Running Head: Need for Cognition and Authoritarianism On Being a Mindful Authoritarian: Is Need for Cognition Always Associated with Greater Tolerance of Deviant Groups?

In much research, the motivation to be mindful is expected to result in greater tolerance of deviant groups. Contrary to this expectation, the present research shows that this is the case only when individuals are relatively low in authoritarianism. Across two studies, we found a robust interaction effect of authoritarianism (right-wing authoritarianism, social dominance orientation) and need for cognition on several measures of intolerance against criminals. Among low authoritarians, need for cognition was negatively associated with punitiveness and dispositional attribution of crimes, and positively with support for rehabilitation of criminals. However, among high authoritarians, a reverse pattern was observed. These findings are discussed in the context of some recent advances in theories of social motivation and functionalist mindset. Need for Cognition and Authoritarianism -3 On Being a Mindful Authoritarian: Is Need for Cognition Always Associated with Greater Tolerance of Deviant Groups? In the literature of social psychology it is often assumed that extreme social judgments (e.g., intolerance towards deviant and stigmatized groups) can be mitigated or moderated when people are motivated to become mindful (e.g., Devine & Monteith, 1999; Stangor, & Thompson, 2002). The current study challenges this assumption; it seeks to show that need for cognition, as an individual difference in the motivation to be mindful, may even strengthen the extreme judgments among people strongly committed to authoritarian ideology, possibly because mindful authoritarians have developed more elaborate ideological structures to support these judgments. The need for cognition (NF) refers to a dispositional tendency to engage in and enjoy thinking (Cacioppo & Petty, 1982). There is consistent evidence that it moderates biased and extreme social judgments. For example, the Need for Cognition Scale is associated with lower levels of dogmatism (Cacioppo & Petty, 1982), punitiveness (Sargent, 2004), racial prejudice (Crawford, Skowronski, & Walker, 1995, cf. Crawford & Skowronski, 1998; Hogan & Mallot, 2005; Waller, 1993), and ingroup favoritism (Stangor & Thompson, 2002). This is the case probably because individuals with higher NF are more attributionally complex, and tend not to form social judgments on the basis of simplistic cognitive structures such as group stereotypes. Need for Cognition and Authoritarianism -4 Contrarily, people with low NF have been described as “chronic cognitive misers” (Cacioppo, Petty, Feinstein, & Jarvis, 1996, p. 197). In short, strengthening the need for cognitive complexity and weakening the one for cognitive economy are widely believed to have the beneficial effect of de-biasing or moderating extreme social judgments (e.g., Van Hiel, 2004). This tendency to equate the motivation to be mindful and the motivation to be accurate and impartial reflects a major theme in much social cognition research: Thinking de-biases judgments. However, some recent developments in the social motivation literature have called for a critical re-examination of this assumption. First, Jost and his colleagues (Jost, Glaser, Kruglanski, & Sulloway, 2003) have proposed a distinction between directional motives and non-directional motives. The former refers to a desire for a specific conclusion (e.g., self-enhancement), and the latter a desire to arrive at a conclusion independent of its content. NF, defined as a need to think about issues that a person confronts, is a non-directional motive. As such, it increases depth of information processing but does not guarantee impartial or moderate judgments. To elaborate, NF incites elaborate thoughts on the confronting issue; it motivates the perceiver to consider arguments that are subjectively perceived to be pertinent and valid. However, what is seen as pertinent and valid depends in part on the perceiver’s chronic functionalist mindset. The idea that deep thoughts result in impartial judgments rests on the Need for Cognition and Authoritarianism -5 assumption that perceivers adopt the mindset of an intuitive scientist, whose thought processes are directed to form valid judgments based on objective evidence (Rucker, Polifroni, Tetlock, & Scott, 2004). However, the perceiver may have other functionalist mindsets, which direct them to emulate the information processing goal of a politician, prosecutor, or theologian (Tetlock, 2002). Of particular relevance to the current study is the functionalist mindset of an intuitive prosecutor. Intuitive prosecutors are people who “seek to defend rules and regimes that they endow with legitimacy” (Tetlock, 2002: p.461). They are motivated by a need to see norm violators atone for what they have done. When people adopt the mindset of lay prosecutors, NF may not moderate the tendency to punish social deviants. Instead, sometimes, more elaborate thoughts could lead information processing further down the alley of finding legitimate reasons for enforcing the status quo and sanctioning deviant behaviors. THE PRESENT RESEARCH The present research tested the above possibility by examining the joint effects of NF and authoritarianism on intolerance against a deviant group, namely, criminals. Authoritarianism constitutes an interesting variable because of its relevance to many social issues, in particular dogmatic intolerance of deviant groups (e.g., Altemeyer, 1996; Peterson, Doty, & Winter, 1993). For example, it has been found to be positively related to punitiveness (e.g., Lerner, Goldberg, Need for Cognition and Authoritarianism -6 Tetlock, 1998). Additionally, authoritarians often act like intuitive prosecutors. Like intuitive prosecutors, authoritarians are concerned about submission to and protection of the established and legitimate norms in a society. They may aggressively punish deviants and regulate others’ behaviors with coercive means (Altemeyer, 1996). The resemblance of authoritarians to intuitive prosecutors is suggested in the following characterization of intuitive prosecutors: “Individuals whose goal is specifically to uphold the social order could be labeled intuitive prosecutors who are upset by and want to punish wrongdoers” (Goldberg, Lerner, & Tetlock, 1999: p.790). Based on the above analysis, we hypothesized that among individuals who disagree with authoritarian ideology, there would be a negative relationship between NF and intolerant attitudes against deviant criminals, as previous research has found. Contrarily, among authoritarians, NF may not be inversely related to the tendency to display these intolerant attitudes. On the contrary, there may be a positive relationship between them. STUDY 1 Method The participants were 137 (43 males, 93 females, 1 did not report) undergraduates recruited from a US public university. Their age ranged from 17 to 23 (M = 19.49, SD = 1.20). They participated in exchange for course requirement credits. To obscure the purpose of the study, the measures of authoritarianism and NF were put Need for Cognition and Authoritarianism -7 together with some filler measures and presented to the participants as a personality study. The dependent variables were included in another questionnaire package, and were mixed with other measures included in the package. Measures Right-wing authoritarianism (RWA). Authoritarianism was measured by the Right-Wing Authoritarianism Scale (Altemeyer, 1996) which comprises 34 items. We excluded the first four items from the present study for two reasons. First, according to the original design of the scale, the first four items are not to be scored. Second, one of these four items (“Life imprisonment is justified for certain crimes”) overlapped with the dependent variables. They were excluded so as to avoid demand characteristics. Participants indicated their agreement with each of the remaining 30 items on a 9-point scale (from -4 “strongly disagree” to +4 “strongly agree”). Two sample items in the scale are “The real keys to the ‘good life’ are obedience, discipline, and sticking to the straight and narrow” and “Our country needs free thinkers who will have the courage to defy traditional ways, even if this upsets many people.” Need for Cognition and Authoritarianism -8 Following the standard scoring procedures, -4 was scored as 1, +4 as 9, and the neutral answer 0 as 5. The right-wing authoritarianism score was the sum of the recoded scores on the 30 items, with the lowest possible score of 30, and the highest possible score of 270. A higher score indicates stronger authoritarianism. The scale had high reliability in this study:  = .92. Need for cognition. Participants indicated how well each of the 18 items on the Need for Cognition Scale (Cacioppo, Petty, & Kao, 1984) described themselves on a 5-point scale (“1” extremely unlike me to “5” extremely like me). Two sample items of the scale are “I only think as hard as I have to” and “The notion of thinking abstractly is appealing to me.” The scale was highly reliable in this study:  = .89. Dependent measures. In this study, two dependent measures were included to assess participants’ intolerance against criminals. The first measure assessed punitiveness towards criminals. Specifically, it measured agreement with death penalty and a retribution orientation in punishing criminals. It consisted of five items: “In general, do you approve or disapprove of the death penalty?” “Do you generally favor or oppose the death penalty in cases where people are convicted of first-degree murder?” “Death penalty is cruel and inhumane.” (reverse-coded) “Criminals should be punished to make the criminals suffer, as the victims of the crimes suffered.” and “Criminals should be punished to make the criminals pay for their crimes.” Need for Cognition and Authoritarianism -9 Participants responded to each of these items on a 7-point scale, with a higher number indicating greater punitiveness. The internal reliability of the five items was .86. Another dependent measure assessed participants’ support for rehabilitating criminals. Specifically, it measured support for rehabilitation programs in prisons and a rehabilitation orientation in punishing criminals. It consisted of four items: “Do you think rehabilitation programs (including education, vocational training, and psychological counseling, etc.) in prisons are helpful for criminals?” “In general do you approve or disapprove of expanding rehabilitation programs in prisons?” “Criminals should be punished to educate the criminals to lead a law-abiding life.” and “Criminals should be punished so as to rehabilitate them.” Participants responded to each of these items on a 7-point scale, with a higher number indicating greater support for rehabilitation. The internal reliability of the four items was .75. Results Punitiveness RWA (mean-centered), NF (mean-centered), and their interaction were included as predictors in a regression model to predict punitiveness. The main effect of RWA was significant, β = 0.32, t(133) = 3.77, p < .001. As in previous research, participants who adhered more strongly to right-wing ideology were more punitive towards criminals. Consistent with our prediction, the interaction of RWA and NF was significant, β = 0.20, t(133) = 2.44, p < .05. Need for Cognition and Authoritarianism -10 To understand this interaction, we performed a simple slope analysis following Aiken and West’s suggestions (1991). As shown in Figure.1, when RWA was low (at -1SD), NF was negatively related to punitiveness; but when RWA was high (at +1SD), NF was positively related to punitiveness. We also followed Preacher, Curran and Bauer’s procedures (2003) to test the region of significance. As show in Figure 2, when RWA was low, the simple slope of NF predicting punitiveness was negative and significantly smaller than 0; but when RWA was high, the slope became positively and significantly larger than 0. This is consistent with our hypothesis. Support for rehabilitation We did a similar analysis on support for rehabilitation, the other dependent measure. The main effects of RWA and NF were not significant. Consistent with our prediction, the interaction of RWA and NF was significant, β = -0.21, t(133) = -2.49, p < .05. To understand this interaction, we performed a simple slope analysis. As shown in Figure 3, when RWA was low, NF was positively related to support for rehabilitation; but when RWA was high, NF was negatively related to support for rehabilitation. As show in Figure 4, when RWA was low, the simple slope of NF predicting punitiveness was significantly larger than 0; but when RWA was high, the slope became smaller than 0, though not significant. This is consistent with our hypothesis. Need for Cognition and Authoritarianism -11 To summarize, NF was negatively associated with punitiveness and positively associated with support for rehabilitation, but only among the low authoritarians (operationalized as RWA). By contrast, among high authoritarians, NF was positively related to punitiveness and negatively related to support for rehabilitation. STUDY 2 The objective of Study 2 was to conceptually replicate the findings in Study 1. There were two major changes. First, we added a new dependent measure into Study 2. It assessed participants’ tendency to make dispositional attributions of crimes. Second, another dimension of authoritarianism was measured. Authoritarianism has been defined as the covariation of conventionalism, authoritarian submission, and authoritarian aggression; what the RWA scale typically measures refers to authoritarian submission (Altemeyer, 1998). Social Dominance Orientation (SDO) has been defined as “a general attitudinal orientation toward intergroup relations, reflecting whether one generally prefers such relations to be equal, versus hierarchical” (Pratto, Sidanius, Stallworth, & Malle, 1994, p. 742). Recently, researchers argue that SDO constitutes another dimension of authoritarianism, namely, authoritarian dominance (Altemeyer, 1998; Caoos, 2002). Also, there has been evidence that both RWA and SDO predict intolerance against criminals (Van Hiel & Kossowska, 2006). In all, Study 2 purported to replicate Study 1’s findings with a different measure of Need for Cognition and Authoritarianism -12 authoritarianism and a new measure of intolerance. If the findings in Study 1 were indeed replicated, we could have a higher confidence in the robustness of the hypothesized effect. Method The participants were 102 (52 males, 49 females, 1 did not report) undergraduates recruited from a US public university. Their age ranged from 17 to 27 (M = 19.25, SD = 1.35). They participated in exchange for course requirement credits. The procedures were similar to Study 1, except the changes as mentioned. Measures SDO. The Social Dominance Orientation Scale (Pratto et al,, 1994) comprises 16 items. Participants indicated their attitudes towards 16 items on a 7-point scale (from 1 “very negative” to 7 “very positive”). Two sample items in the scale are “Some groups of people are simply inferior to other groups”, and “Sometimes other groups must be kept in their place.” The social dominance orientation score was the average of the scores on the 16 items, with a higher score indicating stronger social dominance orientation. The scale had high reliability in the current study:  = .94. Need for cognition. The scale was highly reliable in the current study:  = .89. Dependent measures. The internal reliability of the five-item punitiveness measure was .84, and that of the four-item support for rehabilitation measure was .68. Need for Cognition and Authoritarianism -13 The new dependent measure of dispositional attribution of crimes consisted of five items: “Criminals commit crimes because they have bad character”, “Most criminals commit crimes because they are basically selfish people, unconcerned about the feelings of other people”, “Most criminals deliberately choose to prey on society”, “Most criminals commit crimes because they were just born to be criminals”, and “Most criminals know fully well what they are doing when they break the law.” Participants responded to each of these items on a 7-point scale, with a higher number indicating stronger dispositional attribution of crimes. The internal reliability of the five items was .70. Results Punitiveness SDO (mean-centered), NF (mean-centered), and their interaction were included as predictors in a regression model to predict punitiveness. The main effect of SDO was significant, β = 0.41, t(98) = 4.16, p < .001. As in previous research, participants who had a higher SDO were more punitive towards criminals. Consistent with our prediction, the interaction of SDO and NF was significant, β = 0.21, t(133) = 2.19, p < .05. To understand this interaction, we performed a simple slope analysis. As shown in Figure 5, when SDO was low, NF was negatively related to punitiveness; but when SDO was high, NF was positively related to punitiveness. As shown in Figure 6, when SDO was low, the simple Need for Cognition and Authoritarianism -14 slope of NF predicting punitiveness was significantly smaller than 0; but when SDO was high, the slope became significantly larger than 0. This is consistent with our hypothesis. Support for rehabilitation We did a similar analysis on support for rehabilitation. The main effect of SDO was significant, β = -0.45, t(98) = -4.58, p < .001. The predicted interaction of SDO and NF was not significant however, β = -0.14, t(98) = -1.53, p = .13. We did perform a simple slope analysis despite the non-significance of the effect. The predicted pattern was observed. As shown in Figure 7, when SDO was low, NF was positively related to support for rehabilitation; but when SDO was high, NF was slightly negatively related to support for rehabilitation. This pattern, though statistically non-significant, is consistent with our hypothesis. Dispositional attribution We did a similar analysis on dispositional attribution of crimes. The main effect of SDO was significant, β = 0.41, t(98) = 3.95, p < .001. The predicted interaction of SDO and NF was also significant, β = 0.20, t(98) = 2.04, p < .05. To understand this interaction, we performed a simple slope analysis. As shown in Figure 8, when SDO was low, NF was negatively related to dispositional of crimes; but when SDO was high, NF was positively related to dispositional attribution of crimes. As shown in Figure 9, when SDO was low, the simple slope of NF predicting dispositional attribution of crimes was Need for Cognition and Authoritarianism -15 significantly smaller than 0; but when SDO was high, the slope became significantly larger than 0. This is again consistent with our hypothesis. To summarize, NF was negatively associated with punitiveness and dispositional attribution of crimes, but only among the low authoritarians (operationalized as SDO). By contrast, among high authoritarians, NF was positively related to punitiveness and dispositional attribution of crimes. A reverse, though statistically non-significant, pattern was observed in participants’ support for rehabilitation of criminals. Discussion As mentioned, the assumption that thinking de-biases social judgments is commonly held in much social psychology research. For example, need for complex thoughts (e.g., NF) has been shown to moderate extreme judgments of deviant and stigmatized groups (e.g., Sargent, 2004; Stangor & Thompson, 2002). Contrary to this common view, the present research showed that NF led to greater tolerance of criminals only among low authoritarians; among high authoritarians, NF was associated with even greater intolerance against criminals. This finding is consistent with Jost et al’s (2003) classification of NF and other similar epistemic motives as non-directional motives. These motives represent a desire to arrive at a conclusion independent of what the judgment is about. As such, although NF tends to deepen information processing, it does not guarantee greater tolerance of deviant groups. Need for Cognition and Authoritarianism -16 Our findings also echo Tetlock’s (2002) call for attention to other possible social functionalist mindsets for understanding judgments and decision-making. Tetlock (2002) argued that research on judgment and choice has been dominated by the assumptions that people either function like intuitive scientists in search for accuracy and truth, or like intuitive economists in search for self-interest. He proposed three other possible cognitive mindsets, namely, the mindset of a politician, theologian, and prosecutor. He further pointed out that judgments or choices considered as biases and errors within an intuitive scientist or intuitive economist mindset may seem rational within an intuitive politician, prosecutor, or theologian mindset. As indicated in our findings, authoritarians appear to have a spontaneous tendency to evoke an intuitive prosecutor’s mindset that favors enforcement of conventional morality and sanctioning of deviant groups. Within this mindset, complex thoughts do not necessarily undermine the perceived validity of intolerant attitudes towards deviant groups. In fact, deeper information processing may even push authoritarians to form sophisticated arguments to justify their intolerance against these groups. Although the findings are generally supportive of our idea, future research is needed to establish its generality by substituting authoritarianism with other ideological variables (e.g., conservatism, just world belief, Protestant work ethic), and NF with other nonspecific epistemic needs (e.g., need for cognitive closure, need for structure). Also, attitudes towards various social Need for Cognition and Authoritarianism -17 groups (e.g., ingroup, minority groups, stigmatized groups) can be measured instead. In addition,experimental studies that induce a particular functionalist mindset and manipulate an epistemicneed simultaneously would provide evidence for the causal impact of the interaction offunctionalist mindsets and non-specific information processing motives on tolerance of deviantgroups, supplementing the correlational findings obtained in the present research. Finally, on-linethought processes could be measured in these experiments using the thought-listing technique totrack how deeper information processing fosters generation of more ideology-consistent thoughts,which should mediate the Mindset X Epistemic Need interaction on judgments. In summary, wehope the present research can stimulate future works on the intricate relationship betweenideology, information processing motives, and attitudes towards different social groups.ReferencesAiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpretinginteractions. Thousand Oaks: Sage.Altemeyer, B. (1996). The authoritarian specter. Cambridge, Massachusetts: HarvardUniversity Press.Cacioppo, J.T., & Petty, R.E. (1982). The need for cognition. Journal of Personality andSocial Psychology, 42, 116-131.Cacioppo, J.T., Petty, R.E., & Kao, C.F. (1984). The efficient assessment of need for Need for Cognition and Authoritarianism -18 cognition. Journal of Personality Assessment, 48, 306-307.Cacioppo, J.T., Petty, R.E., Feinstein, J.A., & Jarvis, W.B.G. (1996). Dispositionaldifferences in cognitive motivation: The life and times of individuals varying in need forcognition. Psychological Bulletin, 119, 197-253.Capps, J.S. (2002). Explaining punitiveness : Right-wing authoritarianism and socialdominance. North Americans Journal of Psychology, 4, 263-278.Crawford, M.T., & Skowronski, J.J. (1998). When motivated thought leads to heightenedbias: high need for cognition can enhance the impact of stereotypes on memory. Personality andSocial Psychology Bulletin, 24, 1075-1088.Devine, P.G., & Monteith, M.J. (1999). Automaticity and control in stereotyping. InShelly Chaiken & Yaacov Trope (Eds.), Dual-process theories in social psychology (pp.339-360). New York, NY, US: Guilford Press.Goldberg, J.H., Lerner, J.S., & Tetlock, P.E. (1999). Rage and reason: The psychology ofthe intuitive prosecutor. European Journal of Social Psychology, 29, 781-795.Hogan, D.E., & Mallot, M. (2005). Changing racial prejudice through diversity education.Journal of College Student Development, 46, 115-125.Hong, Y-Y., Coleman, J., Chan, G., Wong, R.Y.M., et al. (2004). Predicting intergroupbias: The interactive effects of implicit theory and social identity. Personality and Social Need for Cognition and Authoritarianism -19 Psychology Bulletin, 30, 1035-1047.Jost, J.T., Glaser, J., Kruglanski, A.W., Sulloway, F.J. (2003). Political conservatism asmotivated social cognition. Psychological Bulletin, 129, 339-375.Lerner, J.S., Goldberg, J.H., & Tetlock, P.E. (1998). Sober second thought: The effects ofaccountability, anger, and authoritarianism on attributions of responsibility. Personality andSocial Psychology Bulletin, 24, 563-574.Maass, A., Castelli, L., & Arcuri, L. (2000). Measuring prejudice: Implicit versus explicittechniques. In D. Capozza & R. Brown (Eds.), Social identity processes: Trends in theory andresearch (pp. 96-116). London: Sage Ltd.Peterson, B.E., Doty, R.M., & Winter, D.G. (1993). Authoritarianism and attitudestowards contemporary social issues. Personality and Social Psychology Bulletin, 19, 174-184.Pratto, F., Sidanius, J., Stallworth, L., &Malle;, B. (1994). Social dominance orientation:A personality variable predicting social and political attitudes. Journal of Personality and SocialPsychology, 67, 741-763.Preacher, K.J., Curran, P.J., & Bauer, D.J. (2003). Simple intercepts, simple slopes, andregions of significance in MLR 2-way interactions. From http://www.unc.edu/~preacher/interact/Rucker, D.D., Polifroni, M., Tetlock, P.E., & Scott, A.L. (2004). On the assignment ofpunishment: The impact of general-societal threat and the moderating role of severity. Need for Cognition and Authoritarianism -20 Personality and Social Psychology Bulletin, 30, 673-684.Sargent, M.J. (2004). Less thought, more punishment: Need for cognition predictssupport for punitive responses to crime. Personality and Social Psychology Bulletin, 30,1485-1493.Stangor, C., & Thompson, E.P. (2002). Needs for cognitive economy andself-enhancement as unique predictors of intergroup attitudes. European Journal of SocialPsychology, 32, 563-575.Tetlock, P.E. (2002). Social functionalist frameworks for judgment and choice: Intuitivepoliticians, theologians, and prosecutors. Psychological Review, 109, 451-471.Van Hiel, A., & Kossowska, M. (2006). Having few positive emotions, or too manynegative feelings? Emotions as moderating variables of authoritarianism effects of racism.Personality and Individual Differences, 40, 919-930.Van Hiel, A., Pandelaere, M., & Duriez, B. (2004). The impact of need for closure onconservative beliefs and racism: Differential mediation by authoritarian submission andauthoritarian dominance. Personality and Social Psychology Bulletin, 30, 834-837.Waller, J. (1993). Correlation of need for cognition and modern racism. PsychologicalReports, 73, 542. Need for Cognition and Authoritarianism -21 Figure 1. Interaction between Need for Cognition (NF) and Right-wing Authoritarianism(RWA) in predicting punitiveness (Study 1)Note. NF and RWA were mean-centered. Need for Cognition and Authoritarianism -22 Figure 2. Simple slope of Need for Cognition (NF) predicting punitiveness as a function ofRight-wing Authoritarianism (RWA) (Study 1)Note. NF and RWA were mean-centered. The middle straight line represents the estimated simpleslope of NF predicting punitiveness at different values of RWA. The two curved lines representthe 95% confidence interval of the estimate. The estimated simple slope is significantly differentfrom 0 ( = .05) at values of RWA that are less than the left vertical dotted line and greater thanthe right one. Need for Cognition and Authoritarianism -23 Figure 3. Interaction between Need for Cognition (NF) and Right-wing Authoritarianism(RWA) in predicting support for rehabilitation (Study 1)Note. NF and RWA were mean-centered. Need for Cognition and Authoritarianism -24 Figure 4. Simple slope of Need for Cognition (NF) predicting punitiveness as a function ofRight-wing Authoritarianism (RWA) (Study 1)Note. NF and RWA were mean-centered. The middle straight line represents the estimated simpleslope of NF predicting support for rehabilitation at different values of RWA. The two curved linesrepresent the 95% confidence interval of the estimate. The estimated simple slope is significantlydifferent from 0 ( = .05) at values of RWA that are less than the vertical dotted line. Need for Cognition and Authoritarianism -25 Figure 5. Interaction between Need for Cognition (NF) and Social Dominance Orientation(SDO) in predicting punitiveness (Study 2)Note. NF and SDO were mean-centered. Need for Cognition and Authoritarianism -26 Figure 6. Simple slope of Need for Cognition (NF) predicting punitiveness as a function ofSocial Dominance Orientation (SDO) (Study 2)Note. NF and SDO were mean-centered. The middle straight line represents the estimated simpleslope of NF predicting punitiveness at different values of SDO. The two curved lines representthe 95% confidence interval of the estimate. The estimated simple slope is significantly differentfrom 0 ( = .05) at values of SDO that are less than the left vertical dotted line and greater thanthe right one. Need for Cognition and Authoritarianism -27 Figure 7. Interaction (non-significant) between Need for Cognition (NF) and SocialDominance Orientation (SDO) in predicting support for rehabilitation (Study 2)Note. NF and SDO were mean-centered. Need for Cognition and Authoritarianism -28 Figure 8. Interaction between Need for Cognition (NF) and Social Dominance Orientation(SDO) in predicting dispositional attribution of crimes (Study 2)Note. NF and SDO were mean-centered. Need for Cognition and Authoritarianism -29 Figure 9. Simple slope of Need for Cognition (NF) predicting dispositional attribution ofcrimes as a function of Social Dominance Orientation (SDO) (Study 2)Note. NF and SDO were mean-centered. The middle straight line represents the estimated simpleslope of NF predicting dispositional attribution at different values of SDO. The two curved linesrepresent the 95% confidence interval of the estimate. The estimated simple slope is significantlydifferent from 0 ( = .05) at values of SDO that are less than the left vertical dotted line andgreater than the right one.