Threats and Prejudice

MISSING IMAGE

Material Information

Title:
Threats and Prejudice Do Threat-Relevant Stimuli Increase Prejudices toward Associated Outgroup Members?
Physical Description:
1 online resource (105 p.)
Language:
english
Creator:
Cook, Corey L
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Psychology
Committee Chair:
Cottrell, Catherine
Committee Members:
Moradi, Banafsheh
Webster, Gregory Daniel
Weigold, Michael F

Subjects

Subjects / Keywords:
atheist -- hiv -- outgroup -- prejudice -- threats
Psychology -- Dissertations, Academic -- UF
Genre:
Psychology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
A threat-based approach to prejudice suggests that specific prejudices often result from specific threats that outgroup members are perceived to pose (Cottrell & Neuberg,2005). In this study, I test whether innocuous threat-relevant stimuli (i.e.,news stories) increase prejudices toward members of a group perceived to pose associated threats. Specifically, I predicted that after reading a news story detailing threats to values (but not mentioning particular social groups),participants would show increased prejudices toward members of a group believed to pose values threats, but not toward members of groups perceived to pose unrelated threats. Participants (156 female, 119 male) read either a news story describing moral decline (threat condition) or dental school expansion (control condition), and were randomly assigned to evaluate and interact with one of three confederates: an atheist (values threat), a student who contracted HIV via blood transfusion (a health threat), or a student who posed no explicit threats (control). Although participants reported greater threat-based prejudices toward atheists in general, participants in the experimental condition (vs. the control) did not display increased prejudices toward atheists. Additionally, participants perceived targets with HIV to pose threats to health, but participants in the experimental condition did not report greater prejudices toward targets with HIV or toward control targets. These results therefore offer a novel contribution to the existing research on threat-based prejudices, but do not suggest that innocuous sources of threat-relevant information (e.g., news or other media) increase prejudices toward groups perceived to pose associated threats.
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility:
by Corey L Cook.
Thesis:
Thesis (Ph.D.)--University of Florida, 2012.
Local:
Adviser: Cottrell, Catherine.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2013-08-31

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2012
System ID:
UFE0044437:00001


This item is only available as the following downloads:


Full Text

PAGE 1

1 THREATS AND PREJUDICE: DO THREAT RELEVANT STIMULI INCREASE PREJUDICES TOWARD ASSOCIATED OUTGROUP MEMBERS? By COREY L. COOK A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2012

PAGE 2

2 2011 Corey L. Cook

PAGE 3

3 To my mom and grandma for their continued love, support, and encouragement

PAGE 4

4 ACKNOWLEDGMENTS I would like to thank Dr. Catherine Cottrell for her continued patience, knowledge, and guidance. I would also like to thank my committee members (Drs. Bonnie Moradi, Gregory Webster, and Michael Weigold) for helping me develop a project that I am proud of and am excited to build f urther research upon. Finally, I would like to thank my psychology support system: the amazing social psychology faculty, my fellow graduate student friends and colleagues, and the dedicated research assistants who helped me through all areas of data colle ction.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF FIGURES ................................ ................................ ................................ .......... 9 ABSTRACT ................................ ................................ ................................ ................... 10 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 12 Automatic Prejudices ................................ ................................ .............................. 12 Effects of Stereotyping ................................ ................................ ............................ 13 Threat based Theoretical Approach to Prejudice ................................ .................... 15 Potential Implications of the Proposed Research ................................ ................... 17 Selection of Threat relevant Stimuli ................................ ................................ ........ 19 Selection of Threat relevant Target Groups ................................ ............................ 21 Evidence of Atheists as a Perceived Threat to Values ................................ ..... 22 Evidence of People with HIV as a Perceived Threat to Health ......................... 23 Preliminary Support for Experimental Manipulations ................................ ........ 24 Pretesting threat perceptions of target groups ................................ ........... 24 Threat perceptions ................................ ................................ ..................... 24 Threat relevant discriminatory behaviors ................................ ................... 26 Emotional reactions ................................ ................................ ................... 27 Pretesting for threat relevant news stories ................................ ................. 29 Conscious awareness of threats ................................ ................................ 30 Reactions toward threat related groups ................................ ..................... 31 2 METHOD ................................ ................................ ................................ ................ 37 Participants ................................ ................................ ................................ ............. 37 Measures and Procedu re ................................ ................................ ........................ 37 Cover Story and Informed Consent ................................ ................................ .. 37 Manipulating Threat Relevant Stimuli ................................ ............................... 38 Target Group Manipulation ................................ ................................ ............... 39 Self r eport Measures of Prejudice ................................ ................................ .... 39 Dyadic Interaction and Behavioral Measures of Prejudice ............................... 41 Final Measures, Demographics, and Debriefing ................................ ............... 42 3 RESULTS ................................ ................................ ................................ ............... 45 Manipulation Checks ................................ ................................ ............................... 45 Self report Measures of Prejudice ................................ ................................ .......... 47

PAGE 6

6 Feelings toward Working with Other Participant ................................ ............... 47 Social Distancing ................................ ................................ .............................. 49 Threat Perceptions ................................ ................................ ........................... 50 General threat ................................ ................................ ................................ .. 51 Values threat ................................ ................................ ................................ .... 52 Health threat ................................ ................................ ................................ ..... 54 Emotional Reactions ................................ ................................ ........................ 55 Moral disgus t ................................ ................................ .............................. 55 Physical disgust ................................ ................................ ......................... 57 Threat related discriminatory intentions ................................ ..................... 59 Values related discriminatory intentions ................................ .................... 59 Health related discriminatory intentions ................................ ..................... 61 Dyadic Interactions and Following Measures ................................ ................... 62 Measures of Physical Distance ................................ ................................ ........ 62 Chairs ................................ ................................ ................................ ........ 62 Physical distance during picture ................................ ................................ 64 4 DISCUSSION ................................ ................................ ................................ ......... 72 Results of the Current Study ................................ ................................ ................... 73 Contributions and Limitations ................................ ................................ .................. 75 Future Directions ................................ ................................ ................................ .... 77 Conclusion ................................ ................................ ................................ .............. 79 APPENDIX A EXPERIMENT AL NEWS STORIES AND FOLLOW UP QUESTIONS ................... 80 B PARTICIPANT INFORMATION ................................ ................................ .............. 81 C MEASURES OF CURRENT CONCERN FOR RELEVANT AND NON RELEVANT THREATS ................................ ................................ ........................... 82 D PARTICIPANT INFORMATION SHEET FOR VALUES THREAT TARGET ........... 83 E PARTICIPANT INFORMATION SHEET FOR HEALTH THREAT TARGET ........... 84 F PARTICIPANT INFORMATION SHEET FOR CONTROL TARGET ....................... 85 G FEELINGS TOWARD OTHER PARTICIPANT ................................ ....................... 86 H SOCIAL DISTANCING SCALE ................................ ................................ ............... 87 I THREAT PERCEPTIONS TOWARD TARGET ................................ ...................... 88 J EMOTIONAL REACTIONS TOWARD TARGET ................................ .................... 89 K THREAT RELEVANT BEHAVIORAL RATINGS TOWARD TARGET .................... 91

PAGE 7

7 L .. 93 M THREE DOMAIN DISGUST SCALE ................................ ................................ ....... 94 N DEMOGRAPHICS ................................ ................................ ................................ .. 96 LIST OF REFERENCES ................................ ................................ ............................. 100 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 105

PAGE 8

8 LIST OF TABLES Table page 1 1 Group p retest d ata ................................ ................................ .............................. 34 3 1 Manipulation check r esults ................................ ................................ ................. 66 3 2 General prejudice m easures ................................ ................................ .............. 67 3 3 Threat measures ................................ ................................ ................................ 68 3 4 Threat relevant d iscriminato ry i ntentions ................................ ............................ 69 3 5 Behavioral distancing and disagreement with target r ecommendations ............. 70 3 6 Correlations of all dependent v ar iables measured in study p roper ..................... 71

PAGE 9

9 LIST OF FIGURES Figure page 1 1 Pretest ratings of negative affect toward target groups ................................ ...... 35 1 2 Pretest ratings of discriminatory intentions toward target groups. ...................... 36

PAGE 10

10 Abstract of Dissertation Presented to the Graduate School of the University of Fl orida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy THREATS AND PREJUDICE: DO THREAT RELEVANT STIMULI INCREASE PREJUDICES TOWARD ASSOCIATED OUTGROUP MEMBERS? By Corey L. Cook August 2012 Chair: Catherine A. Cottrell Major: Psychology A threat based approach to prejudice suggests that specific prejudices often result from specific threats that outgroup members are perceived to pose (Cottrell & Neuberg, 2005). In this study, I test whether innocuo us threat relevant stimuli (i.e., news stories) increase prejudices toward members of a group perceived to pose associated threats. S pecifically, I predict ed that after reading a news story detailing threats to values (but not mentioning particular social groups) participants would show increased prejudices toward members of a group believed to pose values threats, but not toward members of groups perceived to pose unrelated threats. P articipants (156 female, 119 male) read either a news story describing m oral decline (threat condition) or dental school expansion (control condition), and were randomly assigned to evaluate and interact with one of three confederates: an atheist (values threat), a student who contracted HIV via blood transfusion (a health thr eat), or a student who posed no explicit threats (control). Although participants reported greater threat based prejudices toward atheists in general, participants in the experimental condition (vs. the control) did not display increased prejudices toward atheists. Additionally, participants perceived targets with HIV to pose threats to health, but participants in the experimental condition did not

PAGE 11

11 report greater prejudices toward targets with HIV or toward control targets. These results therefore offer a n ovel contribution to the existing research on threat based prejudices, but do not suggest that innocuous sources of threat relevant information (e.g., news or other media) increase prejudices toward groups perceived to pose associated threats

PAGE 12

12 CH APTER 1 INTRODUCTION How do environmental stimuli affect the prejudices that people hold toward others? driven environment we are inundated with information warning us of the various threats that we face on a daily basis. For example, newspapers inform us about crimes that have been committed in our community, making us fear for our safety; health organizations inform us of infection rates and contagion concerns, making us worry about our health; and politicians and pundits warn us of d eclining morals and modern decadence, making us concerned for the future of our value s systems. However, researchers have thus far not explored how exposure to such information might affect the ways people perceive others and subsequently behave toward the m. In the current study I seek to determine whether threat relevant news stories can inadvertently increase prejudices held toward others, particularly when those prejudices morals and values. Automatic Pre judices T he majority of people even those who do not report explicit prejudicial attitudes toward other groups show biases in overt behaviors toward others based on group membership (Fiske, 2002). Additionally, simply thinking of other groups of people (e. g., African Americans, elderly people) can often activate s tereotypes of those groups and can affect subsequent behaviors toward others (e.g., Bargh, Chen, & Burrows, 1996; Devine, 1989; Fiske & Neuberg, 1990). F rom a functional theoretical perspective o ne purpose of the activation of stereotypes is to make snap judgments of the threats and opportunities afforded by members of other groups ( Cottrell & Neuberg, 2005; Kenrick, Griskevicius, Neuberg, & Schaller, 2010 ; Schaller, Faulkner, Park, Neuberg, & Ken rick,

PAGE 13

13 2004). These automatic judgments, however, can result in prejudicial reactions toward members of groups that are perceived to pose some sort of risk or threat For example, if a member of another group is heuristically perceived to threaten safety, t hen that split second judgment could lead to avoidance of the ostensibly dangerous person (Neuberg, Kenrick, & Schaller, 2011). In the current study I propose that innocuous stimuli could make people keenly aware of the threats that members of other group s ostensibly pose. If such stimuli increase threat perception, people might show increased prejudice toward members of groups that are stereotypically associated with relevant threats. For example, when resources are limited, people are likely to show incr eased prejudice toward members of groups who are perceived as a threat to those limited resources (e.g., Sherif et al., 1961/1988). In t he current study I am interested in testing whether news stories containing information relevant to the threats stereoty pically attributed to a group can affect prejudices toward members of that group. I predict that after reading a news story detailing a particular threat similar to media stories we might encounter on any given day people will respond with increase d prejud ice toward members of groups associated with that threat (even if the story does not mention any particular social group), but not members of unrelated groups Effects of Stereotyping In her review on the extant psychological literature, Fiske (2002) label ed display subtle biases and ingroup preferences, while a small minority of people show extreme biases and are openly prejudiced toward other social groups. Modern resea rch has accumulated a wealth of data demonstrating the automatic activation of

PAGE 14

14 stereotypes beliefs about the characteristics, attributes, and behaviors of members of other groups (Hilton & von Hippel, 1996) and the effects of those stereotypes (for a revie w see Devine, 2001). For example, landmark studies on implicit biases showed that both prejudiced and non prejudiced people are equally knowledgeable of the negative stereotypes of other groups and that simple presentation of racial category labels can pri me accessibility of stereotypes (e.g., Devine, 1989; Dovidio & Gaertner, 1986). Knowledge of existing stereotypes has also been shown to influence behaviors and preferences in a number of research paradigms. For example, numerous studies using the implicit association task (IAT; for a review see Greenwald, Poelman, Uhlmann, & Banaji,2009) have demonstrated that both black and white Americans es, suggesting implicit preferences for white people et al., 2007) have shown that both black and white participants (including police officers) are significantly more likely to mistake harmless objects held by black bystanders (e.g., cell phones or wallets) as guns, and subsequently make the split black bystanders significantly more than white bystanders. Despite the accumulated evidence of the negat ive effects of stereotypes, researchers have also concluded that the process of stereotyping is cognitively efficient. Cognitive researchers have argued that stereotypes are a natural part of human cognition and that they help people make sense of their so cial environment by allowing them to form simple, well structured impressions of others (Gilbert & Hixon, 1991; Macrae & Bodenhausen, 2000). Evolutionary researchers have contended that humans

PAGE 15

15 are attuned to the threats and opportunities afforded by others (e.g., Kenrick, et al., 2010; Neuberg et al., 2011) and that stereotypes allow for quick judgment of potential affordances. Along these lines, Schaller and colleagues (2004) have argued that humans have evolved danger avoidance mechanisms that lead people to be vigilantly attuned to the threats afforded by others, including threats to physical health and safety (Schaller, Park, & Faulkner, 2003). Although such threat perception mechanisms could lead people to overperceive threats posed by others (Haselton Nettle, & Andrews, 2004), these mechanisms could also prove beneficial by protecting people from members of groups stereotypically associated with danger. Threat based Theoretical Approach to Prejudice In the current study, I am interested in testing whe ther innocuous information relevant to the threats stereotypically associated with different outgroups affect prejudices toward members of those group s. The hypotheses of the study are based on a functional, threat based theoretical account of prejudice, a s outlined by Cottrell and Neuberg (2005). This approach suggests that outgroup members are believed to pose group resources and therefore evoke qualitatively distinct emotional reactions. Cottr ell and Neuberg (2005) used this approach to generate specific predictions about the threat driven nature of intergroup affect, focusing on specific negative emotions rathe r than general negative affect (the focus of much of the previous research on prejudice). Their predictions were base d on empirical research on the distinct emotions elicited by various perce ived stimulus events and the actions taken to resolve the problems pose d by such events (e.g., Cosmides & Tooby, 2000; Ekman, 1999; Plutchik, 2003 ). According to this functional appr oach to the role of emotions, perceived stimulus events (e.g., obstacles to desired outcomes ) would lead to discrete

PAGE 16

16 emotions (e.g., anger ) and specific actions (e.g., aggression) c ould be taken to remove the obstacle thus offering an adaptive solution t o the perceived problem. In the current study, I am interested specifically in perceptions of threats to values. An evolutionary approach to emotions suggests that moral contamination by unpleasant or disagreeable ideas (i.e., threats to values) should eli cit feelings of disgust and evoke actions to avoid or reject opposing values (Mackie & Smith, 2002). functional threat based hypotheses regarding prejudice were largely supported. T hey found that different groups evoked qual itatively different profiles of specific threat perceptions and emotional reactions These differences in evaluations of groups had been previously overlooked in more traditional studies of prejudice. Most importantly for the purposes of the current study, even when levels of general threat assessment or overall prejudice toward some groups were high, specific threat assessments often differed across these groups. For example, participants reported high levels of general prejudice toward groups such as gay men and fundamentalists Christians, and both groups were rated as posing significantly higher levels of general threat than the control group (European Americans). Threat perceptions attributed toward these groups differed, however participants reported th at they perceived gay men as posing threats to both health and values, while reporting that fundamentalist Christians pose threats to values and personal freedoms. Other research has also suggested that perceptions of threat can increase prejudicial responses. As briefly mentioned earlier, Schaller and colleagues (2004) have argued that humans have evolved danger avoidance mechanisms that lead them to be

PAGE 17

17 immune syste with the spread of disease. These researchers have found that after being primed with contagion cues, individual differences on perceived vulnerability to disease (PVD) can predict antipathy and prejudicial responses toward a wide variety of others, including those of other ethnicities (Faulkner, Schaller, Park, & Duncan, 2004), physically disabled people ( Park, Faulkner, Schaller, & 2003 ), obese people (Park, Schaller, & Crandall, 2007) and elderly people (Duncan, Schaller, & Park, 2009). These numerous studies suggest that perceived threats to personal health, whether or not those threats are indeed valid, lead to increased prejudicial responses. In addition to these findings, Scha ller and colleagues also found that chronic perceptions of threat to personal safety can interact with stereotypes of other groups (Schaller, Park, & Mueller, 2003). Participants who scored high on belief in a dangerous world (BDW) displayed increased prej udice toward black targets a group stereotypically associated with threats to personal safety when they were made to feel vulnerable (i.e., experimenters required them to complete measures in a dark room). Finally, threats associated with fear of n death ( Greenberg, Solomon, & Pyszczynski, 199 esteem and ego ( Fein & Spencer, 199 7) have also shown to increase prejudice toward worth. Potential Implications of the P roposed Research Although previous research has shown that perceptions of threat can indeed increase prejudice, the question proposed in the current study whether threat relevant stimuli, not related to any particular group, will increase prejudice toward members of a group stereotypically associated with those threats has not been addressed directly. In studies on the behavioral immune system, for example, the researchers found that

PAGE 18

18 people who were already chronically concerned about threats to their healt h or safety increased prejudice toward groups based on the stereotypes of those groups. For example, after being primed with pathogen cues, participants who reported high levels of perceived vulnerability to disease reported greater prejudices toward obese people (Park et al., 2007), and those who reported high levels of belief in a dangerous world reported greater prejudice toward black targets when made to feel vulnerable (Schaller et al., 2003). Consistent increases in prejudice were not found for partic ipants who scored lower on perceived vulnerability to disease or belief in a dangerous world. Other research (e.g., Fein & Spencer, 1997; Greenberg, et al., 199 7) found that threats to self concept can lead to outgroup derogation. However, these particular findings have little to do with the specific stereotypes associated with those targets of derogation or the specific threats those particular target groups are perceived to pose. Rather, these studies focus on the act of derogation as a way to bolster one concept and sense of self worth. The current research adds a unique contribution to the extant literature by focusing on whether or not a seemingly innocuous threat relevant cue can affect prejudices displayed toward members of groups stereotyp ically associated with the respective threat. If threat relevant cues indeed increase prejudices toward members of groups stereotypically associated with related threats, this process could have very important real world implications. For example, members of groups stereotypically perceived to threaten health (e.g., people with infectious diseases, obese people, etc.) might experience prejudice from others in response to everyday stimuli that raise awareness of health (e.g., in exercise facilities, food ser vice professions, or by health care

PAGE 19

19 providers). Members of groups stereotypically perceived to threaten morals or values (e.g., atheists, feminist activists, gay men, Muslim Americans, etc.) may experience increased prejudice in areas that raise awareness of, or are associated with, the development of morals or values (e.g., issues regarding marriage, adoption, education, or new institutions) for fear of moral contamination. The current research can help determine whether innocuous cues might lead to increa sed prejudices toward members of particular groups in situations relevant to their associated threats. Selection of Threat relevant Stimuli According to a functional, threat based account of prejudice, it is the respective threats that outgroup members ar e perceived to pose that lead to negative prejudices (Cottrell & Neuberg, 2005). To determine which specific outgroups and threats to focus on in this study, I turned to both the existing psychological literature and to cultural examples of prejudice. Res earchers have noted a cultural shift in the acceptability and expression of prejudices (e.g., Devine & Elliot, 1995; Madon et al., 2001). Traditional forms of prejudice such as overt racism are no longer viewed as acceptable by the majority of Americans (e .g., Dovidio & Gaertner, 1991, 2004). Although racism certainly still exists among some people, many theories regarding modern forms of racism propose that even racist individuals are aware of the norms against racism and therefore seek opportunities to ex press prejudice in more subtle ways (e.g., Dunton & Fazio, 1997; Plant & Devine, 1998). Although overt racism appears to no longer be socially acceptable, simply browsing the news headlines offers evidence that other forms of prejudice are alive and well. At the time of this writing, for example, immigrants, gay and lesbian Americans,

PAGE 20

20 and Muslim Americans are among the social groups currently fighting for equal standing in American society. Heated political attacks based on political ideology are also comm on occurrences in the media. These examples of prejudice could be argued to be more acceptable in society than more traditional forms of prejudice such as racism or sexism. One reason these prejudices might be more acceptable is because membership in one o f the previously mentioned groups is perceived by many to be volitional, and people show greater prejudice toward stigma s that are perceived as controllable or chosen by targets (e.g., Crandall, 1994). of a group that advocates different morals from the majority, the group is seen by the majority group members as a threat to values (Haidt & Joseph, 2004). The behavioral pattern toward people who are perceived to threaten values parallels that of the beh avioral immune system much like perceptions of threats to health (e.g., Faulkner et al., 2004; Park et al., 2003, 2007), perceptions of threats to morals or values result in feelings of disgust and negative evaluations of others (Rozin, Haidt, & McCauley, 2009). For example, in heterosexual Caucasian participants as posing the highest threats to values ( gay men, feminist activists, and fundamentalist Christians) also elicited greate r feelings of disgust than other groups (e.g., African Americans, non fundamentalist Christians, etc.). Because people are more willing to express prejudices toward groups in which membership is perceived as volitional, changes in these prejudices should b e more pronounced than other forms of prejudice (e.g., racism or sexism). I therefore decided to

PAGE 21

21 examine stimuli relevant to threats to values, as they would allow me to better observe changes in prejudicial responses. Selection of Threat relevant Target Groups In their review on the psychological literature exploring social stigma, Hebl and Dovidio (2005) argued that the reliance of stigma researchers on self report measures has overlooked important information that can be obtained from observing person t o person interactions. Although self report measures offer very important information regarding prejudice (e.g., affective reactions and explicit attitudes toward others), the questions explored in the current study have potential everyday implications for behaviors toward members of various groups. Because of these potential behavioral implications toward real people, it is important to not only focus on self report measures of affect and prejudice but to also measure responses toward a target who may pose a specific threat within the context of a face to face interaction. Measuring such behaviors in a controlled laboratory setting requires the use of carefully trained experimental confederates. The decision to use confederates to represent multiple target groups also requires selection of target groups that are not distinguishable by physical appearance alone, and can be blindly manipulated (i.e., the confederate does not know which group he or she ostensibly represents). I therefore chose not to focus on t arget groups based on ethnicity, race, or gender. This decision fits well with the framework involving threat relevant stimuli, as those groups that are perceived to pose threats to values are often groups to which membership is perceived as volitional, wh ich is not the case for group distinctions based on race or gender. To determine which target groups to include in this study, I turned to the extant research on threat perception and prejudice. Because I will be manipulating values

PAGE 22

22 relevant stimuli, it is important to include a target group that is perceived to threaten values. However, to ensure that the manipulations affect responses only toward groups representing corresponding threats (and not toward those representing unrelated threats), I also dec ided to include a target group that represents threats to health (a threat for which saliency will not be manipulated) in addition to a true control group. I chose atheists as the group representing values threats, and people with HIV as the group represen ting health threats, for reasons outlined below. In the control threat condition (i.e., after reading a neutral news story), I predict that participants will show greater prejudice toward atheists and people with HIV compared to control targets. If my hypo thesis regarding the effects of threat relevant stimuli is correct, however, after reading a story detailing threats to values, participants will show increased prejudices toward atheists, but not toward people with HIV or control targets. Evidence of Ath eists as a Perceived Threat to Values Although research on perceptions of atheists is still in its early stages, there is sufficient evidence to believe that atheists are perceived as a threat to values. For example, in a recent survey 57% of American resp ondents reported that they believe that morality is not possible without belief in God (Pew Research Center, 2002). In a nationally representative survey, Edgell, Gerteis, and Hartmann (2006) found that atheists are less likely to be accepted by Americans than a number of other ethnic, religious, and minority groups (e.g., immigrants, Muslims, and non heterosexuals), and Gervais, Shariff, and Norenzayan (2011) likewise f ound in a series of experimental studies that anti atheist prejudice is pervasive and marked by feelings of distrust. Participants in one of their studies, for example, indicated similarly high levels of distrust

PAGE 23

23 toward atheists and rapists compared to Mus lims and Christians. Finally, in a recent study by Ritter and Preston (2011), participants reported increased moral disgust in response to anti The God Delusion (2006). The combination of these diverse data sources gives sufficient reason to believe that participants will view atheist targets as threats to values. I have therefore chosen to use atheists as the group manipulated to represent threats to values. Evi dence of P eople with HIV as a Perceived Threat to Health causing pathogens (e.g., Schaller et al., 2004). Numerous studies have shown that people distance themselves from, and show more prejudicial reactions toward, others who are heuristically perceived as posing threats to health, such as obese people (Park et al., 2007) and physically disabled people (Park et al., 2 003). If people are made aware that another has a communicable disease, their behavioral immune system should respond accordingly and lead them to distance themselves from the threatening target, regardless of whether there is a realistic chance of contami nation. participant who indicates that he contracted HIV from a blood transfusion. I chose HIV over other potential communicable diseases to ensure that participants ar e familiar with the disease and its potential harm. Herek and colleagues have noted that people show consistent prejudices toward people with HIV and AIDS (Herek, Capitanio, & Widaman, 2002), and Crandall and colleagues (Crandall, Glor, & Britt, 1997) foun d that people seek to distance themselves from others with AIDS related afflictions. To avoid activating values related threats, the experimental materials will make it clear that the

PAGE 24

24 target did not contract the disease through sexual behaviors or from ill icit drug use, as these perceptions have been shown to increase prejudice toward those with HIV and AIDS (Capitanio & Herek, 1999). By controlling for increased threats to values, I believe that these targets will be perceived as posing threats to health, and I therefore selected this group to manipulate perceptions of threats to health. Preliminary Support for Experimental Manipulations To ensure that the experimental manipulations elicited the hypothesized threats, pretesting was conducted on threat perc eptions of social groups and on effects of the experimental news stories. Pretesting threat perceptions of target g roups Using the psychology participant pool (the same resource from which participants will be recruited for the actual experiment), I pretes ted responses toward target groups of interest to ensure that they adequately represent the proposed threats. Because this was a preliminary assessment I chose simple, two item measures to determine participant reactions to each of the groups of interest ( atheists, people with HIV, and student/control group). One hundred thirty one participants (84 female, 47 male) were asked to think about their impressions of each target group and to answer items assessing perceptions of threat ascribed to members of each group, affective responses toward members of each group, and behavioral/discriminatory intentions toward members of each target group in threat related situations. Threat perceptions Participants reported, on a six point scale (1 = Strongly disagree 6 = Strongly agree ) the extent to which they perceived members of each group to pose generalized threats (i.e., members of this group pose challenges to people like me and pose

PAGE 25

25 problems for people like me ; mean r = .67 across groups, all p s <.001), threats to health ( members of this group increase spread of illness and endanger the personal health of people like me ; mean r = .71 across groups, all p s <.001), and threats to values ( members of this group promote values that are seen as immoral by and go against the values of people like me ; mean r = .85 across groups, all p s <.001). A repeated measures analysis of variance (ANOVA) suggested that perceptions of generalized threat did not differ across groups [ F (2, 258) = 0.77, p Table 1 1 f or all means]. Perceptions of values threat, however, did differ across groups ( F = 63.95, p threats to values ( M = 4.03, SD = 2.18) than control targets [ M = 2.24, SD = 1.37; t (1 30) = 9.72, p <.001, d = .98] and people with HIV ( M = 2.04, SD = 1.4; t = 10.07, p <.001, d = 1.09). Participants did not report any differences in perceived threats to values posed by control targets and people with HIV ( t = 1.42, p = .16, d = .15). A repeated measure ANOVA also suggested that perceptions of health threat differed across groups [ F (2, 258) = 5.4, p with HIV as posing greater threats to health ( M = 3.48, SD = 1.7) than control targets [ M = 1.84, SD = 1.1; t (130) = 10.19, p <.001, d = 1.14] and atheists ( M = 1.63, SD = 1.00; t = 12.67, p <.001, d = 1.33), as expected. Contrary to predictions regarding perceptions of threat, participants rated control targets to pose greater health threats than atheists, t = 2.21, p <.05, d = .21. In sum, participants rated atheists as posing greater threats to values than control targets or people with HIV, and rated people with HIV as posing greater threats to

PAGE 26

26 health than control targets or atheists. Perceptions of generalized threat, however, did not differ between the groups. Threat relevant discriminatory behaviors In addition to threat perceptions, participants also rated (using a six point scale, 1 = Strongly disagree 7 = Strongly agree ) how they would react toward targets in threat relevant situations. To assess behaviors related to values threats, participants responded to the following two items: to what extent would you recommend a member of this group teach morals and values to your children as a volunteer at a local day care center? and, how likely would you be to vote for a member of this group who is running for political office in your state? These items correlated adequately across groups (mean r = .52, all p s < .001) and were aggregated as a single measure. A repeated measures ANOVA suggested that discriminatory behaviors in values relevant situations differed across groups [ F (2, 258) = 117.94, p Table 1 1 for means]. As expected, participants indicated greater dis criminatory intentions toward atheists in values related situations ( M = 5.21, SD = 1.66) than toward control targets [ M = 2.79, SD = 1.08; t (130) = 14.8, p <.001, d = 1.73] and people with HIV ( M = 3.31, SD = 1.53; t = 11.7, p <.001, d = 1.19) in the sam e situation. Additionally, participants rated greater discriminatory intentions toward people with HIV in values relevant situations than toward control targets, t = 3.94, p <.001, d = .39. To assess behaviors related to health threats, participants responded to two items: to what extent would you recommend a member of this group come into close contact with a sick family member as a volunteer at a loc al hospital?, and, to what extent would y ou recommend a member of this group draw your blood as a volunteer at a local blood bank? These items correlated adequately across target groups (mean r = .54, all ps <

PAGE 27

27 .001) and were therefore combined for each group. A repeated measure ANOVA suggested t hat discriminatory behaviors in health relevant situations differed across groups [F (2, 258) = 175.16, p Table 1 1 ]. As expected, participants indicated greater discriminatory intentions toward people with HIV in health related situations (M = 5.58, SD = 1.5) than toward control targets [M = 2.44, SD = 1.26; t (130) = 20.79, p <.001, d = 2.27] and atheists (M = 3.42, SD = 1.77; t = 11.8, p <.001, d = 1.32) in the same situation. Additionally, participants rated greater discriminatory inten tions toward atheists in health relevant situations than toward control targets, t = 6.58, p <.001, d = .64. In sum, these data also support expectations regarding threat perception of groups. Participants reported greater discriminatory intentions toward atheists than control targets or people with HIV in values related domains, and reported greater discriminatory intentions toward people with HIV than control targets or atheists in health related domains. Emotional reactions As an additional test of thr reported the extent to which they felt different emotional reactions toward members of each group (1 = Not at all 6 = Very much ). Specifically, feelings of general negativity and moral and physical disgu st were recorded. According to a threat based approach (Cottrell & Neuberg, 2005), groups perceived to pose threats to health should elicit increased feelings of physical disgust, while those perceived to pose threats to values should elicit increased feel ings of moral disgust. These specific emotional reactions would be masked by a traditional view of prejudice (i.e., as a general negative reaction toward outgroup others).

PAGE 28

28 reported feeling s of negativity and positivity (reverse scored) felt in response toward targets. These two items correlated adequately across groups (mean r = .57, all p s < .001) and were combined as one measure. A repeated measures ANOVA suggested that feelings of negati vity differed across groups [ F (2, 258) = 78.63, p 1 ], with participants reporting greater negativity toward atheists ( M = 3.76, SD = 1.71) and people with HIV ( M = 2.93, SD = 1.29) compared to control targets [ M = 1.75, SD = 0.93; t (130) = 12.46, p <.001, d = 1.46 for atheists, and t = 9.3, p <.001, d = 1.05 for people with HIV, respectively]. Additionally, participants reported greater feelings of general negativity toward atheists compared to people with HIV, t = 5.39, p <.001, d = 0.55. To assess feelings of moral disgust participants reported the extent to which they felt morally disgusted and morally sickened in response to each target group. These items correlated adequately across groups (mean r = .89, all p s <.001) and wer e combined. Feelings of moral disgust differed across groups [ F (2, 258) = 27.5, p <.001, response to atheists ( M = 2.64, SD = 1.85) than toward control targets [ M = 1.5, S D = .89; t (130) = 7.53, p <.001, d = 0.78] or people with HIV [ M = 1 .79, SD = 1.28; t = 5.08, p <.001, d = 0.53]. Additionally, participants reported greater feelings of moral disgust in response to people with HIV than toward control targets, t = 2.55, p = .01, d = .26. To assess feelings of physical disgust, participants reported to what extent they felt physically disgusted and physically sickened in response to each target group. These items correlated adequately across groups (mean r = .87, all p s <. 001) and were

PAGE 29

29 combined. Feelings of physical disgust differed across groups [ F (2, 258) = 21.32, p response to atheists ( M = 1.71, SD = 1.2) compared to control targe ts [ M = 1 .31, SD = .7; t (130) = 4.21, p <.001, d = 0.41], participants reported greater levels of physical disgust in response to people with HIV than control targets ( t = 5.7, p <.001, d = 0.63) and atheists ( t = 2.57, p =.01, d = 0.24), thus supporting the threat based hypotheses. Overall the preliminary analyses suggest that the proposed groups are indeed perceived to pose the predicted specific threats. The data from these measures suggest that atheists are perceived to pose greater threats to values t han control targets or people with HIV. Participants additionally reported greater discriminatory intentions toward atheists in values related domains and reported greater feelings of moral disgust toward atheists than the other groups. People with HIV, on the other hand, are perceived to pose greater threats to health than control targets or atheists, and are treated with greater discriminatory intentions in health related domains and elicit greater feelings of physical disgust than the other groups. Pret esting for threat relevant news s tories Two brief news stories one relevant to values threats and one control (i.e., not relevant) were written and pretested to determine whether reading the experimental story adequately increases awareness of appropriate threats relative to the control story. The two stories were carefully written to be of similar length and structure. Anthropological research suggests that there are certain traits that are universally valued in human social groups across history and cultu res (Brown, 1991), and modern psychological research has suggested that traits related to trustworthiness are highly valued among those with whom we are socially interdependent (Cottrell, Neuberg, & Li,

PAGE 30

30 2007). To increase awareness of threats to values I w rote a news story specifically discussing violations to such highly valued traits (e.g., fairness, fidelity, honesty, loyalty), An appropriate control story should elicit no elevated awarene ss of threats. For the control story, I wrote an adapted summary of a news story that appeared in the local student newspaper detailing plans for dental school expansion on campus (Jester, One hundred uni versity students (51 female, 49 male) were randomly presented with either the experimental or control story and were asked to answer a number of questions following the story. As a manipulation check of conscious awareness of elevated threats, participants provided ratings of their current concern for values of others in society (i.e., students and members of society as a whole). However, because it is possible that awareness of threats could be activated at an unconscious level but not a conscious level (a s has been found in the terror management theory literature using mortality salience manipulations; e.g., Pyszczynski, Greenberg, Solomon, & Schimel, 2004), participants answered items measuring affective reactions toward groups posing threats to values (a theists), health (people with HIV/AIDS), or neither (college students), and answered items measuring discriminatory intentions toward groups posing threats to values (atheists) or health (people with HIV/AIDS). Conscious awareness of threats Following presentation of the news article participants were asked to indicate to point scale (1 = Not at all 6 = Extremely ). These items correlated highly ( r = .64, p < .001) and were

PAGE 31

31 therefore combined to form one measure of morality concerns. A one way analysis of variance (ANOVA) determined that participants did not report greater concern for morals in the experimental condition ( M = 3.98, SD = 1.16) compared to the control condition [ M = 3.74, SD = 1.09; F (1,98) = 1.14, p of each item of morality concern corroborated these results. Reactions toward threat related grou ps Because it is possible that the threat related news story might not affect conscious awareness of threats while still affecting unconscious processing, I also measured responses toward a threat relevant group (atheists) and threat irrelevant group (peop le with HIV/AIDS and college students) following presentation of the news story. To measure general affective responses toward groups, participants indicated on a six point scale (1 = Not at all 6 = Extremely ) the extent to which they feel tense and anxio us when they think of group members. These items correlated well for each group (mean r = .66, p <.001) and were thus combined into one measure of affective response toward each group. A one way ANOVA indicated that participants reported increased negative affect toward atheists in response to the experimental news story [ M = 3.05, SD = 1.14 vs. M = 2.43, SD = 1.2; F (1, 96) = 6.83, p increased negative affect toward people with HIV ( M = 3.28, SD = .88 vs. M = 2.93, SD = 1.02; F = 3.36, p M = 2.29, SD = .93 vs. M = 2.36, SD = 1.09; F = .14, p Figure 1 1 ). As a measure of discriminatory intentions, participants reported their endorsement of behaviors regarding a theists and people with HIV. For atheists, participants reported (on a 1 to 6 scale; Strongly Agree to Strongly Disagree respectively) whether they would be willing to vote for an atheist presidential candidate, support a local business

PAGE 32

32 run by atheists, a nd whether they believed the US supreme court should include atheists. These items (reverse therefore aggregated into one measure of discriminatory intentions toward atheists. Participants also indic ated their endorsement for organizing a charity event for people with HIV/AIDS, participating in a charity walk for people with HIV/AIDS, and donating .82) and were reverse scored and aggregated as one measure of discriminatory intentions toward people with HIV/AIDS. One way ANOVAs run on these items indicated that participants reported increased discriminatory intentions toward atheists in response to the experimental news story vs. the control story [ M = 3.67, SD = 1.31 vs. M = 3.15, SD = 1.3; F (1, 96) = 3.97, p people with HIV ( M = 2.45, SD = 1.02 vs. M = 2.18, SD = .93; F = 1.94, p = in response to the experimental story vs. the control story ( Figure 1 2 ). In sum, although participants did not indicate an increase in conscious concern for morals in response to the experimental condition, participants displayed consistent changes in responses to the threat relevant group (atheists) but not threat irrelevant groups (people with HIV/ AIDS or college students). These pilot data therefore offer adequate support that the experimental, threat relevant news story raises awareness of threats to values (but not other threats), while the control news story does not. Research predictions and overview of study To test whether innocuous threat relevant stimuli affect prejudices toward members of groups perceived to pose associated threa ts (but not members of groups posing non associated threats), an

PAGE 33

33 experimental situation was developed in which participants interacted with an ostensible member of one of the selected target groups. Participants were led to believe that they would read a n ews story detailing current issues on campus, and that they will be working with another student to brainstorm ways for the university to address these issues. Participants were randomly assigned to either the experimental condition (i.e., they read the ne ws story explaining moral decline among college students; Appendix A) or the control condition (they read the news story describing dental school expansion on campus; Appendix A). After sharing demographic information for any ostensible discussion partners participants learned that their discussion partner is a member of one of the selected target groups (i.e., atheist, person with HIV, or student/control). Participants then completed self report measures of prejudice and affect before interacting with an experimental confederate. A ccording to my threat based hypotheses, after reading a news story describing moral decline (but not mentioning any particular groups of people), participants will report increased prejudicial responses toward members of groups perceived to pose threats to values (i.e., atheists), but not toward members of groups posing unrelated threats (i.e., people with HIV who would be perceived to pose threats to health or college students who act as a control group).

PAGE 34

34 Table 1 1. Group pretest d ata General Threat Values Threat Health Threat Values Behaviors Health Behaviors General Negativity Moral Disgust Physical Disgust Group M SD M SD M SD M SD M SD M SD M SD M SD Control 2.71 1.39 2.24 1.37 1.84 1.1 2.79 1.08 2.44 1.26 1.75 0.93 1.5 0.89 1.31 0.7 Atheists 2.71 1.73 4.03 2.18 1.62 1.0 5.21 1.66 3.42 1.76 3.76 1.71 2.64 1.85 1.71 1.2 HIV 2.52 1.46 2.04 1.4 3.48 1.7 3.31 1.53 5.58 1.5 2.93 1.29 1.79 1.28 2.04 1.4 Note. Higher means indicate more negative responses For all means, n = 131.

PAGE 35

35 Figure 1 1 Pretest ratings of negative affect toward target groups following presentation of threat relevant vs. threat irrelevant news stories. Participants reading the experimental story reported increased negative affect toward atheists, but not toward the other groups, compared to those reading the control story.

PAGE 36

36 Figure 1 2 Pretest ratings of discriminatory intentions toward target groups following presentation of threat relevant vs. threat irrelevant news stories. Participants reading the experimental story reported increased discriminatory intentions toward atheists, but not toward the people with HIV, compared to those read ing the control story.

PAGE 37

37 CHAPTER 2 METHOD Participants This study employ ed a 3 (target group: atheist, person with HIV, or control) by 2 (threat condition: values versus control ) by 2 (participant gender) completely between subjects design Although I do not predict gender differences in responses toward targets, I am including gender as a factor in the overall design to adequately test for gender differences Participants consist ed of 275 ( 156 female and 119 male ) undergraduates recruited from the psychol participant pool and compensated with credit toward their course requirement three female participants chose not to allow use of their data and are therefore not included in analyses Participants averaged 19.16 years of age ( SD = 1.8), an d ranged from 18 to 35. The majority of participants identified as White/Caucasian (65.5%), while 16.7% identified as Hispanic, 15.6% as African American, 8.4% as Asian or Asian American, while 8.4% identified as multi Measures and Proc edure Cover Story and Informed Consent Participants were greeted by an experimenter upon arrival and asked to take a seat at a computer in a small room. Necessary information regarding the study (e.g., costs/benefits, what will be asked of participants, et c.) was provided via computer, and participants were asked to provide consent to participate. After providing informed consent, participants were asked to read the cover story, which led them to believe that the researchers are working in coordination with the university to get student feedback on university involvement in regards to various student issues. Participants were told

PAGE 38

38 are measuring the thoughts and impression s people form of others whe n working together as a group. T his portion of the cover story was designed to allow for measurement of impressions of the confederate while minimizing participant suspicion. Participants were told that they w ould first read a sh ort news story detailing current student issues and would then with a discussion partner to generate ideas to improve university involvement with students. After completing the informed consent and study overview, participants were asked to answer some basic questions about themselves (Appendix B ). They w ere told the information would be shared with any other participants who would be participating as members of the focus group to learn more about them. Manipulating Threat Relevant Stimuli After completing the information sheet, participants were asked to read a short news story describing current issues on campus. Participants were randomly presented with either the threat relevant news story (describing moral decline) or the cover story (d escribing dental school expansion). Participants were asked to pay close attention to the story because they would be asked questions regarding it later. Immediately following the story participants were asked to answer two questions to ensure they paid at tention Immediately following the news story, participants answered items measuring concern for related and unrelated threats. Two items were used to measure current concern for morals/values and two items were used to measure concern for health (Appendix C). Participants indicated the extent of their agreement with each item on a

PAGE 39

39 six point scale (1 = Not at all 6 = Extremely ). If conscious awareness of activated threats is increased in response to the experimental condition, then participants in the experimental condition should report increased concern for values but not health. Using a similar six point scale (1 = Strongly disagree 6 = Strongly a gree ), four additional items asked participants to indicate the extent of their agreement with statements regarding moral behavior. These items were added as an additional manipulation check participants who are more concerned about morality should indicate higher agreement with moral behaviors and less agreement with questionable behaviors. Target Group Manipulation After reading the news story, participants were informed that they would be brainstorming with one disc ussion partner. They were then provided with the profile information ostensibly completed by the other participant (i.e., the target). Participants were randomly assigned to receive one of three different target conditions. For each target, all information on the sheet was the same except the answer to the last question last question served as the target group manipulation. For the threat relevant group (atheists) the answer provide (Appendix D). For the threat irrelevant target (i.e., the health threat target/student with (Appendix E), and for the Self report Measures of Prejudice Participants were asked to take a moment to look over the information sheet then answer q uestions about the target. Participant s were reminded that the researchers are

PAGE 40

40 interested in how people form impressions of others, and they were assured that their responses would remain completely confidential and would not be seen by the experimenter or the other participant. Using a six po int scale (1 = Not at all 6 = Very much ), p articipants first provided responses to questions asking about their feelings toward working with the target garding university ). Participants then answer ed questions measuring general a ) adopted version of the Bogardus (1925) Soc ial Dista ncing Scale (Appendix H ). This seven item scale assesses general Strongly disagree 6 = Strongly a gree ), and has to be reliable in a number of previous studies (e.g., Crandall 1991a, 1991b; Crandall, et al. 1997). Following the social distancing scale, participants were asked to indicate their level of agreement to items measuring the extent to which they perceive the target to pose threats to values and health (items used a six point scale; 1 = Not at all 6 = Very much oppose the values of p of health threat (Appendix I ) Participants next used a similar six point scale (1 = Not at all 6 = V ery much ) to indicate emotional reactio ns toward the target (Appendix J ).

PAGE 41

41 him/he indicated their willingness to perform various behaviors toward targets in situations v These items were measured using a six point scale (1 = Not at all 6 = Very much ; A ppendix K). Dyadic Interaction and Behavioral Measures of Prejudice After reporting their impressions of the target, participants were directed via computer to alert the experimenter that they had finished the first part of the study The experimenter then explain ed that the second part of the study, in which participants will brainstorm ways that students can help solve university issues, would take place in another room The experimenter momentarily stepped out of the room to ostensibly make sure that the other participant was also ready. They then led the participant into a larger room with six chairs arranged in a semi circle, facing a desk. When participants enter ed the room a male confederate was seated in a chair nearest to the entrance The experi The experimenter unobtrusively recorded the number of chairs participant s place d between themselves and the confederate as a behavioral measure of prejudice toward the target similar meas ures of distance have been used in previous research as non verbal indicators of prejudice (Hebl & Dovidio, 2005). Next, the experimenter handed the participant and confederate each a clipboard with a blank sheet of paper and asked them to take a moment t o brainstorm and list

PAGE 42

42 their own ideas of how the university could involve student feedback to address student issues Confederates were trained to write three recommendations that would later be rated by participants ( A ppendix L). A fter approximately three minutes (or until both the participant and confederate appeared to have stopped writing ), the experimenter moved on to the next part of the study. They explained that each participant would be led to their respective computers to input their ideas generat ed during the brainstorm activity. Before moving on, however, the experimenter asked to take a picture of the two participants for documentation purposes. Experimenters assured the participants that the photos would not be made public, and all participants agreed to take a picture. Experimenters first asked the confederate to stand in front of a large wall calendar. Confederates were trained to strategically place their right shoulder parallel to the last day of the week displayed on the calendar. Experimen ters (careful not to give too much direction on where to stand) then asked the participant to stand on the other side for the photo The calendar was strategically placed between two desks to ensure that participants could not stand too far from the confede rate. Because there are seven columns on the wall calendar (one for each day of the week), amount of columns placed by participants and confederates were used as an inconspicuous measure of physical distancing. Final Measures, Demographics and Debriefi ng After taking the picture, experimenters led participants back to their computers where participants entered their ideas generated during the brainstorming activity. They then completed open ended manipulation check items asking them to recall the key point of the news story they had read and to share what they think is the most impor tant issue that students face on a day to day basis. Content coding of responses will

PAGE 43

43 determine if participants in the experimental condition indicate threats to morals and values as an issue that students face significantly more so than students in the co ntrol condition. Participants were then presented with the ideas ostensibly written by the other participant (i.e., the confederate) and were asked to rate their agreement with each idea (on six point scale; 1 = Strongly disagree, 6 = Strongly agree). Agre ement ratings were collected as an additional measure of prejudice, as participants might be less willing to endorse recommendations from a person whom they perceive as a threat to values. completed the Three Domain Disgust Scale (TDDS; Tybur, Lieberman, & Griskevicius, 2009; Appendix M). This scale measures individual differences in disgust sensitivity in domains related to pathogenic transmission, sexuality, and morality. Because I made no specific hypotheses regarding this scale, I do not report it in the following results. However, these data can be used to determine whether responses toward targets differ for those who possess innate sensitivity toward related threats (i.e., values or health). Alt hough this scale is intended to measure state rather than trait characteristics, it could also prove useful as a manipulation check for the experimental condition (i.e., it is possible that people might show an increase in disgust sensitivity in morality d omains after exposure to the experimental condition). Finally, participants were asked to answer a number of demographics questions of interest (e.g., age, ethnicity, gender, sexual o rientation, etc.; Appendix N ) After completing these questions, particip ants were directed to alert the experimenter who then debrief ed the m to the true nature of the experiment and answered any questions

PAGE 44

44 participants might have After debriefing experimenter s thank ed participants for their time and award ed them participation credit.

PAGE 45

45 CHAPTER 3 RESULTS This study tests whether threat relevant stimuli affect responses toward members of groups perceived to pose associated threats. The study employed a 3 (target: atheist, person with HIV, or control) x 2 (threat condition: experi mental vs. control) completely between subjects design. Although I do not predict gender differences in response to any of the groups, participant gender was treated as an additional factor to allow for tests of gender differences. The full study therefore uses a 3 x 2 x 2 between subjects design. Full model ANOVAs were run on each dependent measure of interest to determine the effects of each independent variable on resulting measures of prejudice. If my hypotheses are supported, then for most dependent m easures I should expect to find a significant interaction between threat condition and target. However, planned comparisons, such as those that would be carried out in the event of such an interaction, present more focused tests of my hypotheses. Specifica lly, I predict that after reading a threat relevant news story (detailing moral decline), participants will respond with increased (related) prejudices toward a confederate described as an atheist (a values threat), but not toward confederates described as having HIV (a health threat), or a confederate who poses no specific threats (control targets). Analyses of each dependent measure are presented below. Correlations between all m easures are presented in Table 3 6 Manipulation Checks Manipulation checks w for moral issues were elevated in response to news stories in the experimental vs.

PAGE 46

46 control condition. Participants responded to two items assessing current concern regarding the morals of others ( r = .62, p < .001) and two items assessing current concern for the health of others ( r = .62, p < .001). Responses to each respective concern correlated highly and were thus combined to form one measure for each concern. Mean responses on manipulation check items for women and men on are presented in Table 3 1 Two (threat condition: experimental vs. control) by two (gender) ANOVAs were run on health concern and moral concern. No interaction was found between threat condition and gender for health concern, F (1.270) = .27, p significant two way interaction was found for moral concern, F (1, 270) = 7.55, p < .01, effect of gender on reported moral concern [ F (1, 270) = 6.82, p indicating that women reported greater moral concern overall ( M = 3.99, SD = 1.13) than men ( M = 3 .63, SD = 1.14). Analyses of simple main effects tests show that the experimental condition increased moral concern for women relative to the control condition [ F (1, 153) = 8.92, p F (1, 117) = 1.11, p = These manipulation check data suggest that the experimental condition successful ly raised conscious concern for morals/values for female participants, but not for male participants. However, as previously mentioned in the pilot data analyses, it is possible that moral concern is activated at an unconscious level rather than conscious level. There was one additional manipulation check item aimed at determining whether the experimental condition successfully raised awareness of threats to morals and

PAGE 47

47 values. Toward the end of the study participants were asked to answer an open ended quest ion about what they thought is the most important issue that students face on a day to day basis. These data were carefully coded to determine whether students indicated that the biggest issue students face is morals/values relevant (i.e., if the participa nt clearly indicated concern for morals or morality related behavior; for relevant while indicated that a s ignificantly greater proportion of participants in the experimental condition reported that morality issues are the biggest issues facing students (25/139 or 17.99%) than participants in the control condition (6/136 or 4.41%), x = 12.66, p < .001. Self r eport Measures of Prejudice After being randomly assigned to threat condition (experimental or control), participants were presented with personal information ostensibly provided by another participant in their experimental session. Participants were rando mly assigned to read information presented by an atheist (a values threat), a student who contracted HIV via blood transfusion (a health threat), or another student who provided no threat relevant information (a control target). I predicted that participan ts in the experimental threat condition would report increased prejudices toward atheists especially on measures related to values threats compared to participants in the control threat condition. I predicted that such increase would not be found for the o ther targets. Analyses on self reported measures of prejudice toward targets are presented below. Feelings toward Working with Other Participant Immediately after reading the information ostensibly provided by another participant, participants were asked g eneral questions regarding their feelings about

PAGE 48

48 working with the other participant. These four items reverse coded to represent negativity toward working with members of each target group demonstrated adequate gated to form one measure of negative feelings toward working with the other participant (Table 3 2 ). A full model 3 (target group) x 2 (threat condition) x 2 (gender) ANOVA did not indicate a significant three way interaction on feelings toward working w ith the other participant, F (2, 263) = 0.9, p way interactions between variables (all F s < 1.5, all p s > .20). There was, however, a significant main effect of target group [ F (2, 263) = 10.8, p < effects for threat condition or gender ( F s < 1, p s > .45). Simple effects tests of target group suggest that participants responded with no differences in negativity toward working with an atheist or a control target, t (182) = 1.01, p = .31, d = .15. However, participants reported greater negativity toward working with an atheist than someone with HIV [ t (180) = 4.3, p < .001, d = .64], and greater negativity toward working with a control target than someone with HIV [ t (180) = 3.98, p < .001, d = .59]. Although the full model design did not produce a significant three way interaction, planned comparisons were carried out to determine whether the specific hypotheses were supported. To determine whether participants in the experi mental condition were more negative toward working with atheists (but not toward other groups) compared to participants in the control condition. Collapsing across gender, participants did not report significant differences in negativity toward working wit h any one group in the experimental condition vs. the control condition (all t s < 1.25, all p s > .20). This pattern

PAGE 49

49 did not differ for female participants (all t s < .7, all p s > .50). Male participants indicated no differences in negativity toward working with atheists or people with HIV in the control vs. experimental conditions ( t s < .51, p s > .60). However, men reported greater negativity toward working with control targets (i.e., students) in the control condition vs. the experimental condition, t (37) = 1.99, p = .05, d = .64. In sum, participants indicated less negativity toward working with participants with HIV than toward atheists and control targets. Presentation of the news story did not seem to affect negativity toward working with other targets, ratings toward working with control targets, for whom men in the experimental condition expressed greater negativity than men in the control condition. Social Distancing After the appropriate items were reverse coded, the seven items that comprise the aggregated into one measure of social distancing (Table 3 2 ). The full model ANOVA did not indicate a significant three way interaction on reports o f social distancing, F (2, 263) = .39, p way interactions were found (all F s < 2.0, all p s > .20). Although there were no significant main effects for the threat condition or for gender ( F s < 1.0, p s > .60), there was a significant main effect of target group, F (1, 263) = 5.52, p effects tests suggest that, overall, participants reported marginally greater levels of social distancing toward atheists ( m = 3.36, sd = .81) than tow ard the control group [ m = 3.17, sd = .51; t (183) = 1.84, p = .07, d = .27], and reported significantly greater social distancing toward atheists compared to students with HIV [ m 3.01, sd = .69; t (181) = 3.07, p < .01, d = .45]. Participants also

PAGE 50

50 reporte d marginally greater social distancing toward the control group than toward students with HIV, t (180) = 1.80, p = .07, d = .27. Although the full model design did not produce a significant three way interaction, planned comparisons were carried out to determine whether the specific hypotheses were supported. To determine whether the experimental condition increased prejudices toward associated groups (atheists) but not others (students with HIV and the control group), I tested the simple effects of thr eat condition on social distancing. Collapsing across gender, participants did not report significant differences in social distancing toward any one group in the experimental condition vs. the control condition (all t s < 1.5, all p s > .17). This pattern d id not differ for female (all t s < 1.5, all p s > .20) or male participants (all t s < 1.0, all p s > .50). In sum, although participants did indicate greater social distancing toward atheists compared to other targets, my specific hypotheses were not support ed on reports of social distancing a generalized measure of prejudice. Next, I turned to specific measures of threat based prejudice to test my hypotheses. Threat Perceptions As with the pretesting data, participants reported perceptions of generalized t hreat posed by targets, as well as perceptions of threats to values and health posed by targets. The two questions for each respective threat correlated highly ( r s > .57, p s < .001) and were each combined to form respective threat measures. Means of all th reat perceptions items are presented in Table 3 3 Similar to the social distancing data, I first ran a full model design on reported threat perceptions, then looked at planned comparisons to test my hypothesis.

PAGE 51

51 General threat A full model ANOVA did not in dicate a significant three way interaction on reports of general threat, F (2, 261) = 2.47, p two way interactions between target and gender [ F (2, 261) = 4.27, p threat condition and ge nder [ F (1, 261) = 3.8, p significant main effects for threat condition or gender ( F s < 1.0, p s > .60), there was a significant main effect for target [ F (2, 261) = 5.61, p qualified by the significant interaction between target and gender. To decompose the target by gender interaction, I compared male and female responses on perceptions of general threat posed by members of each target group. Men and women did not differ in their responses toward participants in the control condition [ t (89) = .38, p = .7, d = .08] or students with HIV [ t (88) = 0.95, p = .35, d = .2]. Women, however, rated atheists as posing significantly greater general threats than did men, t (90) = 2.26, p < .05, d = .48. To decompose the threat condition by gender interaction I looked at overall differences in general threat in response to each story reported by men and women. Although the overall interaction was significant, specific contrasts suggested no significant differences in general threat between the experimental and control conditions for women [ t (152) = 0.97, p = .33, d = .16] or men [ t (117) = 1.32, p = .19, d = .24]. To look at the interaction in the other direction (i.e., comparing response s from men versus women) specific contrasts did not indicate significant differences in female and male t (134) = 0.6, p = .55, d = .10] or in the experimental condition, t (135) = 1.66, p = .10, d = .29.

PAGE 52

52 Simple effects tests were next run to further understand the significant main effect of target, indicating that atheists were perceived to pose significantly greater general threats than the targets in control condition [ t (181) = 3.27, p < .001, d = .48] and targets with HIV [ t (180) = 2.41, p < .05, d = .36]. Perceptions of general threat did not differ, however, for control targets or targets with HIV, t (179) = 0.97, p = .33, d = .14. Finally, although the target by threat conditi on interaction was not significant, specific contrasts were run to test whether the hypotheses were supported. Collapsing across gender, participants did not report significant differences in general threat perception toward any one group in the experiment al condition vs. the control condition (all t s < 1, all p s > .45). This pattern did not differ for female (all t s < 1.85, all p s > .07) or male participants (all t s < 1.8, all p s > .07). In sum, participants did not report increased perceptions of general threat toward any of the groups in the experimental condition compared to the control condition. Participants did, however, perceived atheists to pose greater general threats than other targets, and women perceived atheists to pose greater general threats than did men. Values threat As with perceptions of general threat, the full model ANOVA did not indicate a significant three way interaction on reports of values threat, F (2, 262) = 1.46, p = .01. There was, however, a significant two way interaction between threat condition and gender, F (1, 262) = 8.2, p way interactions all F s < 1.8, all p s > .15). Additionally, there was a significant main effec t of target [ F (2, 262) = 34.6, p F s < 1.5, all p s > .2).

PAGE 53

53 To decompose the interaction between threat condition and gender I tested for differences in values threat in respons e to threat condition within each gender. Specific experimental vs. control condition, t (153) = 1.07, p = .28, d = .17. Men, however, reported greater perceptions of val ues threat in the control condition vs. the experimental condition, t (117) = 2.23, p < .05, d = .41. Thus, results for men on perceptions of values threat were in the opposite direction than expected. I also examined the interaction in the other direction testing for differences between men and women within each threat condition. Men and women did not differ in their reports of values threat in the control condition, t (134) = 1.01, p = .32, d = .17. However, women reported greater perceptions of values t hreat in the experimental condition than men did, t (136) = 2.28, p < .05, d = .40. I ran simple effects tests to further understand the significant main effect of target. Simple effects tests indicated that atheists were perceived to pose significantly gr eater values threats than control targets [ t (182) = 6.08, p < .001, d = .90] or targets with HIV [ t (180) = 6.97, p < .001, d = 1.04]. Perceptions of values threat did not differ, however, for control targets or targets with HIV, t (180) = 1.16, p = .25, d = .17. Again I ran specific contrasts to determine whether the hypotheses were supported. Collapsing across gender, participants did not report significant differences in social distancing toward any one group in the experimental condition vs. th e control condition (all t s < 1.2, all p s > .25). Women reported marginally greater perceptions of values threat in response to atheists in the experimental condition than the control condition [ t (49) = 1.88, p = .07, d = .53], but did not differ in respo nses toward students

PAGE 54

54 with HIV or control targets in differing conditions ( t s < ,8, p s > 40). Men, on the other hand, reported that participants with HIV posed significantly greater values threats in the control condition than the experimental condition [ t (37) = 2.12, p < .05, d = .67], but did not report differences for atheists or control targets in the experimental vs. control conditions ( t s < 1.4, p s > .19). In sum, reports of values threat in response to experimental conditions suggest that my hypot heses were only marginally supported. Atheists were indeed perceived as greater threats to values than were other targets. However, while women in the experimental condition reported marginally increased perceptions of values threat toward atheists, men di d not. Men also reported significantly lower ratings of values threat toward groups overall after reading the story detailing moral decline a pattern in the opposite direction than predicted. Health threat As with perceptions of general threat and values t hreat, the full model ANOVA for perceptions of health threat did not indicate a significant three way interaction, F (2, 262) = 0.36, p way interactions between variables (all F s < 1.8, p s > .19). There w as, however a significant main effect of target [ F (2, 262) = 37.52, p condition or gender ( F s < 2.4, p s > .12). As expected, simple effects tests of target condition indicated that targets with HIV were perceived to pose significantly greater health threats than control targets [ t (180) = 6.08, p < .001, d = .90] or atheists [ t (180) = 7.27, p < .001, d = 1.07]. Perceptions of health threats posed by control targets and atheists did not differ, t (18 2) = 1.45, p = .15, d = .21.

PAGE 55

55 I predicted that perceptions of health threat (a threat unrelated to the experimental news story) would not differ for participants in the experimental condition vs. the control condition. This hypothesis was supported, as perc eptions of health threat collapsed across gender did not increase for any target group in the experimental condition compared to the control condition (all t s < 1.3, p s > .20). Analyses within gender suggest that female participants reported marginally gre ater perceptions of health threats toward atheists in the experimental condition compared to the control condition [ t (49) = 1.86, p = .07, d = .53], but did not indicate differences for control targets or targets with HIV ( t s < 1.0, p s > .38). Men did not report any differences in perceptions of health threat toward any target groups in the experimental condition vs. the control condition (all t s < 1.15, all p s > .25). In sum, reading a news story detailing threats to morals/values did not increase percep tions of health threat (an unrelated threat) among target groups, with the atheists the group perceived to threaten values. Emotional Reactions Moral disgust The two items measuring moral disgust correlated highly ( r = .79, p < .001) and were averaged to form one measure (Table 3 4 ). The full model ANOVA for feelings of moral disgust indicated no significant three way interaction, F (2, 263) = 0.24, p < .01. There was a marginally significant two way interaction between threat condition and gender, F (1, 263) = 3.3, p way interactions, F s < .08, p s > .9). There was also a significant main effect of target [ F (2, 262) = 3.98, p < = .03], though no main effects for threat condition or gender ( F s < 0.6, p s > .4).

PAGE 56

56 To decompose the interaction between threat condition and gender I looked at the simple effects of threat condition within each gender. Specific contrasts suggest that neither women nor men reported differences in overall feelings of moral disgust in the experimental vs. control condition [ t (154) = 1.12, p = .26, d = .18 for women, and t (117) = 1.34, p = .18, d within conditions suggests that men and women did not differ in their reported feelings of moral disgust in the control condition [ t (134) = 1.72, p = .08, d = .30] or in the experimental condition, t (137) = 0.68, p = .5, d = .12. As expected, simple effe cts tests indicated that participants reported greater feelings of moral disgust in response to atheists than control targets [ t (183) = 2.71, p < .01, d = .40]. Participants reported no differences, however, in feelings of moral disgust toward atheists an d targets with HIV [ t (181) = 1.37, p = .17, d = .2], or toward targets with HIV compared to control targets, t (180) = 1.54, p = .12, d = .23. I predicted that feelings of moral disgust would increase toward atheists in the experimental condition vs. the control condition, but not for other groups. This hypothesis was not supported as perceptions of moral disgust collapsed across gender did not increase for any target group in the experimental condition compared to the control condition (all t s < 0.5, p s > .65). This pattern did not differ for female participants (all t s < 1.5, p s > .14) or male participants ( t s < 1.05, p s > .3). In sum, participants reported greater feelings of moral disgust toward atheists compared to control targets, as expected. Partic ipants did not, however, report greater feelings of moral disgust toward atheists than toward targets with HIV. More importantly, neither male nor female participants in the experimental condition reported increased

PAGE 57

57 feelings of moral disgust toward atheist s, as predicted, or toward either of the other target groups. Physical disgust were thus averaged to form one measure (Table 3 4 ). The full model ANOVA for feelings of physical disgust indicated a significant three way interaction, F (2, 263) = 3.61, p way interactions (all F s < 1.5, p s > .25). There were, however, significant main effects of target [ F (2, 263) = 9.29, p < .001, .07] and gender [ F (1, 263) = 7.76, p by the interaction. To decompose the three way interaction I examined the effects of threat condition and gender within each target. For the control condition, there was no significant interaction between threat condition and gender [ F (1, 88) = 3.17, p were there significant main effects of gender ( F = 2.93, p F = .15, p For the atheist condition, there wa s a significant threat condition by gender interaction, F (1, 89) 5.19, p < .05, condition or gender ( F s < 1.15, p s > .28). Simple effects tests within threat condition suggest that men reported significantly greater physical disgust toward atheists in the control condition than did women, t (49) = 2.34, p < .05, d = .64. Men and women did not differ in reported physical disgust toward atheists in the experimental condition, however, t (40) = 0.89, p = .38, d = .28. Simple effects of threat condition within gender suggest that neither women nor men reported differences in feelings of physical disgust

PAGE 58

58 toward atheists in the experimental condition vs. the control condition (both t s < 1.8, p s > .10). For responses toward targets with HIV, there was no significant interaction between threat condition and gender, F (1. 86) = 1.58, p was no significant main effect of threat condition ( F = .08, p a marginally significant main effect of gender [ F = 3.74, p men reported greater physical disgust in response toward people with HIV than did women. Finally, specific contrasts were run on reports of physical disgust in response to ea ch target group. Following the threat based hypotheses, participants should report greater feelings of physical disgust toward targets with HIV compared to atheists and control targets, and these should not be affected by the experimental story condition ( as it details unrelated threats). Participants indeed reported greater feelings of physical disgust in response to targets with HIV compared to control targets [ t (180) = 3.07, p < .01, d = .45] and atheists [ t (181) = 3.72, p < .001, d = .55], but did not report any differences in physical disgust felt toward control targets compared with atheists, t (183) = .66, p = .51, d = .1. As predicted, participants reported greater feelings of physical disgust toward participants with HIV than toward atheists or c ontrol targets. Also as predicted, participants did not report increased feelings of physical disgust toward any groups in the experimental condition compared to the control condition. Men, however, reported significantly greater feelings of physical disgu st felt toward atheists in the control condition than did women.

PAGE 59

59 Threat related d iscri minatory i ntentions As a more subtle measure of responses to threat perception, participants indicated their willingness to perform behaviors related to values and health in regards to members of each target group. I predicted that participants would report increased values related discriminatory intentions toward atheists in the experimental condition vs. the control condition, but not toward other groups. I also predict that participants would display greater health related discriminatory intentions toward targets with HIV than toward other targets, but these responses should not change as a result of threat condition. Means for threat related discriminatory intentions ar e presented in Table 3 5 Values related discriminatory intentions Items measuring behavioral intentions in values threat related domains coded to indicate negative responses and were aggreg ated to form one measure of values related discriminatory intentions. The full model ANOVA for values related discriminatory intentions did not indicate a three way interaction, F (2, 263) = 1.11, p significant intera ction between target and gender [ F (2, 263) = 3.5, p though other two way interactions were not significant ( F s < 1.8, p s > .18). Although qualified by the significant target by gender interaction, there was a significant main effect of t arget [ F (2, 263) = 43.58, p F s < .5, p s > .57). To decompose the interaction between target and gender I looked at levels of values related discriminatory intentions reported by men and women in response t o each target. Specific contrasts suggest that women reported marginally greater levels of

PAGE 60

60 values related discriminatory intention toward atheists than did men, t (91) = 1.76, p = .08, d = .37. Men and women did not differ, however, in their levels of valu es related discriminatory intentions reported toward control targets [ t (90) = .61, p = .54, d = .13] or toward targets with HIV, t (88) = 1.5, p = .13, d = .33. Simple effects tests on values related discriminatory intentions toward each target group, collapsed across gender, suggest that participants reported no differences in discriminatory intentions toward atheists and control targets, t (183) = 1.58, p = .12 d = .23. They did, however, report significantly greater values related discriminatory intentions toward atheists than toward targets with HIV [ t (181) = 8.61, p < .001, d = 1.27], and greater discriminatory intentions toward control targets than toward targets with HIV, t (180) = 8.02, p < .001, d = 1.19. Again, although the target group by threat condition interaction was not significant, I ran specific contrasts to determine whether the hypotheses were supported. Collapsed across gender, participants did not report significant differences in values related discriminatory intentions toward any one group in the experimental condition vs. the control condition (all t s < 1.2, all p s > .25). This pattern did not differ for women (all t s < 1.5, p s > .14) or for men (all t s < 1.7, p s > .1). In sum, participants did not show any increase in values related discriminatory intentions toward atheists (or any other group) in response to the experimental condition. Targets did report greater values related discrimi natory intentions toward atheists compared to targets with HIV, as expected, but (unexpectedly) also reported greater values related discriminatory intentions toward control targets than targets with HIV.

PAGE 61

61 Health related discriminatory intentions Items meas uring behavioral intentions in health threat related domains also coded to indicate negative responses and were aggregated to form one measure of values related discriminatory intentions. Th e full model ANOVA for health related discriminatory intentions did not indicate a three way interaction, F (2, 263) = .24, p significant two way interactions in the model (all F s < 1.5, p s > .25). There was a s ignificant main effect of target [ F (2, 263) = 7.92, p other variables ( F s < .5, p s > .55). Simple effects of target collapsed across gender suggest that participants reported greater health related discriminatory inten tions toward control targets than toward atheists, t (183) = 4.59, p < .001, d = .68. Unexpectedly, participants also reported significantly greater health related discriminatory intentions toward control targets than toward targets with HIV, t (180) = 2.0 3, p < .05, d = .3. Last, participants reported marginally greater health related discriminatory intentions toward targets with HIV than toward atheists, t (181) = 1.82, p = .07, d = .27. Finally, although the target group by threat condition interaction was not significant, specific contrasts were run to determine whether the experimental condition increased reported health related discriminatory intentions toward members of any of the target groups. Participants, collapsed across gender, did not report significant differences in health related discriminatory intentions toward any one group in the experimental condition vs. the control condition (all t s < 1.3, all p s > .2). This pattern did not differ for women (all t s < 1.6, p s > .13) or for men (all t s < .4, p s > .7).

PAGE 62

62 As expected, participants did not show any increase in health threat related discriminatory intentions toward members of any group in response to the experimental condition. Targets did report marginally greater health related discriminator y intentions toward people with HIV compared to atheists, also as expected. However, participants unexpectedly reported greater health related discriminatory intentions toward control targets than atheists or targets with HIV. Dyadic Interactions and Following Measures Described below are the results for behavioral measures taken during the dyadic interaction between participants and confederates and measures taken immediately following the interaction. Means for each of these measures are presented in Table 3 6 Measures of Physical Distance Experimenters subtly collected two measures of physical distance during the interaction between participants and confederates: how many chairs the participants placed between themselves and confederates when selec ting a seat and how much space they placed between themselves and confederates when the experimenter took a photo of the two of them. I predicted that participants in the experimental condition would show greater distancing (a standard measure of behaviora l prejudice) toward atheists than would participants in the control condition. It is also possible that participants will show greater distancing toward targets with HIV than toward other targets, in general, due to perceptions of health threat Chairs A f ull model ANOVA was run on the number of chairs participants placed between themselves and confederates when taking a seat during the interaction (ranging from 0 to 4). There was no significant three way interaction for number of chairs, F (2, 261) =

PAGE 63

63 .26, p way interactions in the model (all F s < 1.6, p s > .22). There was a significant main effect of gender, however [ F (1, 261) = 6.39, p ignificantly more from confederates than did women. There were no significant main effects for the other variables ( F s < .5, p s > .6). Although the target by threat condition interaction was not significant, specific contrasts were run to determine wheth er participants in the experimental condition increased the amount of chairs placed between themselves and confederates. Collapsed across gender, participants in the experimental condition placed marginally fewer chairs between themselves and control targe ts than participants in the control condition, t (90) = 1.81, p = .07, d = .37. There were no significant differences in number of chairs placed between atheists or people with HIV in the experimental condition vs. the control condition, however ( t s < .6, all p s > .6). For female participants, there were no differences in number of chairs placed between themselves and confederates in the experimental condition vs. control condition for any of the targets (all t s < 1.4, p s > .18). This pattern was similar fo r male participants (all t s < 1.2, p s > .2). Overall, physical distancing measured by amount of chairs placed between participants and confederates did not support my hypotheses. Participants in the experimental condition, however, placed marginally greate r distance between themselves and confederates than did participants in the control condition, and men placed significantly greater space between themselves and confederates than did women.

PAGE 64

64 Physical distance during picture Independent coders recorded the number of spaces on the wall calendar (indicated by days of the week) participants placed between themselves and averaged to form one measure of physical distance demonst rated during the picture. A full model ANOVA indicated no significant three way interaction for distance during the picture, F (2, 259) = .53, p significant two way interactions (all F s < .5, p s > .5) or main ef fects in the model (all F s < 1.8, p s > .18). Specific contrasts suggest that participants did systematically differ in distance placed between themselves and members of any of the targets in the experimental condition vs. the control condition (all t s < 8, all p s > .4). This pattern did not differ for women (all t s < 1.1, p s > .25) or for men (all t s < .7, p s > .55). In short, measures of physical distancing did not differ and thus did not support my hypotheses. Disagreement with I deas agreement with the ideas ostensibly proposed by confederates. After being reverse averaged to form one measure of disagreement with participants. A full model ANOVA determined that there was no significant three way interaction for disagreement with target, F (2, 263) = .44, p significant interaction betwee n target and gender [ F (2, 263) = 3.21, p There was also a marginal interaction between target and threat condition [ F (2, 263) = 2.45, p F (1, 263) = .93, p

PAGE 65

65 = .34, Finally, although qualified by the significant two way interaction, there was a marginally significant main effect of gender [ F (1, 263) = 3.69, p but no main effects of the other variables ( F s < 1, p s > .35). To decompose the interactio n between target and gender I looked at levels of disagreement reported by men and women in response to each target. Specific contrasts suggest that men reported significantly greater disagreement toward ideas contributed by control targets than women did, t (90) = 3.14, p < .01, d = .64. Men and women did not differ, however, in their levels of disagreement toward ideas contributed by atheists [ t (91) = .4, p = .69, d = .08] or targets with HIV, t (88) = .88, p = .38, d = .19. To decompose the marginal i nteraction between target and threat condition, I looked at the effects of threat condition within each target. These results indicated that there were no differences in ratings of disagreement in the experimental vs. control condition for any of the targe ts (all t s <1.65, all p s > .1). Looking at ratings within gender suggested that this pattern did not differ for women (all t s < 1.2, p s > .24) or men (all t s < 1.3, p s > .2). Although some differences in disagreement were found (i.e., between women and men when rating agreement with ideas proffered by control targets), the data did not support my hypotheses. Participants were not more likely to disagree with atheists than other targets, and there were no effects of the experimental condition.

PAGE 66

66 Tabl e 3 1 Manipulation c heck r esults Health Concern Moral Concern Moral Behaviors Women Men Women Men Women Men Condition M SD M SD M SD M SD M SD M SD Control 3.82 1.21 3.69 1.18 3.73 1.23 3.75 1.18 4.31 .74 4.37 .79 Experimental 3.84 1.07 3.57 1.11 4.26* .97 3.53 1.1 4.57 .68 4.31 .77 Note. Notates a significant difference between responses in the control and experimental condition.

PAGE 67

67 Table 3 2 General prejudice m easures Control Atheists HIV DV Women M ( SD ) Men M ( SD ) Women M ( SD ) Men M ( SD ) Women M ( SD ) Men M ( SD ) Negativity Control 2.73 (.74) 3.12 (.84) 2.97 (1.08) 2.86 (1.01) 2.42 (.91) 2.35 (.67) Experimental 2.76 (.72) 2.66 (.57) 3.15 (1.0) 2.71 (.92) 2.32 (.65) 2.4 (.69) Social Distancing Control 3.19 (.54) 3.33 (.56) 3.3 (.81) 3.21 (.83) 3.01 (.81) 2.99 (.59) Experimental 3.05 (.52) 3.21 (.39) 3.58 (.84) 3.35 (.75) 2.94 (.54) 3.13 (.81) Note. Higher numbers indicate greater levels of respective prejudice toward targets.

PAGE 68

68 Table 3 3 Threat measures Control Atheists HIV DV Women M ( SD ) Men M ( SD ) Women M ( SD ) Men M ( SD ) Women M ( SD ) Men M ( SD ) General Threat Control 1.93 (.64) 1.97 (1.0) 2.33 (1.2) 2.32 (.98) 1.82 (.9) 2.11 (.65) Experimental 1.83 (.84) 1.88 (.87) 2.93 (1.13) 1.79 (.8) 2.04 (.77) 2.1 (1.03) Values Threat Control 1.74 (.67) 1.95 (.96) 2.78 (1.52) 2.95 (1.48) 1.7 (.87) 1.97 (.83) Experimental 1.93 (.93) 1.8 (.78) 2.39 (1.2) 1.5 (.55) 1.75 (.68) 1.5 (.55) Health Threat Control 1.65 (.63) 1.74 (.89) 1.31 (.57) 1.59 (.73) 2.26 (1.28) 2.8 (1.09) Experimental 1.5 (.63) 1.53 (.64) 1.59 (.48) 1.37 (.48) 2.35 (1.07) 2.62 ( 1.39) Note. Higher numbers indicate greater levels of perceived threat toward targets.

PAGE 69

69 Table 3 4 Threat relevant discriminatory i ntentions Control Atheists HIV DV Women M ( SD ) Men M ( SD ) Women M ( SD ) Men M ( SD ) Women M ( SD ) Men M ( SD ) Values related Intentions Control 3.71 (.71) 3.53 (.87) 3.84 (1.02) 3.74 (.95) 2.58 (.97) 3.09 (.67) Experimental 3.68 (.84) 3.66 (.52) 4.22 (.8) 3.57 (.95) 2.57 (.75) 2.65 (.95) Health related Intentions Control 3.84 (.66) 3.62 (.76) 3.23 (1.06) 3.15 (1.19) 3.16 (1.53) 3.49 (.96) Experimental 3.75 (.92) 3.56 (.6) 2.76 (1.17) 3.07 (1.01) 3.32 (1.07) 3.64 (1.5) Note. Higher numbers indicate greater levels of discriminatory intention toward targets.

PAGE 70

70 Table 3 5 Behavioral distancing and disagreement with target r ecommendations Control Atheists HIV DV Women M ( SD ) Men M ( SD ) Women M ( SD ) Men M ( SD ) Women M ( SD ) Men M ( SD ) Number of Chairs Control 1.43 (.59) 1.79 (1.18) 1.43 (.74) 1.64 (.9) 1.32 (.85) 1.61 (1.14) Experimental 1.2 (.66) 1.45 (.6) 1.45 (.74) 1.79 (.98) 1.5 (.65) 1.57 (.68) Calendar Distance Control 1.31 (.64) 1.52 (1.12) 1.24 (.87) 1.3 (.66) 1.52 (1.08) 1.49 (.76) Experimental 1.45 (.8) 1.49 (.57) 1.21 (.74) 1.46 (.91) 1.2 (1.09) 1.52 (.78) Disagreement Control 1.91 (.69) 2.6 (1.06) 1.99 (.52) 1.88 (.78) 1.84 (.57) 2.11 (.9) Experimental 1.88 (.59) 2.22 (.76) 2.28 (1.14) 2.23 (1.24) 2.01 (.66) 2.0 (.49) Note. Higher numbers indicate greater levels of physical distance and disagreement toward targets.

PAGE 71

71 Table 3 6 Correlations of all dependent variables measured in study p roper Variable 1 2 3 4 5 6 7 8 9 10 11 12 1. Negativity .64** .39** .47** .11 .41** .30** .59** .34** .07 .01 .22** 2. Social Distance .43** .49** .24** .48** .35** .63** .48** .09 .00 .15* 3. General Threat .70** .38** .56** .42** .27** .14* .00 .06 .09 4. Values Threat .14* .60** .26** .48** .01 .03 .04 .07 5. Health Threat .31** .54** .08 .34** .02 .04 .05 6. Moral Disgust .60** .30** .16** .00 .01 .10 7. Physical Disgust .04 .30** .03 .03 .13* 8. Values related Intentions .47** .12 .07 .14* 9. Health related Intentions .01 .03 .12* 10. Chairs (Distance) .06 .02* 11. Calendar Distance .04 12. Disagreement Note p < 0.05, ** p < 0.01

PAGE 72

72 CHAPTER 4 DISCUSSION A growing body of literature suggests that humans possess cognitive tools aimed at recognizing both threats and opportunities in th eir social environment (Neuberg et al., 2011). Using a threat based approach to prejudice (Cottrell & Neuberg, 2005), I predicted that innocuous threat relevant stimuli could increas particular threats, and affect subsequent reactions toward groups posing associated threats. I tested this idea by presenting participants with a news story either detailing threats to values or not mentioning any particular threats I predicted that after reading a threat relevant story, participants would increase prejudices toward groups who a lth ough not mentioned in the story a re perceived to pose related threats, but not toward members of groups who pose unrelated threats. Two sets of pilot data were carried out to test the viability of the ideas and manipulations used in the study. Pilot data on potential target groups supported predictions regarding threat perceptions of those respective groups. That is, compared to a control group, participants perceived atheists to threaten values and people with HIV to threaten health. In line with these predictions, participants reported greater feelings of moral disgust toward atheists (compared to a control group and people with HIV) and greater feelings of physical disgust toward people with HIV (compared to a control group and atheists). Participants also indicated greater discriminatory intentions toward atheists in values related domains and greater discriminatory intentions toward peo ple with HIV in health related domains. A second pilot study analyzed responses toward threat relevant vs. threat irrelevant news stories. Although participants did not indicate increased concern for

PAGE 73

73 morals and values following a story detailing moral decline, they did report increased negative affect and increased discriminatory intentions toward atheists (a values threat) but not toward people with HIV (a health threat). Taken together, these pilot studies provided early evidence of the ideas tested i n the current research participants viewed atheists as threats to values and increased prejudicial responses toward them after encountering threat relevant stimuli. Results of the Current Study The current study aimed to test the hypotheses regarding threa t relevant stimuli and responses toward associated groups in further detail, including a real life behavioral context. The goal of measuring behaviors toward targets in an actual interaction was analytic r eview suggesting that actual behaviors are largely overlooked in the prejudice literature. To measure such an interaction, in addition to traditional self report measures of prejudice, an elaborate cover story was created. Convincing participants that they would be interacting with another participant, and that the experimenters were interested in exploring how people form impressions of others, allowed us to both manipulate the desired information about targets and measure actual behavioral responses towar d them. Although the experimenters were trained extensively and the study proper was designed to offer tight control, the results were not as straight forward as those found in the pilot tests. One major concern of the study proper regards the effectivenes s of the news story manipulation. The data do not offer conclusive evidence that this particular manipulation worked as planned. For example, women reported greater concern for morals and values following the experimental news story, though men did not. Wh en asked to state what they thought was the most important issue facing students today,

PAGE 74

74 participants in the experimental condition were more likely to report moral issues than those in the control condition. However, the open ended nature of this question could have been influenced by demand characteristics in the study. In short, any effects of the news story manipulation observed in the study proper should be interpreted with caution, as it is not clear whether the manipulation worked as planned. The da ta do offer strong support for the proposed threat perceptions of the target groups selected for the study. Atheist confederates, for example, were perceived to pose greater values threats than the other targets. Additionally, participants reported greater feelings of moral disgust in response to atheists compared to targets with HIV and greater values related discriminatory intentions toward atheists than control targets. Targets with HIV, on the other hand, were perceived to pose greater threats to health than atheists or control targets, and elicited greater feelings of physical disgust compared to these targets. Likewise, participants indicated greater health related discriminatory intentions toward targets with HIV compared to atheists. There were, howe ver, some unexpected prejudices reported toward the control group (students). For example, participants reported greater values related discriminatory intentions toward control targets than toward targets with HIV and reported marginally greater social dis tancing (a measure of general prejudice) toward control targets than targets with HIV. Although the data suggest that the chosen group manipulations adequately served their intended purposes, the data do not offer strong support for the specific predict ion regarding the effect of threat relevant stimuli on groups posing associated threats. A significant three way interaction between all variables was found for only one dependent measure physical disgust. Additionally, although there were fairly consisten t effects of

PAGE 75

75 target and gender on dependent measures, the hypothesized threat condition by target interaction was not significant for any of the dependent measures. Finally, the specific hypothesis regarding the effects of threat relevant stimuli on member s of groups perceived to pose associated threats was slightly supported in only one instance women in the experimental condition reported marginally increased perceptions of values threats toward atheists compared to women in the control condition. Altoget her, the data gathered from the study proper provide novel support for threat perceptions and resulting prejudices, but do not offer convincing evidence of changes in threat perception in response to innocuous threat relevant stimuli. Contributions and Lim itations Although the specific hypothesis of the current project that threat relevant stimuli would increase prejudices toward members of groups representing associated threats was not supported by the study proper, the data (including those from the prete sting studies) still offer a significant contribution to the literature on threat based prejudices. For example, these data measure threat perceptions of novel groups and offer experimental support for the theoretical conception of threat based prejudices across a variety of measures participants consistently indicated that they perceived atheists to pose threats to values and perceived people with HIV to pose threats to health, and emotional and discriminatory responses toward these groups corresponded wit h these threat perceptions. Additionally, these data contribute evidence to the role of perceptions of threats to values on prejudices toward others, while much of the existing research on threat based prejudices has focused on the effects of perceived thr eats to health and safety (for a review see Neuberg et al., 2011). These data are also the first, to my knowledge, to measure perceptions of atheists as posing threats to values. These

PAGE 76

76 data therefore contribute to the currently small body of research asses sing specific perceptions and prejudices toward atheists (e.g., Gervais et al., 2011). Finally, although the results of the full study and live interaction did not work out as hypothesized, the pilot data did offer support for the key hypothesis of the stu dy. The significant effects of threat relevant information on subsequent prejudices in the pilot study suggest that the key idea of the current project is worth further examination. Comparisons of data recorded in the pilot study and the study proper beg t he question: Why were there significant effects of threat relevant information in one study but not the other ? There are a number of limitations to both studies that could be considered to address this question. For one, the pilot study might have reaped the benefits of truly anonymous data collection. In modern American society it is largely considered unacceptable to display outright biases against members of other groups (Dovidio & Gaertner, 1991; Fiske, 2002) and participants are often motivated to ap pear unprejudiced (Dunton & Fazio, 1997). Participants in the live interaction, who completed all measures in a laboratory setting and were made keenly aware that there responses were being recorded, may therefore have been influenced by social desirabilit y affects. A look at some of the overall patterns in the data might offer indirect support for this possibility although participants in the study proper identified targets with HIV as posing threats to health and responded with increased feelings of physi cal disgust, they evaluated these same targets more positively overall than the other targets, including control targets. This same effect was not found for participants in the pilot study who may have experienced increased perceptions of anonymity by comp leting the study via the internet. Unfortunately, measures of social desirability were not measured in this

PAGE 77

77 study and therefore cannot be tested to determine whether they affected prejudiced responses toward targets. Another potential limitation to the cur rent research is the strength of the experimental manipulation. Although the data do offer some support that the experimental manipulation increased concern for morals/values, this support is not conclusive. It is possible that awareness of threats to mora ls/values was elevated only superficially, never breaching the level of unconscious concern. Similar research paradigms that experimentally manipulate threat awareness typically rely on unconscious activation of respective threats. For example, research on the behavioral immune system consistently finds effects only when participants have been unconsciously primed with cues to pathogen prevalence (Schaller & Park, 2011). Terror Management research, has found that affects of mortality salience manipulations only occur after the manipulation has faded from conscious awareness (e.g., after participants complete some sort of unrelated filler task; Pyszczynski et al., 2004). Additionally, appropriate measures of unconscious awareness for threats (i.e., implicit a ssociation tests or stem completion tasks; Greenwald et al., 2009; Pyszczynski et al., 2004) were not employed in the current study. Future Directions Despite the limitations of the current research, the data offer some novel implications and direction for future studies. Although support for the key hypothesis was inconclusive, participants across studies reported consistent negativity toward atheists. As of 2012, only one article published in popular social psychological literature systematically exami ned prejudices toward atheists (Gervais et al., 2011). Gervais et al. (2011) found that perceptions of atheists are marked by significant distrust. Coupled

PAGE 78

78 with these previous findings, the current data add significantly to the existing knowledge of anti a theist prejudice, and suggest that such prejudice is pervasive and worthy of further exploration. Are atheists actually less trustworthy than others? Are other groups of people (e.g., agnostics or non practicing believers) mistaken as atheists and/or perce ived with similar levels of distrust and perceptions of values threat? Do prejudices toward atheists or other non believers affect their everyday lives? These are among the many empirical questions regarding this social group that have not been explored in the social psychological literature. As previously mentioned, the current studies offer inconclusive results regarding the effects of innocuous threat relevant information on subsequent prejudices. The data do offer evidence, however, that the key hypoth esis of interest is worthy of further exploration. As the data currently stand, participants who reported perceptions under truly anonymous conditions displayed greater prejudices toward associated groups after being exposed to threat relevant information, while those who were being monitored did not. Researchers have long been interested in the effects of media on psychological processes such as aggression (Anderson & Bushman, 2002) and stereotyping (Coltrane & Messineo, 2000). Various forms of media (e.g. the internet, magazines, newspapers, television, etc.) bombard us with threat relevant information on a daily basis. It is possible that threat relevant media affect us in some situations (e.g., when there is not pressure to control prejudiced responses) but not in others. It is also possible that some forms of threat relevant media can be easily dismissed (e.g., information that goes against previous beliefs or comes from an untrustworthy news source) while others linger with us and affect our thoughts and behaviors. It is therefore important for

PAGE 79

79 researchers to continue to explore the effects of seemingly innocuous information on threat based prejudices and other social judgments. Conclusion Taken together, the studies in the current project offer some intriguing findings. First, they provide additional evidence for a threat based theoretical conception of prejudice, while simultaneously focusing on groups (e.g., atheists and people with HIV) and threat perceptions (i.e., values threats) not often explored in this literature. The current project is also the first to systematically measure the threat perceptions and threat based reactions toward atheists, and the results across three studies pro vide consistent evidence that atheists are perceived to threaten values. Data from the pretests also offer preliminary evidence that, in some cases, threat relevant information can indeed affect responses toward seemingly unrelated groups perceived to pose associated threats. In all, the current research suggests that social threat recognition mechanisms absolutely affect our judgments and responses toward others, and there are still many avenues for researchers to explore to gain a better understanding of how these mechanisms affect social cognition and behavior in an increasingly complex society.

PAGE 80

80 APPENDIX A EXPERIMENTAL NEWS ST ORIES AND FOLLOW UP QUESTIONS Experimental (i.e., Threat relevant) News Story Student Morals Declining Student morals appear to be declining. In a recent poll, university students report that traditional values such as loyalty and fidelity are less important to them compared to students in previous years. Additionally, students today report that they lie (t o both their friends and parents) and cheat (on both significant others as well as school work) on a regular basis. The researchers who conducted the study say that they expect to see morals continue to decline until researchers discover the sources of the se problems. Control (i.e., Threat irrelevant) News Story Dental School Expanding University of Florida officials plan to expand the dental school in coordination with other Florida Universities. Florida A&M and University of Central Florida are working with UF to build a joint dental program that will pool resources to improve the educational and service opportunities offered at each respective university. The joint efforts will additionally help fund a numb er of renovations for the College of Dentistry. Representatives from the dental school say that they expect the expansion to offer many benefits for future dental school graduates. Questions Immediately Following News Story Please take a moment to think about and provide your answers to the following questions: What is the key point of the article you just read ? In your opinion, how do the issues presented in the article affect you ?

PAGE 81

81 APPENDIX B PARTICIPANT INFORMATION Please answer the following questions. As mentioned before, your responses are confidential. These demographic questions only serve to help us explore the sample helping us with our study. What is your gender? ______ Male ______ Female How old are you? ______ years Year in school: _______________ Major: _______________ What are your plans for the future? (Please answer briefly.) What is one event in your life that helped define who you are today ? (Please answer briefly.)

PAGE 82

82 APPENDIX C MEASURES OF CURRENT CONCERN FOR RELEVANT AND NON RELEVANT THREATS Please indicate the extent to which you feel each of the following in response to the article: 0 1 2 3 4 5 Not At All Extremely Concerned Concer ned Measures of Concern for Morals/Values Right this moment, how concerned are you with the values of other students? Right this moment, how concerned are you about the moral issues society faces? Measures of Concern for Health Right this moment, ho w concerned are you with the health of other students? Right this moment, how concerned are you about the health issues society faces? Additional Measures of Morality Concerns 0 1 2 3 4 5 Strongly Disagree Strongly Agree for people to have high moral standards, even when it means disagreeing with others. In order to get along with others, sometimes you must do things that go against your values. (reverse scored) Being honest is more important than being nice. Sometimes (reverse scored)

PAGE 83

83 APPENDIX D PARTICIPANT INFORMAT ION SHEET FOR VALUES THREAT TARGET Please answer the following questions. As mentioned before, your responses are confidential. These demographic questions only serve to help us explore the sample helping us with our study. What is your gender? X Male ______ Female How old are you? 19 years Year in school: Junior Major: Psychology What are your plans for the future? (Please answer briefly.) I plan to go to graduate school. What is one event in your life that helped define who you are today ? (Please answer briefly.) When I decided that I no longer believe in God.

PAGE 84

84 APPENDIX E PARTICIPANT INFORMAT ION SHEET FOR HEALTH THREAT TARGET Please ans wer the following questions. As mentioned before, your responses are confidential. These demographic questions only serve to help us explore the sample helping us with our study. What is your gender? X Male ______ Female How old are you? 19 year s Year in school: Junior Major: Psychology What are your plans for the future? (Please answer briefly.) I plan to go to graduate school. What is one event in your life that helped define who you are today ? (Please answer briefly.) When I found out I got HIV from a blood transfusion.

PAGE 85

85 APPENDIX F PARTICIPANT INFORMAT ION SHEET FOR CONTRO L TARGET Please answer the following questions. As mentioned before, your responses are confidential. These demographic questions only serve to help u s explore the sample helping us with our study. What is your gender? X Male ______ Female How old are you? 19 years Year in school: Junior Major: Psychology What are your plans for the future? (Please answer briefly.) I plan to go to graduate school. What is one event in your life that helped define who you are today ? (Please answer briefly.) When I found out I was accepted to UF.

PAGE 86

86 APPENDIX G FEELINGS TOWARD OTHE R PARTICIPANT Based on the information that he/she provided, how much do you look forward to working with the other participant? 1 2 3 4 5 6 7 Not at all Very much Based on the information that he/she provided, how much do you value the other 1 2 3 4 5 6 7 Not at all Very much Based on the information that he/she provided, how likely would you be to be friends with the other partici pant? 1 2 3 4 5 6 7 Not at all Very much Based on the information that he/she provided, how well do you think that you and the other participant will work together? 1 2 3 4 5 6 7 Not at all Very much

PAGE 87

87 APPENDIX H SOCIAL DISTANCING SC ALE Based on the information provided by the other participant, please indicate the degree to which you agree or disagree with the following statements by circling a number on each scale below: He/sh e appears to be a likeable person. Strongly disagree Strongly agree 1 2 3 4 5 6 7 I would like him/her to be a close personal friend. Strongly disagree Strongly agree 1 2 3 4 5 6 7 Strongly disagree Strongly agree 1 2 3 4 5 6 7 I would like him/her to come and work at the same place I do. Strongly disagree Strongly agree 1 2 3 4 5 6 7 He/she is a person who is similar to me. Strongly disagree Strongly agree 1 2 3 4 5 6 7 I would like to have him/her marry into my family. Strongly disagree Strongly agree 1 2 3 4 5 6 7 He/she is the kind of person that I tend to avoid. Strongly disagree Strongly agree 1 2 3 4 5 6 7

PAGE 88

88 APPENDIX I THREAT PERCEPTIONS T OWARD TARGET Based on the information provided by the other participant, please answer the following questions. In general, I think that people like him/her pose a challenge to people like me 1 2 3 4 5 6 7 Not at all Very much In general, I think that people like him/her increase the risk of physical sickness for people like me 1 2 3 4 5 6 7 Not at all Very much In general, I think that people like him/her pose problems for people like me 1 2 3 4 5 6 7 Not at all Very much In general, I think that people like him/her promote values that directly oppose the values of people like me 1 2 3 4 5 6 7 Not at all Very much In general, I think that people like him/her harm the medical health of pe ople like me 1 2 3 4 5 6 7 Not at all Very much In general, I think that people like him/her advocate values that are morally inferior to the values of people like me 1 2 3 4 5 6 7 Not at all Very much

PAGE 89

89 APPENDIX J EMOTIONAL REACTIONS TOWARD TARGET Based on the information provided by the other participant, please answer the following questions. When I think about my impressions of the other participant I feel respect for him/her 1 2 3 4 5 6 7 Not at all Very much When I think about my impressions of the other participant I feel positive toward him/her 1 2 3 4 5 6 7 Not at all Very much When I think about my impressions of the other participant I feel morally sickened by him/her 1 2 3 4 5 6 7 Not at all Very much When I think about my impressions of the other participant I feel dislike for him/her 1 2 3 4 5 6 7 Not at all Very much When I think about my impressions of the other participant I feel admiration for him/her 1 2 3 4 5 6 7 Not at all Very much When I think about my impressions of the other participant I feel physically disgusted by him/her 1 2 3 4 5 6 7 Not at all Very much When I think about my impressions of the other participant I feel negative towards him/her

PAGE 90

90 1 2 3 4 5 6 7 Not at all Very much When I think about my impressions of the other participant I feel morally disgusted by him/her 1 2 3 4 5 6 7 Not at all Very much When I think about my impressions of the other participant I feel a liking for him/her 1 2 3 4 5 6 7 Not at all Very much When I think about my impressions of the other participant I fee l physically sickened by him/her 1 2 3 4 5 6 7 Not at all Very much

PAGE 91

91 APPENDIX K THREAT RELEVANT BEHAVIORAL RATINGS TOWARD TARGE T Based on the information provided by the other participant, please answer the following questions. In general, to what extent would you recommend a member of this group assist you in raising money for an on campus charity? 1 2 3 4 5 6 7 Not at all Very much In general, to what extent would you recommend a member of this group influence your children as a volunteer at a local day care center ? 1 2 3 4 5 6 7 Not at all Very much In general, to what extent would you recommend a member of this group come into close contact with a sick family member as a voluntee r at a local hospital? 1 2 3 4 5 6 7 Not at all Very much In general, to what extent would you recommend a member of this group draw your blood as a volunteer at a local blood bank ? 1 2 3 4 5 6 7 Not at all Very much In general, to what extent would you recommend a member of this group administer justice and interpret laws by serving as a judge in your county ? 1 2 3 4 5 6 7 Not at all Very much In general, how likely would you be to eat a meal prepared by a member of this group ? 1 2 3 4 5 6 7 Not at all Very much

PAGE 92

92 In general, how likely would you be to vote for a member of this group who is running for political office in your state ? 1 2 3 4 5 6 7 Not at all Very much In general, to what extent would you recommend a member of this group teach morals and values to your children as a youth group leader at a local church ? 1 2 3 4 5 6 7 Not at all Very much In general, how likely would y ou be to read a political blog posted by a member of this group ? 1 2 3 4 5 6 7 Not at all Very much In general, how likely would you be to share finger foods (e.g., nachos, cheese/veggie dip, french fries) with a member of this group at a social gathering ? 1 2 3 4 5 6 7 Not at all Very much

PAGE 93

93 APPENDIX L ENDATIONS FOR UNIVER SITY INVOLVEMENT 1. They could survey students like me about our behaviors. 2. Could ask me abou t my opinion before making new policies. 3. Could have an email address or hotline where students like me could give suggestions for new policies.

PAGE 94

94 APPENDIX M THREE DOMAIN DISGUST SCALE Please indicate the extent to which you find each of the following disgu sting: 0 1 2 3 4 5 6 Not at all Extremely Dis gusting Disgusting _____ Shoplifting a candy bar from a convenience store. _____ Hearing two strangers having sex. _____ Stepping on dog poop. _____ Stealing from a neighbor. _____ Performing oral sex. _____ Sitting next to someone who has red sores on their arm. _____ A student cheating to get good grades. _____ Watching a pornographic video. _____ Deceiving a friend. _____ Finding out that someone you _____ Seeing some mold on old leftovers in your refrigerator. _____ Bringing someone you just met back to your room to have sex. _____ Standing close to a person who has body odor. _____ Cutting to the front of a line to purchase the last few tickets to a show. _____ A stranger of the opposite sex intentionally rubbing your thigh in an elevator. _____ Seeing a cockroach run across the floor. _____ Intentionally lying during a business transaction. _____ Having anal sex with someone of the opposite sex.

PAGE 95

95

PAGE 96

96 APPENDIX N DEMOGRAPHICS Please answer the following questions. As mentioned before, your respo nses are confidential. These demographic questions only serve to help us explore the sample helping us with our study. Which of the following ethnic group(s) do you consider yourself a member of? You can check multiple groups. ______ African American ______ Asian American ______ Hispanic ______ Native American ______ Caucasian/White ______ Other: _________________________________________ Which of the following religions best describes you? ______ Buddhist ______ Catholic ______ Fundamentalist/Evangelical Christian ______ Hindu ______ Jewish ______ Muslim ______ Protestant (Methodist, Lutheran, Episcopalian, etc.) ______ No religious affiliation ______ Other: _________________________________________ How important to y our identity are your religions beliefs?

PAGE 97

97 ______ Not at all important ______ Slightly important ______ Somewhat important ______ Very important ______ Extremely important How involved are you with religious organizations? ______ Not at all involved ______ Slightly involved ______ Somewhat involved ______ Very involved ______ Extremely involved How committed are you to your religious beliefs? ______ Not at all committed ______ Slightly committed ______ Somewhat committed ______ Very committe d ______ Extremely committed Using the scale below, please indicate the extent to which you consider yourself to be politically conservative or liberal. 1 2 3 4 5 6 7 Conservative Liberal How important to your identity are your political beliefs?

PAGE 98

98 ______ Not at all important ______ Slightly important ______ Somewhat important ______ Very important ______ Extremely important How involved are you with political organizations? ______ Not at all involved ______ Slightly invo lved ______ Somewhat involved ______ Very involved ______ Extremely involved How committed are you to your political beliefs? ______ Not at all committed ______ Slightly committed ______ Somewhat committed ______ Very committed ______ Extremely committed Which of the following best describes your sexual orientation?

PAGE 99

99 ______ Exclusively lesbian/gay ______ Mostly lesbian/gay ______ Bisexual ______ Mostly heterosexual ______ Exclusively heterosexual

PAGE 100

100 LIST OF REFERENCES Anderson, C. A., & Bushman, B. J. (2002). Media violence and societal violence. Science, 295, 2377 2378. Bargh, J. A., Chen, M., & Burrows, L. (1996). Automaticity of social behavior: Direct effects of trait constructs and stereotype activation on action. Journal of Personality and Social Psychology, 71, 230 244. Bogardus, E. S. (1925). Measuring social distances. Journal of Applied Sociology, 9, 299 308. Brown, D. E. (1991). Human Universals New York: McGraw Hill. Capitanio, J. P., & Herek, G. M. (19 99). AIDS related stigma and attitudes toward injecting drug users among black and white Americans. American Behavioral Scientist, 42, 1144 1157. Coltrane, S., & Messineo M. (2000). The perpetuation of subtle prejudice: Race and gender imagery in 1990s te levision advertising. Sex Roles, 42, 363 389. Correll, J., Park, B., Judd, C. M., Wittenbrink, B., Sadler, M.S., & Keesee, T. (2007). Across the thin blue line: Police officers and racial bias in the decision to shoot. Journal of Personality and Social Psychology, 92, 1006 1023. Cosmides, L., & Tooby, J. (2000). Evolutionary psychology and the emotions. In M. Lewis & J. M. Haviland Jones (Eds.), Handbook of emotions (pp. 91 115). New York: Guilford Press. Cottrell, C. A., & Neuberg, S. L. (2005). Different emotional reactions to different groups: A sociofunctional threat Journal of Personality and Social Psychology, 88, 770 789. Cottrell, C. A., Neuberg, S. L., & Li, N. P. (2 007). What do people desire in others? A sociofunctional perspective on the importance of different valued characteristics. Journal of Personality and Social Psychology, 92, 208 231. Crandall, C. S. (1991a). AIDS related stigma and the lay sense of justic e. Contemporary Social Psychology, 15, 66 67. Crandall, C. S. (1991b). Multiple stigma and AIDS: Medical stigma and attitudes toward homosexuals and IV drug users in AIDS related stigmatization. Social Behavior: Journal of Applied Social and Community Psy chology, 1 165 172. Crandall, C. S. (1994). Prejudice against fat people: Ideology and self interest. Journal of Personality and Social Psychology, 21, 724 735.

PAGE 101

101 Crandall, C. S., Glor, J., & Britt, T. W. (1997). AIDS related stigmatization: Instrumental a nd symbolic attitudes. Journal of Applied Social Psychology, 27, 95 123. Dawkins, R. (2006). The God delusion. New York, NY: Houghton Mifflin Company. Devine, P. G. (1989). Stereotypes and prejudice: Their automatic and controlled components. Journal of Personality and Social Psychology, 56, 5 18. Devine, P. G. (2001). Implicit prejudice and stereotyping: How automatic are they? Introduction to the special section. Journal of Personality and Social Psychology, 81, 757 759. Devine, P. G., & Elliot, A. J. (1995). Are racial stereotypes really fading? The Princeton trilogy revisited. Personality and Social Psychology Bulletin, 21, 1139 1150. Dovidio, J. F., & Ga e rtner, S. L. (Eds.). (1986). Prejudice, discrimination, and racism. San Diego, CA: Academic Press. Dovidio, J. F., & Gaertner, S. L. (1991). Changes in the expression and assessment of racial prejudice. In H. J. Knopke, R. J. Norrell, & R. W. Rogers (Eds.), Opening doors: Perspectives on race relations in contemporary America (pp. 119 148). Tusc aloosa: University of Alabama Press. Dovidio, J. F., & Gaertner, S. L. (2004). Aversive racism. Advances in Experimental Social Psychology, 36, 1 52. Duncan, L. A., Schaller, M., & Park, J. H. (2009). Perceived vulnerability to disease: Development and validation of a 15 item self report instrument. Personality and Individual Differences, 47, 541 546. Dunton, B. C., & Fazio, R. H. (1997). An individual difference measure of motivat ion to control prejudiced reactions. Personality and Social Psychology Bulletin, 23, 316 326. cultural membership in American society. American Sociological Review, 7 1, 211 234. Ekman, P. (1999). Basic emotions. In T. Dalgleish & M. Power (Eds.), The handbook of cognition and emotions (pp. 45 60. Sussex, England: Wiley. Faulkner, J., Schaller, M., Park, J. H., & Duncan, L. A. (2004). Evolved disease avoidance proces ses and contemporary xenophobic attitudes. Group Processes and Intergroup Behavior, 7, 333 353.

PAGE 102

102 Fein, S., & Spencer, S. J. (1997). Prejudice as self image maintenance: Affirming the self through derogating others. Journal of Personality and Social Psycholo gy, 73, 31 44. Fiske, S. T. (2002). What we know now about bias and intergroup conflict, the problem of the century. Current Directions in Psychological Science, 11, 123 128. Fiske, S. T., & Neuberg, S. L. (1990). A continuum model of impression format ion from category based to individuated processes: Influences of information and motivation on attention and interpretation. In M. P. Zanna (Ed.), Advances in Experimental Social Psychology (vol. 3, pp. 1 74). New York: Academic Press. Gervais, W. M., Sha riff, A. F., & Norenzayan, A. (2011). Do you believe in atheists? Distrust is central to anti atheist prejudice. Journal of Personality and Social Psychology, 101, 1189 1206. Gilbert, D. T., & Hixon, J. G. (1991). The trouble of thinking: Activation and application of stereotypic beliefs. Journal of Personality and Social Psychology, 60, 509 517. Greenberg, J., Solomon, S., & Pyszczynski, T. (1997). Terror management theory of self esteem and cultural worldviews: Empirical assessments and conceptual refi nements. Advances in Experimental Social Psychology, 29, 61 139. Greenwald, A. G., Poelman, T. A., Uhlmann, E., & Banaji, M. R. (2009). Understanding and using the Implicit Association Test: III. Meta analysis of predictive validity. Journal of Personalit y and Social Psychology, 97, 17 41. Haidt, J., & Joseph, C. (2007). The moral mind: How 5 sets of innate moral intuitions guide the development of many culture specific virtues, and perhaps even modules. In P. Carrthers, S. Laurence, and S. Stich (Eds.), The innate mind Volume 3: Foundations and the future. (pp. 367 391). New York, NY: Oxford University Press. Haselton, M. G., Nettle, D. and Andews, P. W. (2004): The evolution of cognitive bias. In D. M. Buss (ed.): The evolutionary psychology handbook. New York: Wiley. stigmas. Personality and Social Psychology Review, 9, 156 182. Herek, G. M., Capitanio, J. P., & Widaman, K. F. (2002). HIV related stigma and kno wledge in the United States: Prevalence and trends, 1991 1999. American Journal of Public Health, 92, 371 377. Hilton, J. L., & von Hippel, W. (1996). Stereotypes. Annual Review of Psychology, 47, 237 271.

PAGE 103

103 Jester, E. (2011, October 28). UF, FAMU to expan d dental program. The Independent Florida Alligator. Retrieved from www.alligator.org Kenrick, D. T., Griskevicius, V., Neuberg, S. L., & Schaller, M. (2010). Renovating the pyramid of needs: Contemporary extensions built upon ancient foundations. Perspectives on Psychological Sceince, 5, 292 314. Mackie, D. M., & Smith, E. R. (Eds.) (2002). From prejudice to intergroup relations: Differentiated reactions to social groups. New York: Psychology Press. Macrae C. N., & Bodenhausen, G. V. (2000). Social cognition: Thinking categorically about others. Annual Review of Psychology, 51, 93 120. Madon, S., Guyll, M., Aboufadel, K., Montiel, E., Smith, A., Palumbo, P., & Jussim, L. (2001). Ethnic and national stereo types: The Princeton trilogy revisited. Personality and Social Psychology Bulletin, 27, 996 1010. Neuberg, S. L, Kenrick, D. T., & Schaller, M. (2011). Human threat management systems: Self protection and disease avoidance. Neuroscience and Biobehavioral Reviews, 35, 1042 1051. Park, J. H., Faulkner, J., & Schaller, M. (2003). Evolved disease avoidance processes and contemporary anti social behavior: Prejudicial attitudes and avoidance of people with disabilities. Journal of Nonverbal Behavior, 27, 65 87 Park, J. H., Schaller, M., & Crandall, C. S. (2007). Pathogen avoidance mechanisms and the stigmatization of obese people. Evolution and Human Behavior, 28, 410 414. Pew Research Center (2002). abroad Washington, DC: Author. Plant, E. A., & Devine, P. G. (1998). Internal and external motivation to respond without prejudice. Journal of Personality and Social Psychology, 80, 876 893. Plutchik, R. (2003). Emotions and life: Perspectives from psychology, biology, and evolution. Washington, DC: American Psychological Association. Pyszczynski, T., Greenberg, J., Solomon, S.. & Schimel, J. (2004). Why do people need self esteem? A theoretical and empirical rev iew. Psychological Bulletin, 130, 435 468. Ritter, R. S., & Preston, J. L. (2011). Gross gods and icky atheism: Disgust responses to rejected religious beliefs. Journal of Experimental Social Psychology, 47, 1225 1230.

PAGE 104

104 Rozin, P., Haidt, J., & McCauley, C. (2009). Disgust: The body and soul emotion in the 21 st century. In Olatunji, B. O., & McKay, D. (Eds.), Disgust and its disorders: Theory, assessment, and treatment implications (pp. 9 29). Washington, DC: American Psychological Association. Schaller, M. Park, J., & Faulkner, M. (2003). Prehistoric dangers and contemporary prejudices. European Review of Social Psychology, 14, 105 137. Schaller, M., Faulkner, J., Park, J., Neuberg, S. L., & Kenrick, D. T. (2004). Impressions of danger influence impressi ons of people: An evolutionary perspective on individual and collective cognition. Journal of Cultural and Evolutionary Psychology, 2, 231 247. Schaller, M., Park, J. H., & Mueller, A. (2003). Fear of the dark: Interactive effects of beliefs about danger a nd ambient darkness on ethnic stereotypes. Personality and Social Psychology Bulletin, 29, 637 649. Sherif, M., Harvey, O. J., White, B. J., Hood, W. R., & Sherif, C. W. (1961/1988). The Robbers Cave experiment: Intergroup conflict and cooperation. Middle town, CT: Wesleyan University Press. Tybur, J. M., Lieberman, D., & Griskevicius, V. (2009). Microbes, mating, and morality: Individual differences in three functional domains of disgust. Journal of Personality and Social Psychology, 97, 103 122.

PAGE 105

105 BIOGRAPHICAL SKETCH Corey Cook was born in Glendale, AZ in 1983, and was raised in Show Low, AZ. He graduated m agna cum l aude from Arizona S tate University in 2005 with a Bachelor of S cience degree in p sychology, becoming the first in his family to earn a college degree. After graduating, Corey was employed as an academic counselor and continued to work as a volunteer research assistant in the psychology department at Arizona State. He was accepted to th e social psychology program at University of Florida i n fall 2007, where he earned a Master of Science degree in p syc hology in 2009, and a Ph.D. in social p sychology in 2012. In 2011 Corey was awarded the Gerber Social Psychology O utstanding R esearch A ward from the Department of Psychology at University of Florida, and in 2012 he was awarded the Pearson Teaching Award for Excellence in Undergraduate Education