psychology Bulletin, 24, 715– Kruglanski, A. W. (1996). Motivated social 726. cognition: Principles of the interface. In E. T. Maheswaran, D., & Chaiken, S. (1991). Promot- Higgins & A. W. Kruglanski (Eds.), So- ing systematic processing in low-motivation cial psychology: Handbook of basic principles settings: Effect of incongruent information on (pp. 493–520). New York: Guilford. processing and judgment. Journal of Personality Kruglanski, A. W. (1999). Motivation, cognition, and Social Psychology, 61 , 13–25. and reality: Three memos for the next gen- Martin, L., & Clore, G. C. (Eds.). (2001). Theories eration of research. Psychological Inquiry, 10, of mood and cognition. Mahwah, NJ: Erlbaum. 58. McDonald, H. E., & Hirt, E. R. (1997). When ex- Kruglanski, A. W., & Freund, T. (1983). The pectancy meets desire: Motivational effects in freezing and unfreezing of lay inferences: Ef- reconstructive memory. Journal of Personality fects on impression primacy, ethnic stereotyp- and Social Psychology, 72, 5–23. ing and numerical anchoring. Journal of Exper- McGinnies, E. (1949). Emotionality and percep- imental Social Psychology, 19, 448–468. tual defense. Psychological Review, 56, 244–251. Kruglanski, A. W., & Webster, D. M. (1996). Miller, D. T. (1976). Ego involvement and attri- Motivated closing of the mind: “Seizing” and butions for success and failure. Journal of Per- “freezing.” Psychological Review, 103, 263–283. sonality and Social Psychology, 34, 901–906. Kruglanski, A. W., Webster, D. M., & Klem, A. Molden, D. C., & Higgins, E. T. (2004). Preferred (1993). Motivated resistance and openness to strategies for self-attribution: Motivated inference persuasion in the presence of absence of prior beyond self-serving biases. Manuscript under re- information. Journal of Personality and Social view. 861 876 Psychology, 65, – . Molden, D. C., & Higgins, E. T. (2004). Cate- Kunda, Z. (1987). Motivated inference: Self- gorization under uncertainty: Resolving vague- serving generation and evaluation of causal the- ness and ambiguity with eager versus vigilant ories. Journal of Personality and Social Psychol- strategies. Social Cognition, 22, 248–277. 636 647 ogy, 53, – . Moskowitz, G. B. (1993). Individual differences Kunda, Z. (1990). The case for motivated reason- in social categorization: The influence of per- ing. Psychological Bulletin, 108, 480–498. sonal need for structure on spontaneous trait Kunda, Z., & Santioso, R. (1989). Motivated inferences. Journal of Personality & Social Psy- change in the self-concept. Journal of Experi- chology, 65, 132–142. mental Social Psychology, 25, 272–285. Moskowitz, G. B., Gollwitzer, P. M., Wasel, W., Lerner, J. S., & Tetlock, P. E. (1999). Accounting & Schaal, B. (1999). Preconscious control of for the effects of accountability. Psychological stereotype activation through chronic egalitar- Bulletin, 125, 255–275. ian goals. Journal of Personality and Social Psy- Lewin, K. (1935). A dynamic theory of personality. chology, 77, 167–184. New York: McGraw-Hill. Murray, H. A. (1938). Explorations in personality. Liberman, A., & Chaiken, S. (1992). Defensive New York: Oxford University Press. processing of personally relevant health mes- Murray, S. L. (1999). The quest for conviction: sages. Personality and Social Psychology Bulletin, Motivated cognition in romantic relationships. 18, 669–679. Psychological Inquiry, 10, 23–34. Liberman, N., Molden, D. C., Idson, L. C., & Hig- Neuberg, S. L. (1989). The goal of forming accu- gins, E. T. (2001). Promotion and prevention rate impressions during social interactions: At- focus on alternative hypotheses: Implications tenuating the impact of negative expectancies. P1: GFZ 0521824176c13.xml CB798B/Holyoak 0521824176 October 31, 2004 6:5

316 the cambridge handbook of thinking and reasoning

Journal of Personality and Social Psychology, 56, Santioso, R., Kunda, Z., & Fong, G. T.(1990). Mo- 374–386. tivated recruitment of autobiographical memo- Neuberg, S. L., & Fiske, S. T. (1987). Motiva- ries. Journal of Personality and Social Psychology, tional influences on impression formation: De- 59, 229–241. pendency, accuracy-driven attention, and in- Schuette, R. A., & Fazio, R. H. (1995). Attitude dividuating information. Journal of Personality accessibility and motivation as determinants of and Social Psychology, 53, 431–444. biased processing: A test of the MODE model. Nienhuis, A. E., Manstead, A. S. R., & Spears, Personality and Social Psychology Bulletin, 21, R. (2001). Multiple motives and persuasive 704–710. communication: Creative elaboration as a re- Schwarz, N., & Clore, G. L. (1996). Feelings sult of impression motivation and accuracy mo- and phenomenal experiences. In E. T. Higgins tivation. Personality and Social Psychology Bul- &A.W.Kruglanski (Eds.), Social psychology: letin, 27, 118–132. Handbook of basic principles (pp. 433–465). Nisbett, R. E., & Ross, L. (1980). Human New York: Guilford. inference: Strategies and shortcomings of so- Sinclair, L., & Kunda, Z. (1999). Reactions to cial judgment. Englewood Cliffs, NJ: Prentice- a black professional: Motivated inhibition and Hall. activation and conflicting stereotypes. Journal Pelham, B. W., & Neter, E. (1995). The effect of of Personality and Social Psychology, 77, 885– motivation on judgment depends on the dif- 904. ficulty of the judgment. Journal of Personality Sorrentino, R. M., & Higgins, E. T. (Eds.). (1986). and Social Psychology, 68, 581–594. Handbook of motivation and cognition, vols. 1 –3. Pendry, L. F., & Macrae, C. N. (1994). Stereotypes New York: Guilford. and mental life: The case of the motivated but Spencer, S. J., Fein, S., Wolfe, C., Hodgson, H. L., thwarted tactician. Journal of Experimental So- & Dunn, M. A. (1998). Stereotype activation cial Psychology, 30, 303–325. under cognitive load: The moderating role of Petty, R. E., & Wegener, D. T. (1999). The self-image threat. Personality and Social Psy- elaboration likelihood model: Current status chology Bulletin, 24, 1139–1152. and controversies. In S. Chaiken & Y. Trope Srull, T. K., Lichtenstein, M., & Rothbart, M. (Eds.), Dual-process theories in social psychology (1985). Associative storage and retrieval pro- (pp. 41–72). New York: Guilford. cesses in person memory. Journal of Experimen- Pyszczynski, T., & Greenberg, J. (1987). Toward tal Psychology: Learning, Memory, and Cogni- an integration of cognitive and motivational tion, 11, 316–345. perspectives on social inference: A biased hy- Stephan, W. F., & Gollwitzer, P. (1981). Affect as pothesis testing model. In L. Berokowitz (Ed.), a mediator of attributional egotism. Journal of Advances in experimental social psychology Experimental Social Psychology, 17, 443–458. (Vol. 20,pp.297–340). New York: Academic Tetlock, P. E. (1983). Accountability and com- Press. plexity of thought. Journal of Personality and Roese, N. J. (1997). Counterfactual thinking. Psy- Social Psychology, 45, 74–83. chological Bulletin, 121, 133–148. Tetlock, P. E. (1985). Accountability: A social Roese, N. J., Hur, T., & Pennington, G. L. (1999). check on the fundamental attribution error. So- Counterfactual thinking and regulatory focus: cial Psychology Quarterly, 48, 227–236. Implications for action versus inaction and suf- Tetlock, P. (1998). Close-call counterfactuals and ficiency versus necessity. Journal of Personality belief-system defense: I was not almost wrong and Social Psychology, 77, 1109–1120. but I was almost right. Journal of Personality and Ruscher, J. B., Fiske, S. T., & Schnake, S. B. Social Psychology, 75, 639–652. (2000). The motivated tactician’s juggling act: Tetlock, P. E., & Boettger, R. (1989). Account- Compatible vs. incompatible impression goals. ability: A social magnifier of the dilution effect. British Journal of Social Psychology, 39, 241– Journal of Personality and Social Psychology, 57, 256. 388–398. Sanbonmatsu, D. M., & Fazio, R. H. (1990). Tetlock, P. E., & Levi, A. (1982). Attribution The role of attitudes in memory-based deci- bias: On the inconclusiveness of the cognition- sion making. Journal of Personality and Social motivation debate. Journal of Experimental So- Psychology, 59, 614–622. cial Psychology, 18, 68–88. P1: GFZ 0521824176c13.xml CB798B/Holyoak 0521824176 October 31, 2004 6:5

motivated thinking 317

Thompson, E. P., Roman, R. J., Moskowitz, G. B., nal of Personality and Social Psychology, 65, 261– Chaiken, S., & Bargh, J. A. (1994). Systematic 271. processing and the debasing of covert primacy Verplanken, B. (1993). Need for cognition and effects in impression formation: Unshackling external information search: Responses to time the motivated perceiver from constraints of ac- pressure during decision-making. Journal of Re- cessibility. Journal of Personality and Social Psy- search in Personality, 27, 238–252. 474 489 chology, 66, – . Zelaznik, H. N., Mone, S., McCabe, G. P., & Tyler, T. R., & Blader, S. L. (2000). Cooperation in Thaman, C. (1988). Role of temporal and spa- groups: Procedural justice, social identity, and be- tial precision in determining the nature of havioral engagement. Philadelphia: Psychology the speed-accuracy trade-off in aimed-hand Press. movement. Journal of Experimental Psychology: Webster, D. M. (1993). Motivated augmentation Human Perception and Performance, 14, 221– and reduction of the over-attribution bias. Jour- 230. P1: GFZ 0521824176c13.xml CB798B/Holyoak 0521824176 October 31, 2004 6:5

318