imagining reduce prejudice

7/27/2019 Imagining Reduce Prejudice

1/10

Can Imagined Interactions Produce Positive Perceptions?

Reducing Prejudice Through Simulated Social Contact

Richard J. Crisp University of Kent

Rhiannon N. Turner University of Leeds

The contact hypothesis states that, under the right condi-tions, contact between members of different groups leads tomore positive intergroup relations. The authors track re-cent trends in contact theory to the emergence of extended,or indirect, forms of contact. These advances lead to anintriguing proposition: that simply imagining intergroupinteractions can produce more positive perceptions of out-groups. The authors discuss empirical research supportingthe imagined contact proposition and find it to be anapproach that is at once deceptively simple and remarkably

effective. Encouraging people to mentally simulate a pos-itive intergroup encounter leads to improved outgroupattitudes and reduced stereotyping. It curtails intergroupanxiety and extends the attribution of perceivers positivetraits to others. The authors describe the advantages anddisadvantages of imagined contact compared to conven-tional strategies, outline an agenda for future research,and discuss applications for policymakers and educators intheir efforts to encourage more positive intergroup rela-tions.

Keywords: intergroup contact, imagined contact, prejudice

One day you find yourself on a busy train. You get a seat and start

reading the novel you brought with you to pass the time. At thenext stop, an older Black man boards the train and sits down nextto you. After a few minutes, the man looks at what you arereading and comments that it is one of his favorite books. Thisbegins a discussion in which you share your thoughts on thebook and what you both enjoyed about it. The conversationmeanders, and by the time you get off the train, 30 minuteslater, you have discussed a whole range of topics, from thestresses of having to commute to work every day, to what neigh-borhood you live in, to what your childrens favorite subjects areat school.

Most people will have found it easy to imaginethis scenario. There may have been unfamiliar

elements (perhaps they do not often travel ontrains, or perhaps they do not have children), but imagina-tion is a powerful tool, and people can envisage themselvesin almost any situation. The power of imaginative thoughtlies at the heart of their hopes, fears, aspirations, andinnovations. In this article, we argue that psychologists canharness this power to encourage more positive intergrouprelations.

Take our opening illustration. It did not involve just atrain, a book, and a conversation. If you are young, White,or a woman, it also featured an intergroup encounter (with

an older Black man). What might be the impact of imag-ining this encounter? The imagined interaction was a suc-cess: You found lots to talk about, it was a pleasant en-counter, and your trip went more quickly as a result. Youmight therefore reason that if you actually met an olderBlack man on a train in the future, you might have morepositive expectations about how that encounter would go.These positive expectations may make you feel more pos-itive and open toward older Black men in general. The next

time you actually met an older Black man, you might wellfeel less anxious and more confident about starting a con-versation.

The scenario you imagined is pretty much what wehave used in our research on imagined intergroup contact(although, as we describe later, the benefits of the techniquecan be observed with a much less involving narrative). Webelieve that imagined contact is such an exciting prospectbecause it provides a simple, flexible, and effective meansof promoting more positive perceptions of outgroups. Wedo not advocate imagined contact as a replacement forexisting interventions, such as extended or actual contact(which we expect to have a stronger overall impact on

attitude change). Rather, we assert that the value in imag-ined contact is in its ability to encourage people to seek outcontact, to remove inhibitions associated with existing prej-udices, and to prepare people to engage outgroups with anopen mind. We argue that imagined contact could be highlyeffective as a first step on the route toward reconciliationand reduced prejudice, on a continuum of contact thatprovides a road map for the use of multiple contact strat-egies in improving intergroup relations.

In what follows, we track recent trends in contacttheory, from the use of actual contact to the development ofindirect forms of contact. From here, we outline the imag-ined contact proposition, describe supportive empirical

studies, and discuss how we isolated the most effectiveimagined contact narratives. Finally, we consider how fu-ture research might develop the technique further and therange of potential applications it may offer.

Correspondence concerning this article should be addressed to Richard J.Crisp, Centre for the Study of Group Processes, Department of Psychol-ogy, University of Kent, Canterbury, Kent CT2 7NP, United Kingdom, orto Rhiannon N. Turner, Institute of Psychological Sciences, University ofLeeds, Leeds, LS2 9JT, United Kingdom. E-mail: [email protected] [email protected]

231MayJune 2009 American Psychologist

2009 American Psychological Association 0003-066X/09/$12.00Vol. 64, No. 4, 231240 DOI: 10.1037/a0014718


2/10

Recent Trends in Contact Theory

Promoting the peaceful coexistence of different socialgroups is perhaps the biggest social issue people face. Newand emerging conflicts serve as vivid reminders of theimportance of this endeavor; immigration and globalizationunderscore the need for informed policies that encourageco-operation and tolerance. Do people face an inevitable

clash of civilizations (Huntington, 1996) as tensions eruptand escalate around differences in religion and ethnicity?Social psychologists have long been concerned with thisquestion and have sought to understand the relationshipsamong segregation, social contact, and intergroup relations.This work has maintained sustained interest in one of themost important concepts in psychologists developing un-derstanding of the nature of prejudice: intergroup contact.

In 1954, Gordon Allport proposed the idea that con-tact between members of opposing groups, under the rightconditions, would lessen intergroup hostility and lead tomore positive intergroup attitudes. In the intervening yearsthis contact hypothesis has become one of the most exten-sively researched ideas in psychology (Oskamp & Jones,2000) and a recent meta-analysis of 515 of these studies hasconfirmed the core proposition: There is a robust, highlysignificant, negative relationship between contact and prej-udice (Pettigrew & Tropp, 2006). Of importance, this anal-ysis revealed that, although contact that met Allports op-timal conditions (co-operation, common goals, equal socialstatus, and institutional support) led to the greatest reduc-tions in prejudice, prejudice was still reduced in theirabsence. This indicated that these optimal conditions are, infact, best considered facilitating conditions, rather thannecessary conditions.

Key principles of contact have been distilled intointergroup contact theory (Brown & Hewstone, 2005; Pet-

tigrew, 1998), which specifies how, when, and why contactis associated with reduced prejudice. Contact theory hasbecome a sophisticated account of the conditions underwhich intergroup interactions have benefits for intergrouprelations, and it has provided an important guide for diverseapplications of the core idea. For instance, psychologistsnow know that a unique form of contact, cross-groupfriendship, is more effective at improving outgroup atti-tudes than less intimate forms of contact (Pettigrew, 1997;Pettigrew & Tropp, 2006; Turner, Hewstone, & Voci,2007; Turner, Hewstone, Voci, Paolini, & Christ, 2007).

Despite its success, contact theory has had its critics.For instance, some have argued that researchers have been

overly optimistic about the power of contact and that al-though studies have demonstrated prejudice reduction at anindividual level, it is not yet known whether this translatesinto broader societal change (Dixon, Durrheim, & Tredoux,2005). This is an important question and one that can beanswered as contact interventions are tested with greaterfrequency in field settings. Our concern is with a differentpractical limitation: that contact can work only where theopportunity for contact exists. When groups are highlysegregated, physically or socially, or when there is littlemotivation to engage in contact, the benefits of contact may

remain unrealized. Is this an intractable limitation? Weargue that it is not and that the power inherent to theconcept of contact enables an intriguing and enabling prop-osition: that even imagining intergroup contact could im-prove intergroup attitudes.

Opportunity for Contact

Contact can reduce prejudice only when social groups andgroup members have the opportunity to engage positivelywith one another (e.g., Phinney, Ferguson, & Tate, 1997;Turner, Hewstone, &Voci, 2007). Unfortunately, there aremany examples where the local intergroup relations affordfew such opportunities. In the United States, for example,segregation of Latino and White communities remains per-vasive (Martin, 2006), and the average White person livesin a predominantly White neighborhood with less than 10%Black residents (Logan, 2001). Many Catholic and Protes-tant communities in Belfast, Northern Ireland, include avery low percentage of residents from the other commu-nity, and only 5% of Northern Irish children attend mixedCatholicProtestant schools (Office for National Statistics,

2001). There are many other examples of more extremesegregation, from the Green Line in Cyprus to the WestBank Wall in Israel (Pettigrew, 2008). Even in multicul-tural communities where many different groups live inclose proximity to one another, there can be very littleopportunity for meaningful interaction, and people have atendency to interact mainly with those similar to them-selves, especially with regard to age, race, and gender(Graham & Cohen, 1997). For instance, in the multiracialcity of Bradford, United Kingdom, South Asian and Whitecommunities remain largely isolated from one another,with residents living parallel lives. This social segrega-tion is regarded as contributing to Britains worse ever race

riots in July 2001 (Cantle, 2001). In all of these circum-stances, interventions that involve intergroup contact maybe very difficult to establish. Yet, it is in precisely thesesettings that contact interventions are needed the most.

What if there was a way to produce the positive effectsof contact without there being any actual contact? This wasthe motivation behind the development of an ingeniousindirect form of intergroup contact: extended contact.

Extended Contact

According to the extended contact hypothesis (Wright,Aron, McLaughlin-Volpe, & Ropp, 1997), the benefits ofcontact could accrue from vicarious experienceslearning

that people we know engage in positive interactions withmembers of outgroups (even if we ourselves do not). Anumber of studies have now demonstrated the benefits ofextended contact. In two studies, Wright et al. (1997)showed that White respondents who knew at least oneingroup member with an outgroup friend reported weakerprejudice toward that target group than those who had noextended outgroup friends. Moreover, the more extendedoutgroup friends a participant had, the weaker their preju-dice. In a third study, to demonstrate extended cross-groupfriendship experimentally, Wright et al. initially created

232 MayJune 2009 American Psychologist


3/10

intergroup conflict between two small artificial groups. Oneparticipant from each group was then chosen to take part inan unrelated co-operative game designed to create interper-sonal closeness. Individuals who experienced direct inter-group contact returned to discuss the experience with therest of their group. Participants who merely learned aboutthe intergroup contact from the returning member showedless ingroup favoritism than they had before the extended

contact experience.The extended contact effect has subsequently been

demonstrated in a number of studies. In a survey of Cath-olic and Protestant students in Northern Ireland, partici-pants completed a questionnaire focusing on the number ofingroup friends they had who had outgroup friends, theirattitudes toward the opposing community, and their expe-rience of intergroup anxiety (Paolini, Hewstone, Cairns, &Voci, 2004). To measure intergroup anxiety, participantswere asked to think about how they would feel mixingsocially with a group of people from the other community(e.g., how awkward, self-conscious, and defensive theywould feel). Structural equation modeling revealed that

extended contact was associated with lower levels of out-group prejudice, a relationship that was fully mediated byreduced intergroup anxiety. Research has also found thatextended contact between White and Asian undergraduatesproduced positive outgroup attitudes (Turner, Hewstone,Voci & Vonofakou, 2008). In this research, participantswere asked how many White people they knew who hadSouth Asian friends and how this related to a number ofoutcome measures. These included how anxious they felt atthe prospect of future interactions with South Asians (in-tergroup anxiety), how much they thought their Whitepeers liked South Asians (ingroup norms), how much theythought South Asians liked White people (outgroupnorms), and how much overlap they perceived there to be

between themselves and South Asians (inclusion of theoutgroup in the self; Aron, Aron, & Smollan, 1992). Usingstructural equation modeling, Turner et al. (2008) foundthat extended contact was associated with more positiveoutgroup attitudes. Moreover, this relationship was fullymediated by lower intergroup anxiety, more positive per-ceived ingroup norms, more positive perceived outgroupnorms, and greater inclusion of the outgroup in the self.

Extended contact has also been successfully used asan intervention to reduce prejudice among school children.Following an intervention that involved being read storiesabout an ingroup member and a refugee, children betweenages 5 and 11 had significantly more positive attitudes

toward refugees compared with those in a control condition(Cameron, Rutland, Brown, & Douch, 2006). In otherresearch, Finnish teenagers read stories in their schoolsabout same-age ingroup peers engaging in close friendshipswith foreigners (Liebkind & McAlister, 1999). Tolerancetoward foreigners increased or remained stable in schoolswhere the intervention was introduced, whereas it remainedstable or decreased in some of the control schools.

This research is important because it shows that actualexperience of contact with outgroups is not the only waythat contact can have benefits for intergroup relations.

Extended contact may be especially useful when there isless opportunity for contact because it does not depend onpersonal experience, but depends simply on the existenceof some contact somewhere in ones social network. Theimportance of this idea for policymakers and educatorsseeking to develop interventions to reduce prejudice cannotbe understated because it suggests that contact may be a farmore powerful and flexible means of improving intergroup

relations than previously thought.So, just how powerful is the concept of contact?

Researchers know that actual contact is not (at least notalways) a requirement: Simply learning about positive in-tergroup interactions can change intergroup attitudes. Doesextended contact solve the opportunity-for-contact prob-lem? The answer is yes, to some extent, but it cannot do soentirely. Under some circumstances, extended contactcould suffer the same limitation as actual contact. In highlysegregated settings, one simply might not know anyonewho has an outgroup friend (or even anyone who wouldinteract in any meaningful way with an outgroup member).We could call this a lack of extended opportunities for

contact, where no outgroup friends exist within ones widersocial network. This is a characteristic of many real con-texts of conflict, ranging from the Middle East to Cyprus tothe former Yugoslavia. However, if the concept of contactis powerful enough to extend to extended contact, perhapsit can be extended just a little further.

Imagerys Social ImpactPettigrew (1997, 1998, 2008) proposed that to take inter-group contact theory forward, researchers should focus onunderstanding in greater depth how unique forms of contactreduce prejudice. Here, we describe one such new approachthat capitalizes on the basic idea of contact but does so ina way that discounts the limits of opportunity, an approach

based on the mental simulation of social contact.The minds tremendous capacity for imagination has

captivated psychologists since the earliest enquiries (Gal-ton, 1883; James, 1890), and it is a concept that hasenjoyed enduring appeal. Mental imagery has been foundto elicit emotional and motivational responses similar tothose of real experiences (Dadds, Bovbjerg, Redd, & Cut-more, 1997), and neuropsychological studies have shownthat mental imagery shares the same neurological basis asperception and uses neurological mechanisms similar tothose for memory, emotion, and motor control (Kosslyn,Ganis, & Thompson, 2001). Functionally, mental imageryserves as an important element in the selection, rehearsal,

preparation, and planning of action and goal-directed be-havior (Marks, 1999), and it has been shown to havesignificant benefits for activities ranging from academicability to sporting performance (Taylor, Pham, Rivkin, &Armor, 1998).

There is a precedent for the social impact of imagery.Recent research has shown that simply imagining a partic-ular social situation can have the same effect as the expe-rience itself. Research has found that after imagining a(counterstereotypic) strong woman, participants demon-strated fewer implicit stereotypes than participants who



4/10

engaged in neutral or stereotypical mental imagery (imag-ining a weak woman or a strong man) or participants whohad not engaged in any imagery (Blair, Ma, & Lenton,2001). Other research has investigated the role of imagin-ing a social context on the bystander apathy effect, the ideathat people are less likely to help others if other people arepresent. Research has found that simply imagining being ina large group can lead to significantly less helping behavior

on a subsequent task (Garcia, Weaver, Moskowitz, & Dar-ley, 2002). In their study, Garcia et al. (2002) found thatparticipants who imagined having a meal out with 10 otherpeople were subsequently less likely to help the experi-menter by participating in a second study than those whoimagined having a meal out with just 1 other person. Ittherefore seems that imagining both people and situationshas a significant impact on perceptions, attitudes, and at-tributions. It turns out that when person and situation arecombined in imagined interactions, something similar oc-curs.

Imagined Intergroup Contact

Imagined intergroup contact is the mental simulation of asocial interaction with a member or members of an out-group category. The basic idea is that mentally simulatinga positive contact experience activates concepts normallyassociated with successful interactions with members ofother groups. These can include feeling more comfortableand less apprehensive about the prospect of future contactwith the group, and this reduced anxiety should reducenegative outgroup attitudes. According to Garcia et al.(2002), imagery increases the accessibility of abstract con-cepts associated with that social context. Imagining beingin a crowd, for example, activates feelings of being lost ina crowd and unaccountable, feelings that are associatedwith less helping behavior in real situations. Similarly,

when people imagine intergroup contact, they should en-gage in conscious processes that parallel the processesinvolved in actual intergroup contact. They may, for ex-ample, actively think about what they would learn aboutthe outgroup member, how they would feel during theinteraction, and how this would influence their perceptionsof that outgroup member and the outgroup more generally.In turn, this should lead to more positive evaluations of theoutgroup, similar to the effects of face-to-face contact.

At the start of this article, we presented a fairly elab-orate imagined contact narrative. In fact, we have found amuch more minimal instructional set is sufficient to yieldpositive effects. The core task instructions are very simple:

We would like you to take a minute to imagine yourselfmeeting [an outgroup] stranger for the first time. Imaginethat the interaction is positive, relaxed, and comfortable.This simple phrase includes two key elements that we havefound to be the critical components. First is the instructionto engage in simulation. We have found that runningthrough the mental script of an interaction is critical forobserving positive effects (thinking, in contrast, of just anoutgroup member in the absence of any simulated interac-tion has no positive effects on attitudes; Turner, Crisp, &Lambert, 2007, Experiment 2). Second is a positive tone.

We know that this is important for direct contact, and it isthe same for imagined contact. Empirically, we have shownthat imagined contact works better when it is positivecompared to neutral (Stathi & Crisp, 2008, Experiment 1).A positive tone is also important to guard against a possiblenegative tone, which might emerge if participants are left totheir own devices (this may be likely, especially in contextsof conflict or extreme segregation). Other than these two

core elements, we have found that other embellishmentshave very little impact on the basic effect. For instance,describing the imagined encounter as interesting has noadditional impact (Stathi & Crisp, 2008, Experiments 1 and3), and there are no effects of adding detail like that used inthe narrative at the start of this article (Turner, Crisp, &Lambert, 2007, Experiment 1).

In summary, unlike actual contact, from the outset theimagined contact effect seems unfettered by a raft of nec-essary conditions, other than the two core principles wehave identified. Indeed, its applicability and effectivenessare part of the power and flexibility of the approach. Ofcourse, as with actual contact, further research may well

uncover a range of facilitating conditions that improve itseffectiveness. We return to a discussion of these possibleconditions at the end of the article. Next, however, wediscuss the empirically observed impacts of imagined con-tact, before finally discussing its potential applications forpolicy and practice.

Impacts of Imagined Contact

Improved Intergroup Attitudes

In three studies, we found that participants who were askedto imagine a positive interaction with an outgroup membersubsequently expressed more positive attitudes and stereo-typed less than participants who did not (Turner, Crisp, &

Lambert, 2007). Two studies showed that young partici-pants who imagined a scenario in which they participatedin a short positive interaction with an older person showedless ingroup favoring bias in subsequent evaluations (as-sessed immediately after the task). This was the casewhether participants imagined contact compared to simplyimagining an outdoor scene (Experiment 1) or compared tosimply thinking about an older person (Experiment 2; i.e.,without any interaction involved). In a third study, wefocused on attitudes of heterosexual men toward homosex-ual men (see Herek & Capitanio, 1996; Vonofakou, Hew-stone, & Voci, 2007, for actual contact studies with thesegroups). Participants who imagined talking to a homosex-ual man on a train subsequently evaluated homosexual menin general more positively and stereotyped homosexualmen less (perceived less homogeneity) than participantswho imagined an outdoor scene.

Projection to Outgroups

Research has shown that positively toned imagined contactleads to greater projection of positive traits to the targetoutgroup (Stathi & Crisp, 2008). Projection is a process bywhich traits and attitudes are attributed to others and canconstitute a fundamental cognitive basis for ingroup fa-



5/10

voritism (Robbins & Krueger, 2005, p. 42; see alsoCadinu & Rothbart, 1996). This is because projection ofpositive self traits to similar others (i.e., the ingroup) isgenerally stronger for ingroups than outgroups (Clement &Krueger, 2002) and occurs in lieu of stereotyping (Ames,2004). If positive contact can make relevant outgroupmembers seem more similar to self, which we know is thecase for actual and extended contact (Eller & Abrams,

2004; Wright et al., 1997), then those outgroup membersshould benefit from the positive projection that ingroupmembers typically benefit from (Robbins & Krueger,2005). Imagined contact did indeed lead to more positivetrait projection to outgroups following imagined contact,compared to controls, with a variety of target groups,including Mestizos in Mexico and international students inthe United Kingdom (Stathi & Crisp, 2008). Notably, thisprojection was consistent with the lower perceived homo-geneity of outgroups observed by Turner, Crisp, and Lam-bert (2007, Experiment 3).

Consistent with projection theory and the idea thatimagined contact activates cognitivebehavioral processes

similar to those of actual contact, projection was greatestunder three conditions. Projection should be higher whenthe personal self is more salient than the collective selfbecause the personal self is the informational base used forprojection (Cadinu & Rothbart, 1996). Correspondingly,majorities (who have low personal self-salience) projectedmore than minorities (who have high collective self-sa-lience). Moreover, lower identifiers (who have high per-sonal self-salience) projected more than higher identifiers(who have high collective self-salience), and a direct ma-nipulation of personal self-salience versus social self-sa-lience also led to greater projection (Stathi & Crisp, 2008).

Reduced Anxiety

A key mediating process that accounts for the positiveimpact of actual contact on intergroup attitudes is anxiety(Islam & Hewstone, 1993). Correspondingly, our researchhas established intergroup anxiety as a mediator of imag-ined contact effects (Turner, Crisp, & Lambert, 2007).Intergroup anxiety is the negative emotional reaction thatcan occur at the prospect of having to engage in an inter-group encounter. It is most likely to arise when there hasbeen minimal previous contact and when there are largedifferences in status (Stephan & Stephan, 1985). Whenindividuals have had a successful interaction with an out-group member, however, their level of intergroup anxiety islikely to be reduced, as they come to realize they have

nothing to fear from such interactions. Consistent with thisreasoning, a number of studies in diverse intergroup set-tings have found the effect of intergroup contact on reduc-ing prejudice to be mediated by intergroup anxiety (Islam& Hewstone, 1993; Paolini et al., 2004; Turner, Hewstone,&Voci, 2007; Voci & Hewstone, 2003).

In one study, male heterosexual participants wereasked to imagine a positive interaction with a homosexualman (Turner, Crisp, & Lambert, 2007, Experiment 3). Asnoted earlier, participants who spent a few minutes imag-ining intergroup contact subsequently had a more positive

attitude toward gay men in general and also perceived thereto be greater variability among the outgroup than partici-pants in the control condition. We found that the positiverelationship between imagined contact and outgroup atti-tude was mediated by reduced intergroup anxiety. Anxietydid not, however, mediate the relationship between imag-ined contact and outgroup variability. This parallels therelationships observed in research on actual contact. Al-

though some direct contact studies have found intergroupanxiety to mediate the effect of intergroup contact onoutgroup variability (e.g., Islam & Hewstone, 1993; Voci& Hewstone, 2003), the evidence for this is less consistentthan the effect of intergroup anxiety on affective outcomemeasures, such as feelings of intergroup comfort and liking(e.g., Tropp & Pettigrew, 2005; Turner, Hewstone, & Voci,2007).

Reduced Stereotype Threat

Imagined contact can also reduce self-stereotyping and,correspondingly, the impact of negative self-stereotypes onquantitative performance (the stereotype threat effect,

Steele, 1997; see Abrams et al., 2008; Crisp & Abrams,2008). Imagined contact could serve a protective functionfor older people exposed to contexts where they mightotherwise suffer performance decrements. Research witholder people has found that self-stereotyping affects a rangeof cognitive abilities consistent with the stereotype thatcognitive performance declines with age (Hess, Hinson, &Statham, 2004; Levy, 1996). The application of imaginedcontact to this domain was based on the premise thatintergenerational contact is generally limited (Hagestad &Uhlenberg, 2005) and that actual contact has been found toreduce threat effects in older people through reduced anx-iety (Abrams, Eller, & Bryant, 2006). Research has foundthat older people (all over 60 years old) who imagined ashort social interaction with a young stranger (compared toa control) were subsequently immune to the depletingeffects of a threat comparison (i.e., a stated comparison ofthe performance of older and younger people) on cognitivetest performance (see Abrams et al., 2008).

Alternative Explanations

Although the research programs outlined here demonstratethe range of benefits that accrue from imagined contact, wehave also been careful to rule out alternative explanations,such as cognitive load, demand characteristics, priming,and generalized positive affect.

Cognitive LoadPerhaps imagined contact effects are not about imaginingcontact at all but are just about the cognitive activityrequired to engage with the task. A number of studies havedemonstrated, however, that imagined contact has benefitsnot only compared to baseline (no task), but also relative tocontrol conditions that have equivalent cognitive load andrequire identical mental operations. We have found, forexample, that imagined contact reduced the expression ofbiased evaluations compared to imagining an outdoor scene



6/10

(Turner, Crisp, & Lambert, 2007, Experiments 1 and 3).This was similarly the case when we examined the effect ofimagined contact on trait projection (Stathi & Crisp, 2008,Experiment 3). Although a priori it is difficult to see howsimply engaging in mental imagery in and of itself mightlead to more positive attitudes (i.e., vs. specifically engag-ing in imagined contact), these studies help to rule thispossibility out as a competing explanation.

Demand Characteristics

Perhaps the imagined contact effect is not about imaginingcontact at all but is just about participants guessing whatthe task is meant to do. We have, however, found noevidence to suggest that imagined contact effects can beexplained by demand characteristics. Typically, no partic-ipants report any awareness of the experimental hypothesesat feedback. For instance, in Turner, Crisp, and Lambert(2007) only four participants reported any suspicion aboutthe purpose of the experiment, and not one participantsuccessfully identified the aims of the experiment. None-theless, to better rule out this explanation, we have exam-

ined the effects of imagined contact on implicit measures ofattitudes (Turner & Crisp, in press). Whereas explicit atti-tudes are conscious, deliberative, and controllable (and areusually captured by traditional self-report measures), im-plicit attitudes are unintentionally activated by the merepresence (actual or symbolic) of an attitude object and aretherefore less likely to be influenced by social desirabilitythan are explicit measures. We asked young participants toimagine talking to an older stranger, using the same in-structions as in Turner, Crisp, and Lambert (2007), beforecompleting an explicit outgroup attitude measure and ameasure of implicit attitude, the Implicit Association Test(Greenwald, McGee, & Schwartz, 1998). Participants who

had imagined contact with an older person not only sub-sequently showed more positive explicit, self-reported, out-group attitudes toward older people in general, but alsoshowed more positive implicit outgroup attitudes on ayoungolder version of the Implicit Association Test.

Stereotype Priming

Perhaps the imagined contact effect is not about imaginingcontact at all but is just about thinking about the outgroup.The argument here is that imagined contact simply primesparticipants to think of the outgroup (e.g., older people),which then invokes a conscious attempt to regulate behav-ior to appear nonprejudiced (e.g., Devine & Monteith,

1999). On the basis of feedback received from participants,we are confident that a priming effect cannot account forour findings. However, to empirically rule out this expla-nation, we carried out an experiment in which we sought toreplicate the imagined contact effect but used a controlcondition in which we simply primed the outgroup (Turner,Crisp, & Lambert, 2007, Experiment 2). We observed lessbias in the imagined contact condition compared to thecontrol condition. This is consistent with the idea thatimagined contact does not simply involve category primingand associated self-regulation, but that there is something

special about mentally simulating the intergroup interac-tion.

Positive Affect

Perhaps the imagined contact effect is not about imaginingcontact at all but is just about the positive feelings that areassociated with the interactive nature of the task. It iscertainly the case that contact works better when it is

positively toned (Pettigrew & Tropp, 2006). Consistentwith this, imagining positive interactions is more effectivethan imagining neutrally valenced interactions (Stathi &Crisp, 2008, Experiment 1). However, this also raises thequestion of whether it is, in fact, just the positive charac-teristics of imagined contact that are responsible for thepositive effects. In other words, perhaps simply thinkingabout some positive interaction (regardless of whether itfeatures the outgroup) yields more positive evaluations.

To rule out this possibility, an experiment testedwhether imagined contact has benefits (for perceptions ofFrench targets by British participants) because of the iden-tity of the target or just because of the interactive nature of

the imagined scene (Stathi & Crisp, 2008, Experiment 2).This was achieved by comparing imagined scenarios thatwere identical in both conditions, except that the interact-ing partner was a specified outgroup member in only onecondition (in this case, a French person for our Britishparticipants). In both conditions, participants imagined in-teracting with a stranger at a party, but only in the exper-imental condition was the interaction partner an outgroupmember. Imagined contact was significantly more effectivein the condition featuring the outgroup member.

Future Research

It is important to extend our knowledge of imagined con-

tact effects in several areas. We must (a) specify optimizingconditions, (b) distinguish underlying mechanisms, and (c)elaborate on the range of positive impacts that the inter-vention can have.

Optimizing Conditions

Future research should establish the conditions that bestfacilitate the positive impacts of imagined contact. It isimportant to be mindful that the same intervention canmean different things to different people (see Crisp, Stone,& Hall, 2006, for a detailed discussion of such issues). Forinstance, imagined contact can lead to greater projection ofpositive traits to outgroups for those individuals who are

less committed to their ingroup (Stathi & Crisp, 2008).Uncovering how such individual differences moderateimagined contact effects is an important endeavor, espe-cially if psychologists are to move toward effective imple-mentation. Perceived authenticity of the imagined encoun-ter is also important. In clinical and counseling psychology,role playing is perceived as a more authentic representationof reality when it features actors who are culturally similar(Pedersen, 2000). In imagined contact, by definition, theactors are mismatched (and the lower likelihood of actualinteractions is the motivation to encourage imagined inter-



7/10

actions). Nonetheless, perceived authenticity of the imag-ined encounter may be an important variable in futureresearch.

Other important questions are how long do the effectsof the intervention last (in the studies we have reported,attitudes are typically assessed immediately after the imag-ined contact task), and how many times must the task berepeated to internalize attitude change? These are imple-

mentation issues that are common to all forms of contactintervention (or, indeed, to all attitude change interven-tions). However, there may be some possible routes forfurther research that are unique to imagined contact. Forinstance, longer term experience of imagined contact mayhelp to promote an internalized self-supervision process,whereby more general adherence to egalitarian norms be-comes a guiding principle. How many times imaginedcontact should be implemented to achieve such internalizedchanges and what form this implementation should take arepertinent questions for future research.

Underlying Mechanisms

We have found that, like actual and extended contact,imagined contact can improve attitudes by reduced anxiety(Turner, Crisp, & Lambert, 2007). A reasonable question iswhether imagined contact shares other mediating mecha-nisms with established forms of contact. For instance,extended contact works not only by reducing intergroupanxiety, but also by generating positive norms for inter-group relations and increasing the extent to which theoutgroup is included in the self. These processes are basedon a strong theoretical rationale (Wright et al., 1997) andare supported by empirical evidence in a variety of inter-group contexts testing extended contact (e.g., British chil-drens attitudes toward refugees, Cameron et al., 2006;Catholics and Protestants in Northern Ireland, Paolini et al.,

2004; Whites and Asians in the United Kingdom, Turner etal., 2008; Turner, Hewstone, & Voci, 2007). To appropri-ately position imagined contact in line with other methods,researchers should examine the possible role these multiplemediators play in imagined contact effects.

We believe that although imagined contact shares anumber of similarities with extended contact, it may bedistinct in terms of the social norm mechanism. In extendedcontact, participants observe an ingroup member behavingpositively toward an outgroup member, apparently reflect-ing positive regard. This positive model constitutes aningroup norm that uniquely mediates extended contact butnot actual contact (see Turner et al., 2008). This is because

extended contact involves a perceptual focus on anotheringroup member, whereas actual contact does not. In actualcontact, the perceiver is focused only on the outgroup andhas no ingroup model from which to derive a behavioralnorm. Because the instructional set used in imagined con-tact, like that for actual contact, focuses participants on theoutgroup, it is likely that ingroup norms will be unaffected.Thus, although extended and imagined contact are bothindirect forms of contact, they can be distinguished by (atleast) one mechanism. This also highlights a similaritybetween actual and imagined contact that extended contact

does not share: the mental simulation of ones personalengagement with the outgroup. This brief analysis illus-trates some of the theoretical similarities and differencesamong actual, extended, and imagined contact. Furtherspecifying the psychological and functional distinctionsamong all three approaches is an important task for futureresearch.

Positive ImpactsWe have seen that imagined contact reduces anxiety aboutthe prospect of future intergroup encounters (Turner, Crisp,& Lambert, 2007). As such, the positive impact of imag-ined contact may not be restricted to changes in attitudesbut could have more direct benefits for actual interactions.

To elaborate, intergroup anxiety can arise as a conse-quence of negative expectations of rejection or discrimina-tion during cross-group interactions or because of fears thatthe interaction partner, or the respondents themselves, maybehave in an incompetent or offensive manner (Stephan &Stephan, 1985). These fears may lead people to avoidintergroup contact altogether. It has been found, for exam-

ple, that the more negative White participants expectationswere about interacting with Black people, the more theyreported avoiding interracial encounters (Plant & Devine,2003). Similarly, people often explain their failure to ini-tiate intergroup contact in terms of their fear of beingrejected by the outgroup member (Shelton & Richeson,2005). Even if contact is initiated, intergroup anxiety mayincrease the likelihood that individuals will enter the en-counter with feelings of hostility. In turn, this increases thelikelihood that group members will interpret the interactionin a negative light, with negative consequences for inter-group relations. If, however, participants first spend sometime imagining the encounter, their levels of anxiety will belower and their attitudes more positive when they subse-quently enter the interaction. Imagined contact may there-fore increase the chances that a subsequent intergroupencounter will be of a high quality and will result in furtherpositive attitude change. Examining the direct behavioralimpact of imagined contact is therefore an important focusfor future research.

Limitations, Implications, andApplications

We have demonstrated the benefits of imagined contact ina variety of intergroup contexts. We have, for example,showed how imagined contact can change young peoples

attitudes toward older people (Turner, Crisp, & Lambert,2007, Studies 1 and 2), straight mens attitudes toward gaymen (Turner, Crisp, & Lambert, 2007, Study 3), and Mex-ican peoples attitudes toward Mestizos in Mexico (Stathi& Crisp, 2008). We have also observed benefits for implicitintergroup attitudes (Turner & Crisp, in press) and even apositive impact on self-stereotyping and stereotype threat(Abrams et al., 2008; Crisp & Abrams, 2008). Nonetheless,there are some things that imagined contact cannot do, andunderstanding its limitations is an essential part of devel-oping effective implementation programs.



8/10

The most important limitation is that imagined contactprobably does not have as powerful or long lasting an effectas more direct forms of intergroup contact. This is becausedirect experiences are thought to produce stronger attitudeson an issue than indirect experiences (Fazio, Powell, &Herr, 1983; Stangor, Sullivan, & Ford, 1991). Supportingthis premise, research comparing actual and extended con-tact typically shows actual contact to have the stronger

impact on prejudice (Paolini et al., 2004; Turner, Hew-stone, & Voci, 2007). Given that imagined contact, likeextended contact, is less direct than actual contact, wemight expect it to also have a weaker or more temporaryeffect on intergroup attitudes.

To counter this limitation, imagined contact has adistinctive advantage over both direct and extended con-tact. As we discussed at the start of this article, actualcontact can yield benefits only when there is opportunityfor contact. Although extended contact avoids this problemto some extent, it still requires knowledge of outgroupcontact somewhere in ones wider social network. Imag-ined contact requires no experience of outgroup contact,

either direct or extended.It is this unique contribution of imagined contact thatwe believe presents the most exciting prospects for futureapplication. This is not because it can serve as a replace-ment for existing approachesit cannot. We know thatexisting contact interventions have a highly significant (andstronger) impact on intergroup attitudes. Rather, it is be-cause it provides access to these other forms of contactwhere they might otherwise be unavailable. We believethat one of the most tangible benefits of imagined contactis its ability to break inhibitions that go hand in hand withexisting prejudices. Imagined contact encourages people toseek out contact and prepares them to engage members ofoutgroups with an open mind. We know that imagining an

event reliably increases the likelihood that the event willoccur and that individuals are more likely to carry out animagined target behavior (Carroll, 1978); imagined contactmay have a similar impact on future contact intentions.Used in this way, imagined contact might be highly effec-tive as a first step on the route toward reconciliation andreduced prejudice. For this reason, imagined contact mightbe usefully applied immediately before an intervention thatinvolves extended or direct contact.

Thinking a little more broadly, an intriguing possibil-ity is that imagined, extended, and actual contact mightform a continuum of psychological interventions that aremaximally effective at different stages of social integration.

At early stages of coexistence, there may be high segrega-tion and little opportunity for contact. In this situation,imagined contact may be the only viable intervention tohelp encourage attitude change and intentions to engage inpreliminary contact (or to ensure that when that contactdoes occur, it does so with open minds and an increasedchance of success). At intermediate stages, when bound-aries have begun to permeate and some positive interac-tions have been initiated, extended contact works well toreinforce the impact of isolated (but known) contact en-counters. Increasing extended contact may then lead to a

cascade of positive interactions, along with all the benefitsassociated with actual intergroup contact.

From a practical perspective, imagined contact mayprove important for the application of contact interventionsby policymakers and educators. Community leaders in ar-eas with pervasive segregation and tensions between com-munities recognize that positive intergroup contact mightreduce prejudice, but they are often faced with practical

difficulties that make such changes difficult. White flightfrom inner city areas to the suburbs has resulted in ethnicminority enclaves in the inner cities of Britain, whereasWhite communities dominate the outer suburbs. A similarphenomenon has occurred in public schools in San Jose,California, where the proportion of White students hasdropped by half over the past 10 years (Hwang, 2005).Such situations cannot easily be remedied because they areoften related to economic and social inequalities in thewider society. When intergroup contact does occur, theanticipation of the encounter for those who have had littleprior experience with the outgroup may lead to intergroupanxiety, with potentially negative consequences.

Imagined contact, however, is an inexpensive andpractical means of reducing intergroup anxiety and improv-ing attitudes, and it can yield benefits even when opportu-nities for direct and extended contact are limited. In socialsettings where positive contact is not possible because ofsegregation or lack of opportunity, the knowledge thatimagining contact can create positive attitudes that approx-imate those associated with actual contact is invaluable forpolicymakers and educators. For example, schools coulddevelop and apply teaching techniques that encourage con-tact imagery to bring groups closer together and promotetolerance. To date, relatively few social psychological in-terventions to reduce prejudice have been implemented in

schools. However, we believe that imagined contacteither alone or in combination with existing interven-tionscan provide a simple and practical means of inject-ing social psychological content into educationalinterventions.

Concluding Remarks

In this article, we described a new intervention strategy forimproving intergroup relations: imagined intergroup con-tact. Preliminary empirical research has shown that theapproach is deceptively simple but remarkably effective.Encouraging people to mentally simulate a positive en-counter with someone from an outgroup promotes morepositive attitudes toward that group. Imagined contact pro-vides a firmly grounded intervention strategy with signifi-cant potential application for policymakers and educatorsseeking to promote tolerance for social diversity. We be-lieve this is the strongest testament to the power, flexibility,and enduring appeal of the contact hypothesis. Furtherrefinement of this and other applications of the contactconcept will elaborate, extend, and establish such methodsas viable, defined, and tested strategies for improving in-tergroup relations.



9/10

REFERENCES

Abrams, D., Crisp, R. J., Marques, S., Fagg, E., Bedford, L., & Provias,D. (2008). Threat inoculation: Experienced and imagined intergenera-tional contact prevents stereotype threat effects on older peoples mathperformance. Psychology and Aging, 23, 934939.

Abrams, D., Eller, A., & Bryant, J. (2006). An age apart: Effects ofintergenerational contact and stereotype threat on performance andintergroup bias. Psychology and Aging, 2, 691702.

Allport, G. W. (1954). The nature of prejudice. Cambridge, MA: Perseus

Books.Ames, D. (2004). Inside the mind-readers toolkit: Projection and stereo-

typing in mental state inference. Journal of Personality and SocialPsychology, 87, 340353.

Aron, A., Aron, E. N., & Smollan, D. (1992). Inclusion of Other in theSelf Scale and the structure of interpersonal closeness. Journal ofPersonality and Social Psychology, 63, 596612.

Blair, I. V., Ma, J. E., & Lenton, A. P. (2001). Imagining stereotypesaway: The moderation of implicit stereotypes through mental imagery.

Journal of Personality and Social Psychology, 81, 828841.Brown, R., & Hewstone, H. (2005). An integrative theory of intergroup

contact. In M. P. Zanna (Ed.), Advances in experimental social psy-chology (Vol. 37, pp. 255343). San Diego, CA: Academic Press.

Cadinu, M. R., & Rothbart, M. (1996). Self-anchoring and differentiationprocesses in the minimal group setting. Journal of Personality andSocial Psychology, 70, 661677.

Cameron, L., Rutland, A., Brown, R., & Douch, R. (2006). Changingchildrens intergroup attitudes towards refugees: Testing different mod-els of extended contact. Child Development, 77, 12081219.

Cantle, T. (2001). Community cohesion: A report of the independentreview team. London: Home Office.

Carroll, J. S. (1978). The effort of imagining an effect on expectations forthe event: An interpretation in terms of the availability heuristic. Jour-nal of Experimental Social Psychology, 14, 8896.

Clement, R. W., & Krueger, J. (2002). Social categorization moderatessocial projection. Journal of Experimental Social Psychology, 38, 219231.

Crisp, R. J., & Abrams, D. (2008). Improving intergroup attitudes andreducing stereotype threat: An integrated contact model. In W. Stroebe& M. Hewstone (Eds.), European review of social psychology (Vol. 19,pp. 242284). Hove, England: Psychology Press.

Crisp, R. J., Stone, C. H., & Hall, N. R. (2006). Recategorization andsubgroup identification: Predicting and preventing threats from com-

mon ingroups. Personality and Social Psychology Bulletin, 32, 230243.

Dadds, M. R., Bovbjerg, D. H., Redd, W. H., & Cutmore, T. R. (1997).Imagery in human classical conditioning. Psychological Bulletin, 122,89103.

Devine, P. G., & Monteith, M. J. (1999). Automaticity and control instereotyping. In S. Chaiken & Y. Trope (Eds.), Dual-process theories insocial psychology (pp. 339360). New York: Guilford Press.

Dixon, J., Durrheim, K., & Tredoux, C. (2005). Beyond the optimalstrategy: A reality check for the contact hypothesis. American Psychol-ogist, 60, 697711.

Eller, A., & Abrams, D. (2004). Come together: Longitudinal compari-sons of Pettigrews reformulated intergroup contact model and thecommon ingroup identity model in AngloFrench and MexicanAmeri-can contexts. European Journal of Social Psychology, 34, 229256.

Fazio, R. H., Powell, M. C., & Herr, P. M. (1983). Toward a process

model of the attitudebehavior relation: Accessing ones attitude uponmere observation of the attitude object. Journal of Personality andSocial Psychology, 44, 723735.

Galton, F. (1883). Inquiries into human faculty and its development.London: Macmillan.

Garcia, S. M., Weaver, K., Moskowitz, G. B., & Darley, J. M. (2002).Crowded minds: The implicit bystander effect. Journal of Personalityand Social Psychology, 83, 843853.

Graham, J. A., & Cohen, R. (1997). Race and sex as factors in childrenssociometric ratings and friendship choices. Social Development, 6,355372.

Greenwald, A. G., McGee, D. E., & Schwartz, J. L. K. (1998).Measuring individual differences in implicit cognition: The implicit

association test. Journal of Personality and Social Psychology, 74,14641480.

Hagestad, G. O., & Uhlenberg, P. (2005). The social separation of old andyoung: A root of ageism. Journal of Social Issues, 61, 343360.

Herek, G. M., & Capitanio, J. (1996). Some of my best friends:Intergroup contact, concealable stigma, and heterosexuals attitudestoward gay men and lesbians. Personality and Social Psychology Bul-letin, 22, 412424.

Hess, T. M., Hinson, J. T., & Statham, J. A. (2004). Explicit and implicitstereotype activation effects on memory: Do age and awareness mod-

erate the impact of priming? Psychology and Aging, 19, 495505.Huntington, S. P. (1996). The clash of civilizations and the remaking of

world order. New York: Simon & Schuster.Hwang, S. (2005). The new white flight. Wall Street Journal: Online.

Retrieved March 19, 2008, from http://online.wsj.com/article_email/SB113236377590902105lMyQjAxMDE1MzEyOTMxNjkzWj.html

Islam, R. M., & Hewstone, M. (1993). Dimension of contact as predictorsof intergroup anxiety, perceived outgroup variability and outgroupattitudes: An integrative model. Personality and Social Psychology

Bulletin, 19, 700710.James, W. (1890). The principles of psychology (Vol. 2). New York:

Dover.Kosslyn, S. M., Ganis, G., & Thompson, W. L. (2001). Neural founda-

tions of imagery. Nature Reviews Neuroscience, 2, 635642.Levy, B. R. (1996). Improving memory in old age by implicit self-

stereotyping. Journal of Personality and Social Psychology, 71, 1092

1107.Liebkind, K., & McAlister, A. L. (1999). Extended contact through peer

modelling to promote tolerance in Finland. European Journal of SocialPsychology, 29, 765780.

Logan, J. (2001). Ethnic diversity grows, neighborhood integration lagsbehind. Albany: State University of New York at Albany, LewisMumford Center.

Marks, D. F. (1999). Consciousness, mental imagery and action. BritishJournal of Psychology, 90, 567585.

Martin, M. E. (2006). Residential segregation patterns of Latinos in theUnited States, 19902000. New York: Routledge.

Office for National Statistics. (2001). Neighbourhood statistics. RetrievedJuly 24, 2006, from http://neighbourhood.statistics.gov.uk

Oskamp, S., & Jones, J. M. (2000). Promising practice in reducingprejudice: A report from the Presidents Initiative on Race. In S.Oskamp (Ed.), Reducing prejudice and discrimination (pp. 319 334).

Mahwah, NJ: Erlbaum.Paolini, S., Hewstone, M., Cairns, E., & Voci, A. (2004). Effects of direct

and indirect cross-group friendships on judgments of Catholics andProtestants in Northern Ireland: The mediating role of an anxiety-reduction mechanism. Personality and Social Psychology Bulletin, 30,770786.

Pedersen, P. (2000). Hidden messages in culture centered counseling.Thousand Oaks, CA: Sage.

Pettigrew, T. F. (1997). Generalized intergroup contact effects on preju-dice. Personality and Social Psychology Bulletin, 23, 173185.

Pettigrew, T. F. (1998). Intergroup contact: Theory, research and newperspectives. Annual Review of Psychology, 49, 6585.

Pettigrew, T. F. (2008). Future directions for intergroup contact theoryand research. International Journal of Intercultural Relations, 32, 187199.

Pettigrew, T. F., & Tropp, L. R. (2006). A meta-analytic test of intergroup

contact theory. Journal of Personality and Social Psychology, 90,751783.Phinney, J. S., Ferguson, D. L., & Tate, J. D. (1997). Intergroup attitudes

among ethnic minority adolescents: A causal model. Child Develop-ment, 68, 955969.

Plant, E. A., & Devine, P. G. (2003). The antecedents and implications ofinterracial anxiety. Personality and Social Psychology Bulletin, 29,790801.

Robbins, J. M., & Krueger, J. I. (2005). Social projection to ingroups andto outgroups: A review and meta-analysis. Personality and SocialPsychology Review, 9, 3247.

Shelton, J. N., & Richeson, J. A. (2005). Intergroup contact and pluralisticignorance. Journal of Personality and Social Psychology, 88, 91107.



10/10

Stangor, C., Sullivan, L. A., & Ford, T. E. (1991). Affective and cognitivedeterminants of prejudice. Social Cognition, 9, 359380.

Stathi, S., & Crisp, R. J. (2008). Imagining intergroup contact promotesprojection to outgroups. Journal of Experimental Social Psychology,44, 943957.

Steele, C. M. (1997). A threat in the air: How stereotypes shape theintellectual identities and performance of women and African Ameri-cans. American Psychologist, 52, 613629.

Stephan, W. G., & Stephan, C. W. (1985). Intergroup anxiety. Journal ofSocial Issues, 41, 157176.

Taylor, S. E., Pham, L. B., Rivkin, I. D., & Armor, D. A. (1998).Harnessing the imagination: Mental simulation, self-regulation, andcoping. American Psychologist, 53, 429439.

Tropp, L. R., & Pettigrew, T. F. (2005). Differential relationships betweenintergroup contact and affective and cognitive dimensions of prejudice.Personality and Social Psychology Bulletin, 31, 11451158.

Turner, R. N., & Crisp, R. J. (in press). Imagining intergroup contactreduces implicit prejudice. British Journal of Social Psychology.

Turner, R. N., Crisp, R. J., & Lambert, E. (2007). Imagining intergroupcontact can improve intergroup attitudes. Group Processes and Inter-group Relations, 10, 427441.

Turner, R. N., Hewstone, M., & Voci, A. (2007). Reducing explicit and

implicit prejudice via direct and extended contact: The mediating roleof self-disclosure and intergroup anxiety. Journal of Personality andSocial Psychology, 93, 369388.

Turner, R. N., Hewstone, M., Voci, A., Paolini, S., & Christ, O. (2007).Reducing prejudice via direct and extended cross-group friendship.

European Review of Social Psychology, 18, 212255.Turner, R. N., Hewstone, M., Voci, A., & Vonofakou, C. (2008). A test of

the extended intergroup contact hypothesis: The mediating role ofintergroup anxiety, perceived ingroup and outgroup norms, and inclu-sion of the outgroup in the self. Journal of Personality and Social

Psychology, 95, 843860.Voci, A., & Hewstone, M. (2003). Intergroup contact and prejudice

towards immigrants in Italy: The mediational role of anxiety and themoderational role of group salience. Group Processes and Intergroup

Relations, 6, 3754.Vonofakou, C., Hewstone, M., & Voci, A. (2007). Contact with outgroup

friends as a predictor of meta-attitudinal strength and accessibility ofattitudes toward gay men. Journal of Personality and Social Psychol-ogy, 92, 804820.

Wright, S. C., Aron, A., McLaughlin-Volpe, T., & Ropp, S. A. (1997).The extended contact effect: Knowledge of cross-group friendships andprejudice. Journal of Personality and Social Psychology, 73, 7390.


imagining reduce prejudice

Documents