Printer Friendly

The effect of counterexemplars and victim expectations on crime perceptions and hostile attitudes toward racial minorities.

Heavy consumers of crime news are more likely to hold negative attitudes and stereotyped expectations of racial minorities. Attempts to reduce stereotyping with counterexemplars that portray racial minorities in a positive light have yielded mixed results. Do prior crime expectations or frequency of crime news exposure moderate the efficacy of counter-exemplar interventions? An experiment (N = 240) was conducted to test this question using a 2 (exemplar typicality: atypical victim vs. no victim control) x 2 (victim expectations: White vs. Black) between-subjects design. Results revealed that counterexemplars that conflicted with viewers' preexisting victim expectations decreased hostile attitudes among light consumers of crime news, but increased racial stereotyping among heavy news consumers. The theoretical and practical implications of these findings for media psychology are discussed.

Keywords: racial stereotyping, media effects, intervention, counter-exemplar, construct accessibility

Ethnic minorities are often underrepresented across popular media (Dixon, Azocar, & Casas, 2003; Dixon & Linz, 2000; Tukachinsky, Mastro, & Yarchi, 2015). When African Americans or Latino characters do appear on television or in newspapers, content analyses show that they tend to be portrayed in an undesirable manner (Dixon et al., 2003; Dixon & Linz, 2000). One particularly problematic example of these trends in representation is the domain of news, where past work has found that members of minority groups are more likely than Whites to be portrayed as perpetrators of criminal acts (Entman, 1994; Oliver, 1994; Romer, Jamieson, & De Coteau, 1998). These unrepresentative, stereotyped portrayals have a range of negative consequences for the ways that racial minorities are perceived, particularly among heavy news consumers (Dixon & Maddox, 2005; Hurley, Jensen, Weaver, & Dixon, 2015). For example, several studies have found that frequent readers of crime news are more likely to assume that an unidentified criminal is Black rather than White (Dixon, 2007; Oliver & Fonash, 2002), which scholars have attributed to the increased accessibility of race-related exemplars that are made chronically available by exposure to crime news over time (Dixon & Azocar, 2007).

What steps can be taken to reverse the negative effects of heavy crime news consumption on race-related judgments and attitudes? One possibility is media-based interventions that expose audiences to counterexemplars of minority groups, such as portrayals that depict non-White groups in positive or sympathetic roles not typically shown in news reports (Mastro, 2015; Scharrer & Ramasubramanian, 2015). Past research on the efficacy of counter-based exemplars for decreasing racial stereotyping has yielded mixed findings. Some studies have found evidence that exposure to counterexemplars that positively depict African Americans can reverse preexisting negative attitudes toward minorities (Dasgupta & Greenwald, 2001; Paluck, 2009; Power, Murphy, & Coover, 1996). With that said, boomerang effects have also been found, as intervention attempts aimed at decreasing the accessibility of stereotypes can actually lead to more negative attitudes toward minorities and increased racial stereotyping (Bodenhausen, Schwarz, Bless, & Wanke, 1995; Devine & Monteith, 1999; Ramasubramanian & Oliver, 2007).

Building on this intervention paradigm, the present work examines whether the effect of counterexemplars on racial stereotyping and attitudes toward Blacks is moderated by viewers' prior victim expectations and crime news consumption. Furthermore, the present work also explores possible mediating variables responsible for the effects elicited by counter-exemplar exposure, focusing specifically on the role played by universal orientation and perceived realism.

Media Effects and Racial Stereotyping

A large body of research from the media effects tradition has examined the effects of exposure to crime news on racial stereotyping. Exposure to newscasts that feature a majority of Black suspects is associated with increased perceptions of suspect culpability (Dixon, 2006), heightened agreement that the world is "a mean and dangerous place" (Dixon & Maddox, 2005), and more negative attitudes toward racial minorities (Tukachinsky et al., 2015), among an array of other negative outcomes (Mastro, 2009). One specific strand of this program of research has focused on how heavy exposure to crime news influences individuals' expectations of perpetrator and victim race. Findings from this work have revealed that heavy news viewers are more likely to assume that unidentified criminals are Black rather than White (Dixon, 2007; Dixon & Azocar, 2007). Along similar lines, exposure to crime-related news has also been associated with a greater likelihood of misidentifying perpetrators as Black (Oliver & Fonash, 2002) and increased support for punitive treatment of criminals (Dixon, 2008).

The media's role in shaping news readers' perceptions of social reality is typically theorized to be caused by differences in construct accessibility (Sanbonmatsu & Fazio, 1991; Scheufele & Tewksbury, 2007; Shrum, 2002). From an accessibility approach, trends in the way that minorities are portrayed in news influence the race-related exemplars that are chronically accessible for viewers. The degree to which exemplars are cognitively accessible has consequences for race-related judgments, as exemplars that are easily retrieved from memory are more likely to subsequently influence media effects (Shrum, 2002; Zillmann, 2002). If stereotyped exemplars are contributing to the negative effects of news consumption on perceptions of racial groups, is it possible that counterexemplars of racial minorities as victims (rather than criminals) can be employed to reverse this pattern?

Multiple studies have tested the utility of media-based interventions using counterexemplars from popular media (Bodenhausen et al., 1995; Mastro & Tukachinsky, 2011; Ramasubramanian, 2011, 2015). In some cases, interventions that employed counterexemplars have had success in reducing racial stereotyping. For example, Dasgupta and Greenwald (2001) found that participants exposed to an "admired Black" exemplar were less likely to exhibit positive implicit attitudes toward Whites. Other scholars (Ramasubramanian, 2007) have compared the effects of counterexemplars with the success of media literacy interventions, with results revealing that the endorsement of implicit stereotypes is significantly reduced by a combination of message and media literacy approaches. The effect of counterexemplars embedded in news stories has also been tested, with findings revealing that counterexemplars decrease prejudice relative to stereotypical exemplars (Ramasubramanian & Oliver, 2007). However, not all intervention attempts and methods have been successful--in some cases either failing to affect explicit attitudes or even increasing negative attitudes toward minorities (e.g., Scharrer & Ramasubramanian, 2015).

One possible reason for the equivocality of past research on media-based interventions is that the counter-exemplar in question must be capable of overriding heavy news consumers' default tendency to generalize from chronically accessible exemplars derived from frequent news consumption. To that end, the present study considers whether the effect of counterexemplars on racial stereotyping is moderated by participants' preceding exemplar expectations (e.g., expectations of whether a hypothetical victim is either White or Black). Exemplification theory (Zillmann, 2002; Zillmann & Brosius, 2012) posits that individuals tend to generalize from examples to the population overall because exemplars tend to be vivid and easy to process, which heightens their memorability and subsequent retrieval at the time that judgments are rendered (Zillmann, 1999). Extending this logic to the current study, counterexemplars may be more memorable among participants whose expectations are disconfirmed by the atypicality of counterexemplars, thus leading to less frequent racial stereotyping on subsequent tasks due to the heightened accessibility of the counter-exemplar. In other words, the logic of exemplification might suggest that heavy consumers of crime news will be less likely to expect that Blacks are victims of crime when the judgment is preceded by a media-based counter-exemplar that disconfirms their prior expectations (e.g., a viewer who expects a victim of a violent crime to be White, but instead is exposed to a media exemplar of a violent crime against a victim who is Black). More formally, the following hypothesis is proposed:
H1:  The effect of counterexemplars on stereotyping and attitudes
     toward African Americans will be moderated by the frequency of
     crime news exposure and previous exemplar expectations.

Possible Mediators of Intervention Effects: Universal Orientation

Prejudice is often explained as a function of humans' automatic tendency to employ categorizations based on perceived differences between the self and others (e.g., Allport, 1954; Brewer, 1979; Tversky, 1977). Although work on measuring out-group categorization has yielded insight into the factors that contribute to the development of prejudicial attitudes, scholars have argued that assuming prejudice is a constant excludes the possibility of studying perceptions of similarity with others (Phillips & Ziller, 1997). Specifically, Phillips and Ziller propose the concept of "universal orientation," or the perceptual tendency to focus on shared similarities between social actors. Put simply, universal orientation is the absence of prejudice. Participants scoring highly on universal orientation are more likely to perceive others in terms of common experiences universal to the human experience. This has significant consequences for racial stereotyping, because past work has found that higher levels of universal orientation are related to lower levels of racial bias.

For scholars interested in reversing racial stereotyping, universal orientation may be one mechanism through which the possible effects of a counter-exemplar intervention might occur. As a result, interventions using media to reduce racial stereotyping may do so in part through their potential influence on universal orientation. For example, one study (Oliver et al., 2015) found that inspiring media portrayals can increase positive attitudes toward out-group members through the mediating pathway of universal orientation, providing tentative evidence that universal orientation is not just a trait but also a state capable of being influenced by preceding media experiences. More broadly, a long tradition of research from the field of media effects has studied viewers' tendency to vicariously experience narrative worlds. For example, a large body of work on "identification" (e.g., Cohen, 2001; de Graaf, Hoeken, Sanders, & Beentjes, 2011; Hefner, Klimmt, & Vorderer, 2007; Tal-Or & Cohen, 2010) has revealed that viewers can experience media from the perspective of fictional characters. In other words, identification refers to the process by which individuals temporarily become disassociated from the present self to have a vicarious experience through characters shown in media (e.g., Igartua, 2010; Moyer-Guse, Chung, & Jain, 2011; van Looy, Courtois, de Vocht, & de Marez, 2012).

In the context of counterexemplars, viewers who vicariously identify with minorities may be more likely to perceive similarities between the self and others. In other words, counter-exemplar interventions that portray racial minorities as the victim (rather than the perpetrator) may affect universal orientation by calling attention to the consequences of victimhood experienced by humans regardless of demographic affiliation. In turn, heighted levels of universal orientation are likely to elicit positive out-group perceptions, given the findings observed in past work. More formally, the following hypothesis is proposed:
H2:  Counter-exemplar exposure (relative to a no-exemplar control)
     will decrease stereotyping and hostile attitudes toward
     minorities through the indirect pathway of universal orientation.

Possible Mediators of Intervention Effects: Perceived Realism

An examination of mediating variables also affords the possibility to examine psychological mechanisms potentially responsible for the activation of boomerang effects, or interventions that heighten rather than reduce racial prejudices. One such variable that could decrease the effectiveness of a counter-exemplar-based intervention is the perceived realism (Busselle, 2001; Green, 2004; Hall, 2003; Pouliot & Cowen, 2007; Potter, 1988) of the out-group portrayal. Perceived realism can refer to one of two possible types of judgments: (1) the degree to which media are perceived to be consistent with everyday life or (2) the degree to which the properties of the narrative in question are internally consistent (Bilandzic & Busselle, 2011). In terms of the former, scholars have categorized the degree to which media are perceived as reflective of everyday reality as "external realism" (Busselle & Bilandzic, 2008; Busselle & Greenberg, 2000). Media are presumed to be high in external realism when there is high similarity between the fictional world presented in media and viewers' mental models concerning the nature of the corresponding event in everyday life. Put another way, external realism depends on assessments of whether the media narrative is consistent with one's perception of the accompanying real-world situation that is being depicted. Considerations of media realism have important consequences for the effects that media elicit. For example, past studies on the effects of realism have found that narratives scoring high on external realism are more likely to elicit subsequent psychological effects than comparable narratives scoring low on external realism (Busselle & Shrum, 2003; Cho, Shen, & Wilson, 2014).

Often individuals have very little direct experience with violent crime in everyday life. As a result, viewers' perceptions of crime tend to be derived from the way that violent crime is portrayed on television. As cultivation theory describes (Gerbner, Gross, Morgan, Signorielli, & Shanahan, 2002), individuals' perceptions of crime are likely to be consistent with the demographic portrayals of criminals in media. Content analyses of television programming and news reports have shown that minority groups tend to be underrepresented as victims of violent crime and overrepresented as perpetrators of crime. As a result, heavy viewers of crime news are likely to hold expectations of crime in real life that mirror the unrealistic and disproportionate patterns that appear in media. Consistent with this prediction, past studies have found that viewers tend to perceive portrayals of African Americans on television as realistic of how African Americans are also perceived in nonmediated contexts (Fujioka, 1999; Gunter, 1998; Punyanunt-Carter, 2008).

If this is the case, then counterexemplars that depict minorities as victims of crime might be inconsistent with viewers' mental models regarding crime and demographics in everyday life. In other words, counter-exemplar portrayals might be perceived as lower in external realism, which past research has shown can decrease media effects. If this is the case, then perceived realism may be a possible mediating variable through which counterexemplars could elicit negative effects on viewers' subsequent racial attitudes. Given this logic, the following hypothesis is proposed:
H3:  Counter-exemplar exposure (relative to a no-exemplar control)
     will increase stereotyping and hostile attitudes toward
     minorities through the indirect pathway of perceived realism
     among heavy consumers of crime news.


An online experiment (N = 240) was conducted to test a 2 (exemplar typicality: atypical exemplar vs. no exemplar control) x 2 (victim expectations: White vs. Black) between-subjects design. Participants were asked to read a film script, then select a female actor (from a collection of four photographs) who would be appropriate for the role of victim described in the script. Afterward, participants watched a film clip showing either the counter-exemplar video clip (modeled after the scene that was just described in the script) or an unrelated scene taken from the same film. Finally, participants (in a supposedly unrelated task) were asked to read a news article that described a violent crime. Following the experimental tasks, participants completed a questionnaire that measured the anticipated race of the victims described in the news article, their feelings of universal orientation, and their attitudes toward African Americans.


Two hundred and forty people were recruited to participate in the study using the online crowdsourcing website Amazon Mechanical Turk (Buhrmester, Kwang, & Gosling, 2011). Participants were compensated 75 cents for completing the study, which was approved by the institutional review board at the primary investigator's university. The sample was evenly divided in terms of participant sex (50% men, n = 119) with a mean age of 38.95 (SD = 12.50). When asked to self-report their race, 77% reported White/Caucasian (n = 185), 10% reported Black/African American (n = 25), 5% reported Hispanic/Latino/Latina (n = 12), 5% reported Asian/Asian American (n = 11), and 3% reported biracial (n = 7). All participants were given a written description of the research and consented to participate before completing the study.


Film Casting Activity

Measures of racial bias are often limited by social desirability bias among respondents. Thus, it is common for studies to measure racial stereotyping using experimental procedures that measure racial attitudes with implicit techniques so that participants are less able to regulate their responses to provide normative answers. Following this logic, the present study measured participants' expectations of victim race by using a hypothetical film casting activity. The activity asked participants to read a hypothetical film script while imagining the type of actor they would cast for each of the roles described in the scene. The 240-word script was modeled after the opening scene portrayed in the 1996 film A Time to Kill, which features a violent crime committed by a pair of drunk men against a young girl. The scene was selected for its low familiarity among participants given that 20 years have passed since the film's original release. After reading the 240-word script, participants were directed to a second screen that asked them to select one of four possible actresses to play the role of victim in the script. The race of the actresses (as shown in the photographs presented to participants) was varied to include two actresses who were White and two actresses who were Black.

Video Clip Counter-Exemplar

The second portion of the study asked participants to watch a three-minute video clip taken from A Time to Kill, the same film described in the preceding script activity. Participants watched one of two possible scenes from the film depending on their experimental condition. In the atypical exemplar condition, participants watched the opening exposition of the film, which featured a series of aggressive actions by a pair of White men against a series of Black victims. The scene included images of the men throwing a bottle at African American youth, shoplifting from a grocery store managed by an African American owner, then attacking a young African American girl who is walking down the street alone. In the no exemplar control condition, participants watched a scene from the film where a lawyer and his wife (both of whom are White) discuss his upcoming case, which features vague references to the aforementioned crimes, but no explicit indication of the victims or perpetrators' demographic traits.

To ensure the counter-exemplar portrayal was perceived as intended among participants, an independent sample of participants (N = 49) was recruited from Amazon Mechanical Turk to watch the same scene used to operationalize the counter-exemplar condition. Most participants from the independent sample correctly recalled the race of the victim (98%) and criminal (100%) from the atypical exemplar clip. Furthermore, participants generally disagreed that the victim (M = 3.78, SD = 1.53) and crime (M =3.18, SD = 1.62) depicted in the atypical exemplar condition were "typical" examples of violence in everyday life. In sum, the overwhelming majority of participants correctly recalled the race of the victim and criminals depicted in the scene. Furthermore, participants' perceived the victim and crime as relatively atypical, an outcome indicative of the manipulation's intended reception by participants.

News Story Attitude Object

A common dependent variable measured in past media effects research is the perceived race of unidentified victims and perpetrators of crime in written news stories (e.g., Dixon, 2007). To that end, the final section of the study asked participants to read a 150-word news article entitled, "Mall Shooting Leaves 1 Dead, 9 Hurt: Police Hunt for Suspects." The news stimulus was created based on a published news article that described the shooting of 10 people at a Pennsylvania shopping mall. The article described the location and time of the shooting as well as the status of the incident, which identified the case as ongoing with suspects yet to be apprehended. Importantly, the assumed identity of the perpetrator and victims was omitted from the story during creation of the stimuli. The article was attributed to USA Today, a relatively nonpartisan news outlet, to hold source credibility and partisanship constant across conditions. The crime news story served as the attitude object for items pertaining to dependent variables regarding the assumed race of the victims described in the article, which were included in a questionnaire that was given to participants' after exposure to the news article.

Experimental Conditions

Exemplar Typicality

As described, participants were randomly assigned to watch one of two possible video clips: either (1) a video clip that featured a violent act by a White perpetrator against a Black victim or (2) a control video clip that featured a conversation between a lawyer and his wife where the same violent act was discussed, but no details of the crime (e.g., the race of the victim or criminal) were revealed. Most participants reported that they had not viewed the film prior to the study (82%, n = 196). Given that the race of characters varied between the control and treatment conditions, an independent sample (N = 80) of participants from Amazon Mechanical Turk was run to compare responses to the control condition with a baseline who did not view any stimuli. No difference between the control and no video baseline were observed for any of the dependent variables of interest, all ts < 1.99, all ps > .50, thus suggesting that the portrayal depicted in the control group elicited no effect on respondents' victim expectations relative to a no-exposure baseline.

Victim Expectations

As described, participants completed a film casting activity where they were asked to read a short scene, then select an actor from a collection of four photos whom they perceived to be most appropriate for the role of victim in the scene. A subsequent variable was created where participants were assigned based on their actor selection to either the Black actress condition (46%, n = 110) or the White actress condition (54%, n = 54). In other words, "victim expectations" was a measured variable created based on participants' selection of either a White or Black actor during the film casting activity.

Mediating Variables

Universal Orientation

Six Likert-type items ranging from 1 (strongly disagree) to 7 (strongly agree) adapted from previous research (Oliver et al., 2015; Piedmont, 1999) asked participants to identify the extent to which they felt connected to others. Sample items included agreement with statements such as, "On a higher level, all of us share a common bond," "All life is interconnected," and "At one level of thinking, all humans are the same." An index was formed by combining the six items, which was reliable (Cronbach's [alpha] = .93; M = 5.17, SD = 1.41).

Perceived Realism

Five Likert-type items ranging from 1 (strongly disagree) to 7 (strongly agree) adapted from previous research (Busselle, 2001) asked participants to identify the perceived realism of the video clip they watched. Sample items included the extent to which participants agreed that, "The video clip showed life as it really is," "The video clip I watched let me see how other people live," and "The criminals in the video clip are just like criminals in the real world." An index was formed by combining the five items, which was reliable (Cronbach's [alpha] = .84; M = 4.49, SD = 1.26).

Dependent Variables

Victim Likelihood

Two Likert-type items ranging from 1 (not at all likely) to 7 (very likely) adapted from previous research (Dixon, 2007) asked participants to identify the race of the victims from the news story they read. Specifically, participants were asked, "How likely is it that the shooting victims from the news story were White?" (M = 5.20, SD = 1.23), and "How likely is it that the shooting victims from the news story were Black?" (M = 4.61, SD = 1.48).

Hostile Feelings Toward African Americans

Five Likert-type items ranging from 1 (not at all) to 7 (very much) adapted from previous research (Ramasubramanian & Oliver, 2007) asked participants to identify the extent to which the following adjectives described their emotional reaction toward African Americans: fear, nervousness, discomfort, dislike, and anger. An index was formed by combining the five items, which was reliable (Cronbach's [alpha] = .95; M = 2.04, SD = 1.35).

Other Measured Variables

Crime News Attention

A single Likert-type item ranging from 1 (no attention at all) to 5 (a lot of attention) adapted from previous research (Dixon, 2008) asked participants, "How much attention do you pay to news about crime?" (M = 3.58, SD = 1.08).


Participants were asked to identify their age, sex, and race for the purpose of compiling descriptive statistics about the study sample.


Victim Likelihood

A three-way analysis of variance (ANOVA) with exemplar typicality, victim expectations, and crime news attention as the independent variables and perceived likelihood that the victim was Black as the dependent variable revealed a significant three-way interaction, F(1, 232) = 4.63, p = .03, [[eta].sup.2.sub.p] = .02. To probe the interaction, an ordinary least squares regression was employed using Model 3 of the SPSS PROCESS macro (Hayes, 2013) with 5,000 bootstrapped samples and 95% bias-adjusted confidence intervals. As shown in Figure 1 and Table 1, the interaction effect between exemplar typicality and victim expectations was significant only for participants who scored high in crime news attention. Specifically, among those scoring high in crime news attention, atypical exemplar exposure decreased Black victim likelihood estimates relative to the control condition. A similar ANOVA pertaining to whether the victim was expected to be White revealed no main or interaction effects, all Fs < 3.56, all ps > .07.

To probe whether universal orientation or perceived realism was responsible for the three-way interaction, moderated mediation analyses were conducted using Model 13 of the PROCESS macro (Hayes, 2013) with 5,000 bootstrapped samples and 95% bias-adjusted confidence intervals. As shown in Figure 2, neither universal orientation (.07; lower-limit confidence interval [LLCI] = -.17, upper-limit confidence interval [ULCI] = .23) nor perceived realism (.001; LLCI = -.17, ULCI = .23) mediated the interaction effect, because the 95% bias adjusted confidence interval for both indices of moderated mediation included zero.


Hostile Feelings Toward African Americans

A similar ANOVA with hostile feelings toward African Americans as the dependent variable revealed a significant two-way interaction between exemplar exposure and victim expectations, F(1, 232) = 4.44, p = .04, [[eta].sup.2.sub.p] = .02. As shown in Figure 3, the effect of exemplar typicality on hostile feelings was moderated by victim expectations. Among participants who selected a White actor in the casting activity, atypical exemplars (relative to a no-exemplar control condition) decreased hostile attitudes toward African Americans. By comparison, there was relatively little difference in hostile attitudes between the atypical exemplar and control conditions among participants who selected a Black actor in the film casting activity (see Table 2 for estimates). No other main or interaction effects were significant, all Fs < 1.15, all ps > .27.

A similar series of moderated mediation models were run to test whether universal orientation or perceived realism mediated the effect of counter-exemplar exposure on hostile feelings toward African Americans. As shown in Figure 4, the conditional indirect effect of atypical exemplar exposure on hostile racial attitudes was mediated by universal orientation (-.16, LLCI = -.43, ULCI = -.02), as the indirect effect was only significant among light crime news viewers who selected a White actor in the casting activity. No moderated mediation effects via perceived realism were observed.



Summary of Results

The results of the present study found that the influence of atypical exemplars on anticipated victim race was moderated by preexisting victim expectancies and prior attention to crime news. Specifically, atypical exemplars decreased the perceived likelihood that crime victims were Black among those scoring high in attention to crime who expected a White victim in the video clip. Effects were also observed for hostile feelings toward African Americans, with exposure to the atypical exemplar decreasing hostile attitudes among those who previously expected a White victim. Subsequent moderated mediation analyses revealed that the influence of counterexemplars on racial attitudes among light crime viewers was mediated by increased levels of universal orientation among those who expected a White victim, whereas no moderated mediation outcomes via perceived realism were observed.


Frequent exposure to crime news can increase the likelihood of perceiving minorities as criminals (Mastro, 2009). Given the range of negative outcomes associated with distorted racial perceptions (Dixon, 2007; Oliver & Fonash, 2002; Ramasubramanian & Oliver, 2007; Tukachinsky et al., 2015), there has been an increased call for media-based interventions that may reverse the negative effects of heavy news consumption on race-based judgments such as perceptions of victim race (Mastro, 2015; Scharrer & Ramasubramanian, 2015). One popular intervention method that has been studied in past research is the influence of exposure to counterexemplars of minorities that portray ethnic groups with nonstereotyped, positive portrayals. The effects of these interventions have been mixed, however, with the influence of counterexemplars yielding both decreased and heightened racial stereotyping following media exposure (Bodenhausen et al., 1995; Devine & Monteith, 1999; Ramasubramanian & Oliver, 2007).

Past studies on racial stereotyping have found that responses to media depictions of race are moderated by individual differences such as preexisting racial attitudes (Mastro & Tukachinsky, 2011; Ramasubramanian, 2007) and frequency of news consumption (Dixon, 2008; Dixon & Azocar, 2007). Building on this earlier work, the present experiment tested whether the efficacy of counterexemplars is moderated by viewers' prior expectations of victim race. Results revealed that the effects of the counter-exemplar intervention on participants' victim expectations and racial attitudes toward African Americans were moderated by two key variables: (1) their preceding expectation of victim race and (2) the frequency of their attention to crime news. For individuals who expected victims of violence to be White, exposure to a counter-exemplar that conflicted with their prior victim expectations decreased hostile attitudes toward African Americans. By comparison, boomerang effects were observed for heavy consumers of crime news, as exposure to the counter-exemplar that conflicted with their prior victim expectations decreased perceptions of African Americans as victims of violence.

This pattern of results holds various theoretical implications for the application of exemplification theory (e.g., Zillmann, 2002) to the domain of racial stereotyping. Specifically, this study provides a unique test of how chronically accessible exemplars derived from crime news consumption interact with short-term counterexemplars primed by media-based interventions. Given that boomerang effects were observed among heavy consumers of crime news, it appears that a boundary condition for exemplification theory may be the presence of chronically accessible exemplars primed repeatedly by media over time. In other words, reactance to counterexemplars appears to be a function of exemplars derived from heavy news use competing with counterexemplars primed by the intervention. More broadly, it is important to note that the effects observed are likely short-term and small in magnitude. Attempts to use counterexemplars to elicit long-term attitude change will likely require cumulative, repeated exposure to media messages containing competing portrayals of ethnic minorities. Nonetheless, the results of our short-term effects reveal that such effects may be possible, at least in the short-term context of the experimental setting.

A second key theoretical contribution of the present study is provided by testing the possible theoretical mechanism through which the effects of counterexemplars occur. To that end, moderated mediation analyses were conducted to test whether differences in universal orientation or perceived realism mediated the effects of the counter-exemplar intervention on perceived victimhood and hostile racial attitudes. Findings show that universal orientation mediated the conditional effect of counterexemplars on hostile racial attitudes, such that participants exposed to a counter-exemplar who scored low in attention to crime news reported higher levels of universal orientation, which was subsequently related to decreased hostile attitudes toward African Americans. By comparison, no moderated mediation via perceived realism was observed, nor was the boomerang effect of counterexemplars among those scoring high in attention to crime news mediated by either perceived realism or universal orientation.

This finding contributes to our understanding of media-based interventions for decreasing racial attitudes by revealing that universal orientation is one mechanism responsible for the efficacy of counter-exemplar exposure. From a theoretical approach, this finding links the assumptions of exemplification theory with concepts from the domain of positive psychology such as universal orientation. Among light consumers of news, counterexemplars appear to shift viewers' perceived similarity to others, leading to downstream psychological outcomes such as increased levels of universal orientation. In other words, a previously unobserved consequence of exemplification effects appears to be increased identification and connection to others, a mechanism potentially responsible for the persuasive effects of counterexemplars observed in past work (Ramasubramanian, 2015), alongside other previously documented mediating variables such as internal and external attributions (Mastro, 2009; Ramasubramanian, 2011). It should also be noted that mediation analyses did not reveal indirect effects responsible for effects related to perceived victimhood, thus suggesting that alternative mechanisms independent of perceived realism and universal orientation are likely responsible for counter-exemplar effects on judgments of demographic traits and crime likelihood.

Practical Implications

For groups interested in the potential of media-based interventions for reducing racial stereotyping, our study offers both insights and cautions. In terms of insights, the findings appear to suggest that certain groups--those who are infrequent consumers of crime news--can be influenced by counterexemplars from popular media, particularly when the counter-exemplar conflicts with their prior victim expectations. By extension, groups that are infrequent consumers of news overall (e.g., young adults) may be an ideal demographic for counter-exemplar interventions due to their low level of attention to crime news overall. On the other hand, heavy crime news consumers appear to be the most difficult to affect through counterexemplars, particularly when the exemplar in question is atypical, thus suggesting that methods other than using counterexemplars (e.g., a media literacy approach, see Ramasubramanian, 2007) might be more appropriate for groups with particularly high exposure to stereotyped portrayals. More broadly, it should also be noted that the effect sizes observed in the present study were small, both for the intervention and related media exposure variables. Any attempt to reduce racial stereotyping should thus consider the relative magnitude of effects when considering whether other risk factors should be targeted or considered with potentially greater influence on racial stereotyping overall.

Limitations and Future Research

The present study should be contextualized according to several limitations. In terms of stimuli, one specific video clip adapted from a single film was used for the key experimental manipulations. Although this approach is common for media-based interventions, it would be ideal for future research to build on these results by testing the effects associated with multiple counterexemplars of the same type to ensure that any effects observed are generalizable to multiple types of counterexemplars. The effects observed in this study are also short-term, as is the case for most intervention-based studies in laboratory contexts. As a result, subsequent work could examine the cumulative effect of counter-exemplar exposure using a longitudinal design to complement the findings of the present work. In terms of measurement, it would be valuable for future studies to measure not only universal orientation but also the role possibly played by character identification as a sequential mediator of counter-exemplar effects. As for the sampling frame, this study focused specifically on responses to counterexemplars with a sample that was primarily White. Study findings may therefore not generalize to the responses of other racial groups or, more broadly, to groups not represented by samples recruited from online crowdsourcing websites. Finally, the limitations typically inherent to experimental research are also applicable, including the use of artificial procedures that may best be applied in intervention settings rather than observed in everyday life.


Despite the potential opportunities for refinement, the results of the present study nonetheless offer the unique theoretical contribution of revealing that counterexemplars can decrease some news readers' hostile attitudes toward African Americans due to increased feelings of universal orientation. Unintended effects are also possible, however, as frequent consumers of crime news appear to become more polarized in their racial beliefs following exposure to counterexemplars that conflict with their victim expectations. Media-based interventions that feature counterexemplars thus appear to be a double-edged sword, with those most heavily exposed to crime news the most difficult to persuade.


Allport, G. (1954). The nature of prejudice. Cambridge, MA: Addison-Wesley.

Bilandzic, H., & Busselle, R. W. (2011). Enjoyment of films as a function of narrative experience, perceived realism, and transportability. Communications, 36(1), 29-50. doi:10.1515/comm.2011.002

Bodenhausen, G. V., Schwarz, N., Bless, H., & Wanke, M. (1995). Effects of atypical exemplars on racial beliefs: Enlightened racism or generalized appraisals. Journal of Experimental Social Psychology, 31(1), 48-63. doi:10.1006/jesp.1995.1003

Brewer, M. B. (1979). In-group bias in the minimal group situation: A cognitive-motivational analysis. Psychological Bulletin, 86, 307-324. doi:10.1037/0033-2909.86.2.307

Buhrmester, M., Kwang, T., & Gosling, S. D. (2011). Amazon's Mechanical Turk: A new source of inexpensive, yet high-quality, data? Perspectives on Psychological Science, 6(1), 3-5. doi:10.1177/1745691610393980

Busselle, R. W. (2001). Television exposure, perceived realism, and exemplar accessibility in the social judgment process. Media Psychology, 3(1), 43-67. doi:10.1207/S1532785XMEP0301_03

Busselle, R. W., & Bilandzic, H. (2008). Fictionality and perceived realism in experiencing stories: A model of narrative comprehension and engagement. Communication Theory, 18(2), 255-280. doi:10.1111/j.1468-2885.2008.00322.x

Busselle, R. W., & Greenberg, B. S. (2000). The nature of television realism judgments: A reevaluation of their conceptualization and measurement. Mass Communication and Society, 3(2/3), 249-268. doi:10.1207/S15327825MCs0323_05

Busselle, R. W., & Shrum, L. J. (2003). Media exposure and exemplar accessibility. Media Psychology, 5(3), 255-282. doi:10.1207/S1532785XMEP0503_02

Cho, H., Shen, L., & Wilson, K. (2014). Perceived realism: Dimensions and roles in narrative persuasion. Communication Research, 41(6), 828-851. doi:10.1177/0093650212450585

Cohen, J. (2001). Defining identification: A theoretical look at the identification of audiences with media characters. Mass Communication and Society, 4(3), 245-264. doi:10.1207/S15327825MCS0403_01

Dasgupta, N., & Greenwald, A. G. (2001). On the malleability of automatic attitudes: Combating automatic prejudice with images of admired and disliked individuals. Journal of Personality and Social Psychology, 81(5), 800-814. doi:10.1037/0022-3514.81.5.800

de Graaf, A., Hoeken, H., Sanders, J., & Beentjes, J. W. J. (2011). Identification as a mechanism of narrative persuasion. Communication Research, 39(6), 802-823. doi:

Devine, P. G., & Monteith, M. J. (1999). Automaticity and control in stereotyping. In S. Chaiken & T. Yaacov (Eds.), Dual-process theories in social psychology (pp. 339-360). New York, NY: Guilford Press.

Dixon, T. L. (2006). Psychological reactions to crime news portrayals of Black criminals: Understanding the moderating roles of prior news viewing and stereotype endorsement. Communication Monographs, 73(2), 162-187. doi:10.1080/03637750600690643

Dixon, T. L. (2007). Black criminals and White officers: The effects of racially misrepresenting law breakers and law defenders on television news. Media Psychology, 10(2), 270-291. doi:10.1080/15213260701375660

Dixon, T. L. (2008). Crime news and racialized beliefs: Understanding the relationship between local news viewing and perceptions of African Americans and crime. Journal of Communication, 58(1), 106-125. doi:10.1111/j.1460-2466.2007.00376.x

Dixon, T. L., & Azocar, C. L. (2007). Priming crime and activating blackness: Understanding the psychological impact of the overrepresentation of Blacks as lawbreakers on television news. Journal of Communication, 57(2), 229-253. doi:10.1111/j.1460-2466.2007.00341.x

Dixon, T. L., Azocar, C. L., & Casas, M. (2003). The portrayal of race and crime on television network news. Journal of Broadcasting & Electronic Media, 47(4), 498-523. doi:10.1207/s15506878jobem4704_2

Dixon, T. L., & Linz, D. (2000). Overrepresentation and underrepresentation of African Americans and Latinos as lawbreakers on television news. Journal of Communication, 50(2), 131-154. doi:10.1111/j.1460-2466.2000.tb02845.x

Dixon, T. L., & Maddox, K. B. (2005). Skin tone, crime news, and social reality judgments: Priming the stereotype of the dark and dangerous black criminal. Journal of Applied Social Psychology, 35(8), 1555-1570. doi:10.1111/j.1559-1816.2005.tb02184.x

Entman, R. M. (1994). Representation and reality in the portrayal of Blacks on network television news. Journalism & Mass Communication Quarterly, 71 (3), 509-520. doi:10.1177/107769909407100303

Fujioka, Y. (1999). Television portrayals and African-American stereotypes: Examination of television effects when direct contact is lacking. Journalism & Mass Communication Quarterly, 76, 52-75. doi:10.1177/10776990 9907600105

Gerbner, G., Gross, L., Morgan, M., Signorielli, N., & Shanahan, J. (2002). Growing up with television: Cultivation processes. In B. Jennings & D. Zillmann (Eds.), Media effects: Advances in theory and research (pp. 43-67). Mahwah, NJ: Lawrence Erlbaum.

Green, M. C. (2004). Transportation into narrative worlds: The role of prior knowledge and perceived realism. Discourse Processes, 38(2), 247-266. doi:10.1177/107769909907600105

Gunter, B. (1998). Ethnicity and involvement in violence on television: Nature and context of on-screen portrayals. Journal of Black Studies, 28, 683-703.

Hall, A. (2003). Reading realism: Audiences' evaluations of the reality of media texts. Journal of Communication, 53(4), 624-641. doi:10.1111/j.1460-2466.2003.tb02914.x

Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. New York, NY: Guilford Press.

Hefner, D., Klimmt, C., & Vorderer, P. (2007). Identification with the player character as determinant of video game enjoyment. In L. Ma, M. Rauterberg, & R. Nakatsu (Eds.), Entertainment computing--ICEC 2007: Lecture notes in computer science (Vol. 4740, pp. 39-48). Berlin, Germany: Springer. doi:10.1007/978-3-540-74873-1_6

Hurley, R. J., Jensen, J. J., Weaver, A., & Dixon, T. (2015). Viewer ethnicity matters: Black crime in TV news and its impact on decisions regarding public policy. Journal of Social Issues, 71 (1), 155-170. doi:10.1111/josi.12102

Igartua, J. J. (2010). Identification with characters and narrative persuasion through fictional feature films. Communications, 35(4), 347-373. doi:10.1515/comm.2010.019

Mastro, D. (2009). Effects of racial and ethnic stereotyping. In J. Bryant and M. B. Oliver (Eds.), Media effects: Advances in theory and research (pp. 325-341). New York, NY: Routledge.

Mastro, D. (2015). Why the media's role in issues of race and ethnicity should be in the spotlight. Journal of Social Issues, 71(1), 1-16. doi:10.1111/josi.12093

Mastro, D., & Tukachinsky, R. (2011). The influence of exemplar versus prototype-based media primes on racial/ethnic evaluations. Journal of Communication, 61(5), 916-937. doi:10.1111/j.1460-2466.2011.01587.x

Moyer-Guse, E., Chung, A. H., & Jain, P. (2011). Identification with characters and discussion of taboo topics after exposure to an entertainment narrative about sexual health. Journal of Communication, 61(3), 387-406. doi:10.1111/j.1460-2466.2011.01551.x

Oliver, M. B. (1994). Portrayals of crime, race, and aggression in "reality-based" police shows: A content analysis. Journal of Broadcasting & Electronic Media, 38(2), 179-192. doi:10.1080/08838159409364255

Oliver, M. B., & Fonash, D. (2002). Race and crime in the news: Whites' identification and misidentification of violent and nonviolent criminal suspects. Media Psychology, 4(2), 137-156. doi:10.1207/S1532785XMEP0402_0

Oliver, M. B., Kim, K., Hoewe, J., Chung, M. Y., Ash, E., Woolley, J. K., & Shade, D. D. (2015). Media-induced elevation as a means of enhancing feelings of intergroup connectedness. Journal of Social Issues, 71(1), 106-122. doi:10.1111/josi.12099

Paluck, E. L. (2009). Reducing intergroup prejudice and conflict using the media: A field experiment in Rwanda. Journal of Personality and Social Psychology, 96(3), 574-587. doi:10.1037/a0011989

Phillips, S. T., & Ziller, R. C. (1997). Toward a theory and measure of the nature of nonprejudice. Journal of Personality and Social Psychology, 72(2), 420-434. doi:10.1037/0022-3514.72.2.420

Piedmont, R. L. (1999). Does spirituality represent the sixth factor of personality? Spiritual transcendence and the five-factor model. Journal of Personality, 67(6), 985-1013. doi:10.1111/1467-6494.00080

Potter, W. J. (1988). Perceived reality in television effects research. Journal of Broadcasting & Electronic Media, 32(1), 23-41. doi:10.1080/08838158809386682

Pouliot, L., & Cowen, P. S. (2007). Does perceived realism really matter in media effects? Media Psychology, 9(2), 241-259. doi:10.1080/15213260701285819

Power, J. G., Murphy, S. T., & Coover, G. (1996). Priming prejudice: How stereotypes and counterstereotypes influence attribution of responsibility and credibility among ingroups and outgroups. Human Communication Research, 23(1), 36-58. doi:10.1111/j.1468-2958.1996.tb00386.x

Punyanunt-Carter, N. (2008). The perceived realism of African American portrayals on television. Howard Journal of Communications, 19(3), 241-257. doi:10.1080/10646170802218263

Ramasubramanian, S. (2007). Media-based strategies to reduce racial stereotypes activated by news stories. Journalism & Mass Communication Quarterly, 84(2), 249-264. doi:10.1177/107769900708400204

Ramasubramanian, S. (2011). The impact of stereotypical versus counterstereotypical media exemplars on racial attitudes, causal attributions, and support for affirmative action. Communication Research, 38(4), 497-516. doi:10.1177/0093650210384854

Ramasubramanian, S. (2015). Using celebrity news stories to effectively reduce racial/ethnic prejudice. Journal of Social Issues, 71(1), 123-138. doi:10.1111/josi.12100

Ramasubramanian, S., & Oliver, M. B. (2007). Activating and suppressing hostile and benevolent racism: Evidence for comparative media stereotyping. Media Psychology, 9(3), 623-646. doi:10.1080/15213260701283244

Romer, D., Jamieson, K. H., & De Coteau, N. J. (1998). The treatment of persons of color in local television news: Ethnic blame discourse or realistic group conflict? Communication Research, 25(3), 286-305. doi:10.1177/009365098025003002

Sanbonmatsu, D. M., & Fazio, R. H. (1991). Construct accessibility: Determinants, consequences, and implications for the media. In J. Bryant & D. Zillmann (Eds.), Responding to the screen: Reception and reaction processes (pp. 45-62). Mahwah, NJ: Lawrence Erlbaum.

Scharrer, E., & Ramasubramanian, S. (2015). Intervening in the media's influence on stereotypes of race and ethnicity: The role of media literacy education. Journal of Social Issues, 71(1), 171-185. doi:10.1111/josi.12103

Scheufele, D. A., & Tewksbury, D. (2007). Framing, agenda setting, and priming: The evolution of three media effects models. Journal of Communication, 57(1), 9-20. doi:10.1111/j.0021-9916.2007.00326.x

Shrum, L. J. (2002). Media consumption and perceptions of social reality: Effects and underlying processes. In B. Jennings & D. Zillmann (Eds.), Media effects: Advances in theory and research (pp. 69-95). Mahwah, NJ: Lawrence Erlbaum.

Tal-Or, N., & Cohen, J. (2010). Understanding audience involvement: Conceptualizing and manipulating identification and transportation. Poetics, 38(4), 402-418. doi:10.1016/j.poetic.2010.05.004

Tukachinsky, R., Mastro, D., & Yarchi, M. (2015). Documenting portrayals of race/ethnicity on primetime television over a 20-year span and their association with national-level racial/ethnic attitudes. Journal of Social Issues, 71(1), 17-38. doi:10.1111/josi.12094

Tversky, A. (1977). Features of similarity. Psychological Review, 84, 327-352. doi:10.1037/0033-295X.84.4.327

van Looy, J., Courtois, C., de Vocht, D., & de Marez, L. (2012). Player identification in online games: Validation of a scale for measuring identification in MMOGs. Media Psychology, 15(2), 197-221. doi:10.1080/15213269.2012.674917

Zillmann, D. (1999). Exemplification theory: Judging the whole by some of its parts. Media Psychology, 1(1), 69-94. doi:10.1207/s1532785xmep0101_5

Zillmann, D. (2002). Exemplification theory of media influence. In B. Jennings & D. Zillmann (Eds.), Media effects: Advances in theory and research (pp. 19-41). Mahwah, NJ: Lawrence Erlbaum.

Zillmann, D., & Brosius, H.-B. (2012). Exemplification in communication: The influence of case reports on the perception of issues. New York, NY: Routledge.


University of Florida, USA

T. Franklin Waddell:

Date submitted: 2016-12-22
Table 1. Conditional Effect of Exemplar Typicality and Victim
Expectations at Values of Crime News Attention.

Crime news    Estimate   LLCI  ULCI

Low (-1 SD)    -0.43    -1.49  0.63
Moderate (M)    0.39    -0.35  1.14
High (+1 SD)    1.22     0.15  2.29

Note. Estimate = the product of interaction terms at each level of the
moderator; LLCI = 95% lower-limit confidence interval; and ULCI = 95%
upper-limit confidence interval.

Table 2. Exemplar Typicality x Victim Expectation Interaction, Hostile
Attitudes Toward Blacks.

                          Victim expectations:    Victim expectations:
                          White                   Black
                          M              SE        M           SE

No-exemplar control      2.30           .17       2.00        .19
Atypical exemplar        1.74           .17       2.16        .18
COPYRIGHT 2017 University of Southern California, Annenberg School for Communication & Journalism, Annenberg Press
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2017 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Waddell, T. Franklin
Publication:International journal of communication (Online)
Article Type:Report
Date:Jun 1, 2017
Previous Article:Emerging currents in communication/LGBTQ studies: A review of LGBTQ-related articles published in communication journals from 2010 to 2015.
Next Article:Dan Caspi and Nelly Elias (Eds.), Ethnic Minorities and Media in the Holy Land.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |