The current research integrates theories of morality and social identity in order to identify the ways in which moralization of social groups can produce hostile intergroup relations. Two studies investigated two distinct modes of moralization: the creation of entirely new morality-based social groups and the moralization of preexisting groups. Study 1 tested the effects of moralizing the behavior of texting while driving on attitudes, emotions, and beliefs about individuals who either frequently text while driving (the outgroup) or those who do not (the ingroup). Results indicated that moralization alone (perceiving harm from the outgroup) was sufficient to strengthen social identification with non-frequent texters, and framing the behavior in terms of two distinct groups had no additional impact. The lines between ingroup favoritism and outgroup derogation were also blurred in the moralization condition. Identification was directly related to outgroup negativity for the moralized group, and both positive ingroup-directed emotions and negative outgroup-directed emotions uniquely predicted attitudes toward the moralized ingroup. In Study 2, moralization through an incidental manipulation (thinking of a friend's moral/immoral behavior) led to automatic outgroup derogation against Black Americans by White Americans who felt strong positive emotions toward their ingroup. Those in the two control conditions (sadness and neutral) showed no relationship between positive ingroup emotions and outgroup derogation. Moreover, the total amount of implicit bias observed was entirely a function of the association between concepts of "Black" and "Bad" for the moralization conditions. However, the association between "White" and "Good" predicted implicit bias in the sadness and control conditions. These results suggest moralization can be an automatic process that can happen without awareness, and that outgroup derogation can be the primary motivation for intergroup bias. Together, these studies provide evidence that morality has a strong influence on intergroup relations, and this influence extends beyond groups defined by moral conviction. Specifically, moralization leads to an increased motive for outgroup derogation, a reduced motive for ingroup favoritism, and this can happen regardless of whether moral judgments are made about an outgroup specifically (Study 1), or whether the moral judgment is incidental to the target outgroup (Study 2).

Four studies explore the relationship between power--control over others' outcomes--and impression formation. Participants in each study occupied different power roles: the powerful (managers) controlled others' outcomes, the powerless (employees) were contingent on the powerful for outcomes, and the power-irrelevant (alternate participants) neither controlled the powerless nor were contingent on the powerful. Dependent measures included attention to target trait information and impression ratings. Power-irrelevant targets served as the experimental control condition. Participants were predicted to ignore both stereotype-consistent and stereotype-inconsistent information about these targets, forming moderately stereotypic but relatively less confident impressions. Powerful participants were predicted to stereotype subordinates by default (ignoring stereotype-inconsistent information) and by design (effortfully attending to stereotype-consistent information). As a result, powerholders' impressions of subordinates were predicted to be most stereotypic and most confident. In contrast, the powerless were predicted to individuate the powerful, effortfully attending to stereotype-inconsistent information and forming less stereotypic impressions, relative to the other groups. The results of Study 1 support the hypotheses regarding attention to trait information; attention to trait information varied as a function of perceiver/target power roles. The impression rating data did not support the hypothesized relationship between power and stereotyping. Studies 2-4 failed to replicate the attention data of Study 1. Intergroup versus intragroup contexts are discussed as a possible explanation for the failure to replicate across studies.

Research suggests that (a) people conform more with members of their own social groups than with members of different social groups and (b) people assume that it is easier to convince members of their own social groups than members of different social groups. Based on the assumption that the extent to which people will reveal their attitudes to others primarily depends on the pressure to conform when the issue is low in personal relevance and on the possibility of convincing when the issue is high in personal relevance, it was predicted that issue relevance and majority similarity interact in determining public expressions of minority views. Two experiments were performed to test this prediction. In Experiment 1, participants expected to discuss an issue with four other students who opposed their position. Moreover, participants believed that the other students belonged either to the same social group as they or to a different social group than they. Results revealed the predicted interaction between issue relevance and majority similarity. When the issue did not affect them, the attitudes participants reported deviated more from the position of the dissimilar majority than from the position of the similar majority. When the issue did affect them, reported attitudes deviated slightly more from the position of the similar majority than from the position of the dissimilar majority. Experiment 2 attempted to replicate this finding for expressions of newly adopted minority views. Moreover, because Experiment 1 failed to find a difference between privately and publicly reported attitudes, Experiment 2 tried to make the distinction between the private and public response conditions stronger. Yet, inasmuch as participants' attitudes remained very close to the majority position, Experiment 2 did not provide a good test of the study's hypotheses. To explore the processes underlying the effects predicted for public expressions of minority views, both experiments also assessed participants' expectations for the group discussion. Although results supported some predictions, the evidence was not completely conclusive. Recommendations for future research include more systematic and direct ways of addressing the issue of mediation.

Moods and emotions can bias our memories of the past as well as our expectations for the future. What we believe is likely to happen in the future can further influence our behavior today. To demonstrate this, in experiment one, people were placed in a sad or angry mood, and then asked to rate how likely different events were to occur in the future. Events framed such that circumstances were the cause were seen as more likely to sad participants, while angry participants found events more likely when people (either themselves or another) were the cause of the event. In experiment two, the influence of moods on present behavior was demonstrated. People in sad or angry moods were presented with the opportunity to participate in programs that could help themselves or another. Sad people had more positive attitudes towards the programs when it was emphasized how much the program could help, thus they were more persuaded by circumstantial frames. Angry people had more positive attitudes towards programs that emphasized how much they, themselves, could help, thus they were more persuaded by human agency frames. Sad people were not persuaded when they, themselves, were portrayed as being the active agent for change. Likewise, angry people were not persuaded to help when the program, itself, was portrayed as the agent of change. This research likely has practical application in helping sad and angry people help others or themselves in counseling environments.

Two studies investigated the moderating effects of a series of variables on the relation between subjective norms and behavioral intentions. Study 1 probed the moderating effects of real vs. hypothetical construal of behavioral engagement on the relation between subjective norms and intentions to donate money. Contrary to hypotheses, expectations of real behavioral engagement attenuated the relation between subjective norms and behavioral intentions. In support of hypotheses, descriptive norms proved to be a more accessible source of normative information. Study 2 probed the moderating effects of experience with the target behavior and framing of the behavior for binge drinking. While there was no effect of experience with the target behavior on the descriptive norm-intention relation, experience served to attenuate the injunctive norm-intention relation. There was no effect framing of the behavior on the relation between subjective norms and behavioral intentions. While increases in social desirability lead to stronger relations between subjective norms and behavioral intentions for donating money, there were no moderating effects of social desirability on the relations when it came to drinking behaviors. There were also no moderating effects of need for cognition on the relations for either behavior.

Recent research suggests that forgiveness allows transgressions to continue, possibly by signaling acceptance. Three studies tested the prediction that the implications of forgiveness for future transgressions/benevolent behavior and/or happiness depend on whether forgivers also engage in communication behaviors that communicate a lack of acceptance of the transgression. Study 1 was a longitudinal study of newlywed couples in which couples first reported their tendencies to forgive, were observed during problem-solving discussions, and provided reports of their partners' considerateness every 6 months for 4 years; Study 2 was a daily diary study of established relationships in which participants reported transgressions, forgiveness, partner-regulation behaviors, and happiness every day for 2 weeks; Study 3 was an experimental study of undergraduate couples in which participants' perceptions of their partners' forgiveness and tendencies to engage in partner-regulation behaviors were first experimentally manipulated and participants were then given the opportunity to transgress against their partners. Results from Studies 1 and 2 demonstrated that forgiveness was more strongly associated with partners engaging in more compromise (Study 1), partners behaving more considerately (Study 1), and partners engaging in fewer transgressions (Study 2) that led the forgivers to experience greater happiness (Study 2) among forgivers who demanded more changes compared with forgivers who demanded less changes. Unexpectedly, Study 3 did not replicate these results, possibly because the partner-regulation manipulation failed to affect participants' perceptions of their partners' regulatory behaviors. Implications and suggested directions for future research are discussed. / A Dissertation submitted to the Department of Psychology in partial fulfillment of the Doctor of Philosophy. / Spring Semester, 2015. / March 23, 2015. / Close relationships, Forgiveness, Marriage, Partner regulation / Includes bibliographical references. / James K. McNulty, Professor Directing Dissertation; Frank D. Fincham, University Representative; Roy F. Baumeister, Committee Member; Jesse R. Cougle, Committee Member; E. Ashby Plant, Committee Member.

Despite social pressure for White Americans to be nonprejudiced, Black Americans still regularly experience discrimination. We argue that bias persists because although many White Americans espouse nonprejudiced beliefs, far fewer actively work to combat discrimination. Previous research on a newly developed scale of antiprejudice, or the belief that White people should proactively fight discrimination, indicates that higher levels of antiprejudice are associated increased proactive support among White people for multiple actions that would help put an end to discrimination. Drawing from research on prescriptive moral convictions (i.e., what people should do), we predicted that teaching White Americans four reasons why White people should be proactive in the fight against systemic racism would increase perceptions that White involvement is necessary in order for real change to occur and increase antiprejudiced beliefs. Results of a pilot study supported our predictions. Moreover, in a second study we replicated our results and extended them by demonstrating that our intervention not only increased perceptions about the necessity of White involvement and antiprejudice, it was also associated with a greater likelihood of volunteering for an equal rights organization. / A Dissertation submitted to the Department of Psychology in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Spring Semester 2018. / March 21, 2018. / Includes bibliographical references. / E. Ashby Plant, Professor Directing Dissertation; Patricia Y. Hightower, University Representative; Jon Maner, Committee Member; Paul Conway, Committee Member; Colleen Kelley, Committee Member.

The present research examined how group members construe events of conflict and violence in which their ingroup was involved and shed light on the relationship between ingroup identification and those construals of intergroup conflict. I proposed that construals of intergroup conflict vary along two main dimensions, attributions of responsibility and perceived severity of harm. Drawing on social identity theory, in three survey studies, I derived and tested hypotheses regarding cross-group and within-group variations in these dimensions. The first study examined the influence of ingroup identification (identification with Turkish nationality) on individuals' construals of a historical period characterized by intergroup violence (Armenian-Turkish conflict). Study 2 sought to extend this investigation in a different conflict context (Turkish-Kurdish conflict) by examining construals of a recent conflict which occurred during respondents' lifetime. Furthermore, it examined whether the influence of ingroup identification on construals of conflict varies as a function of one's group membership (Turks vs. Kurds). Study 3 extended the prior findings to examine whether individual differences in the strength of ingroup (Hutu vs. Tutsi) identification are an important predictor of construals of conflict in a context of ongoing extreme violence between groups (ethnic conflict in Burundi).

Two studies established that there are previously unexamined negative consequences of stereotype suppression beyond the traditional stereotype rebound effect. It is suggested that stereotype suppression has more in common with the literature on the suppression of emotionally arousing thoughts than the general thought suppression literature. Suppressing emotionally arousing thoughts often leads to emotional aftereffects (i.e. emotional rebound), but limited cognitive rebound. Both studies established the presence of emotional aftereffects following stereotype suppression using both direct and indirect measures of emotion. In addition, Study 1 found reduced stereotype activation following suppression relative to the control condition, whereas Study 2 attempted to determine whether these emotional aftereffects influenced the desire for intergroup contact. Future research involving people's attempts to correct for emotional aftereffects is suggested.

This research examined the impact of stereotype threat and identity salience on academic performance in male students at a highly-selective liberal arts college, as well as whether fear of negative evaluation and individuals' perception of role conflict between athlete- and student-identities moderate this effect. In Study 1, seventy-seven student-athletes were recruited to participate in a study on personality and problem-solving. Participants were randomly assigned to read one of five experimental conditions (athlete success, athlete failure, academic success, academic failure or a control), and then completed a brief math test. Results indicated that athletes in the athlete-success, academic-failure, and control condition performed significantly worse on a math test than those who in the athletic-failure and academic-success conditions. Moreover, the negative effects of stereotype threat were moderated by both role separation (RS), with athletes high in RS showing the most debilitating effects, role interference (RI), with student-athletes high in RI showing the most debilitating effects. Study 2 demonstrated that student-athletes who perform an athletic task prior to an academic task perform worse than those who perform the academic task first. Order of task does not matter for non-student-athletes. The discussion focuses and explains a potential identity shift hypothesis.