SciELO - Scientific Electronic Library Online

 número45La lealtad hacia la televisión de pago en tiempos de dificultades económicas en México y en Brasil índice de autoresíndice de materiabúsqueda de artículos
Home Pagelista alfabética de revistas  

Servicios Personalizados




Links relacionados

  • En proceso de indezaciónCitado por Google
  • No hay articulos similaresSimilares en SciELO
  • En proceso de indezaciónSimilares en Google


versión impresa ISSN 0719-3661versión On-line ISSN 0719-367X

Cuad.inf.  no.45 Santiago dic. 2019 


The Joint Effects of Imagined Contact and Similarity with the Protagonist of Testimonial Messages Through Identification and Transportation

Efeitos conjuntos do contato imaginado e a semelhança com o protagonista de mensagens testemunhais através da identificação e o transporte narrativo

Efectos conjuntos del contacto imaginado y la similitud con el protagonista de mensajes testimoniales a través de la identificación y el transporte narrativo

Juan-José Igartua* 

Magdalena Wojcieszak** 

Nuri Kim*** 

*University of Salamanca, Salamanca, Spain (

**University of California, Davis, USA (

***Nanyang Technological University, Singapore, Singapore (


Two experiments carried out in Spain and the Netherlands tested the joint effects of imagined contact and similarity with a narrative protagonist on attitudes and behavioral intentions related to stigmatized immigrants. We advance a concept of optimal reception condition: imagining a positive interaction with an immigrant before reading a testimonial of an immigrant who is similar to the native audience. The optimal reception condition induced greater identification and transportation than the reference condition, leading to more positive attitudes and a higher intention of inter-group contact. The findings are discussed in the context of research on narrative persuasion and prejudice reduction.

Keywords: narrative persuasion; testimonial messages; identification with characters; transportation; character similarity; imagined intergroup contact


Por meio de dois experimentos realizados na Espanha e nos Países Baixos observamos o contraste entre os efeitos conjuntos do contato imaginado e a semelhança com o protagonista de mensagens narrativas nas atitudes e intenção de conduta sobre imigrantes estigmatizados. Avançamos o conceito condição de recepção ótima: imaginar uma interação positiva com um imigrante antes de ler uma mensagem testemunhal de um imigrante que se apresenta como semelhante à audiência. A condição de recepção ótima provocou maior identificação e transporte que a condição de referência, o qual tem produzido atitudes mais positivas e uma maior intenção de contato intergrupal. As descobertas são discutidas no contexto da pesquisa sobre persuasão narrativa e redução do preconceito.

Palavras-chave: persuasão narrativa; mensagens testemunhais; identificação com os personagens; transporte narrativo; semelhança com os personagens; contato intergrupal imaginado


Mediante dos experimentos realizados en España y Países Bajos se contrastaron los efectos conjuntos del contacto imaginado y la similitud con el protagonista de mensajes narrativos en las actitudes e intención de conducta hacia inmigrantes estigmatizados. Avanzamos el concepto condición de recepción óptima: imaginar una interacción positiva con un inmigrante antes de leer un mensaje testimonial de un inmigrante que se presenta como similar a la audiencia. La condición de recepción óptima indujo mayor identificación y transporte que la condición de referencia, lo que llevó a actitudes más positivas y una mayor intención de contacto intergrupal. Los hallazgos se discuten en el contexto de la investigación sobre persuasión narrativa y reducción del prejuicio.

Palabras clave: persuasión narrativa; mensajes testimoniales; identificación con los personajes; transporte narrativo; similitud con los personajes; contacto intergrupal imaginado

“I was born in Guadalajara Mexico. My parents came to the USA when I was 8 years old. I’m now 36 years old and unemployed. Being undocumented has made life so difficult. I became pregnant with twins at age 16; my kids are now 19 and my youngest 13. I have been living in constant fear of deportation and not been able to obtain stability and provide for my younger daughter. I feel like everyone around me is moving up in life and I’m stuck. I literally cry every day because of my immigration status, all my family is here legally except me. I want to do so much with my life but it seems impossible. I stay optimistic but it doesn’t change the fact that I’m undocumented. Life is just so unfair.” (


People often encounter personal narratives, whether in newspaper articles, in magazine testimonial ads, in online blogs, or through social media. Research shows that these narratives have effects on people’s attitudes, beliefs, and behaviors (Braddock & Dillard, 2016), effects that emerge partly through identification with characters and narrative transportation (Tal-Or & Cohen, 2016; Tukachinsky & Tokunaga, 2013). There has been less work done, however, on how to enhance these processes (de Graaf, Sanders, & Hoeken, 2016; Tukachinsky, 2014).

This project adds to this line of inquiry, examining factors that increase identification and transportation and indirectly influence attitudes and behavioral intentions as related to outgroups (i.e., those people excluded from or not belonging to one’s own group). We analyze similarity and imagined intergroup contact and propose that integrating both into a media reception process should generate the strongest effects. We introduce the concept of optimal reception condition, one in which the imagination of a message’s recipient is trained through imagined intergroup contact before reading a personal narrative written by a protagonist presented as very similar to the recipient.

We apply this focus to the context of stigmatized immigrants within the European Union, where countering prejudice toward immigrant groups is an important challenge (Wojcieszak, Azrout, Boomgaarden, Alencar, & Sheets, 2017). We focus on a specific category of narrative messages, short testimonials delivered first-hand by an affected individual (Braverman, 2008; de Wit, Das, & Vet, 2008). Such messages are not only used in public health campaigns (Boeijinga, Hoeken, & Sanders, 2017; Kim & Lee, 2017; Kim & Shapiro, 2016) but are also prevalent in the online environment (Kim & Wojcieszak, 2018; Wojcieszak, Azrout, Boomgaarden, Alencar, & Sheets, 2017; Wojcieszak & Kim, 2016).

Narrative persuasion: identification and transportation

A message can be considered a narrative if it tells a story about a specific event(s), has a clearly identifiable structure, and communicates the ups and downs of characters (Kreuter et al., 2007). Narrative messages can range from simple forms such as personal first-person testimonials shared on social media to audio-visual narratives, such as television shows or full-length feature films.

Research on the effects of narratives identifies two important mechanisms through which persuasion emerges: identification with characters and transportation (Green & Brock, 2000; Moyer-Gusé, 2008; Slater & Rouner, 2002). Identification is a cognitive-affective process that takes place during reception (reading, viewing) of a narrative message and entails perspective-taking or cognitive empathy (putting oneself in the shoes of the character), emotional empathy (feeling the same emotions as the character), internalizing the character’s goal, and the psychological merging of character and audience member (Cohen, 2001; Cohen, 2014: Igartua & Barrios, 2012; Moyer-Gusé, 2008). Identification is “an imaginative experience in which a person surrenders consciousness of his or her own identity and experiences the world through someone else’s point of view” (Cohen, 2001, p. 248). Research finds that identification with a character from a narrative leads to changes in beliefs or opinions (de Graaf, Hoeken, Sanders, & Beentjes, 2012; Hoeken, Kolthoff, & Sanders, 2016; Igartua & Barrios, 2012; Moyer-Gusé, Chung, & Jain, 2011).

In turn, transportation is a psychological process during which the audience focuses its attention on the story, becomes emotionally involved in it, and forms clear and vivid mental images about the story (Green & Brock, 2000). According to the Transportation-Imagery Model, these three processes comprise transportation, and lead to changes in beliefs about the social milieu represented in the narrative. In fact, consistent evidence shows that transportation explains the impact of narrative messages (e.g., Cohen, Tal-Or, & Mazor-Tregerman, 2015; Kim & Shapiro, 2016).

Because transportation and identification facilitate narrative persuasion, it is important to understand how to increase both mechanisms. Research in this field focuses on two aspects: altering narrative messages themselves (e.g., similarity between the audience and the message’s protagonist, narrative voice, etc.) and manipulating exposure conditions (e.g., distractions during message reception) (see, Tukachinsky, 2014). This project focuses on two factors that can enhance both transportation and identification, and -crucially- manipulates both factors in our experimental design.

Similarity in terms of social identity

First, we test similarity between the protagonist of the message and its recipient. Similarity describes a situation in which a message receiver shares certain traits with the protagonist. Similarity can be based on objective traits (e.g., demographics or nationality) and/or psychological or subjective characteristics (e.g., opinions, values, or experiences), both of which may increase identification and transportation and, indirectly, affect attitudes.

Although similarity is often studied, empirical evidence regarding its effect on identification and transportation is inconclusive (Chen, Bell, & Taylor, 2016; Cohen, Weimann-Saks, & Mazor-Tregerman, 2018; de Graaf, 2014; de Graaf et al., 2016; Hoeken et al., 2016; Kim, Shi, & Cappella, 2016). In one review, Tukachinsky (2014) shows that similarity increases transportation and perceived similarity, but not identification. That review, however, only considers the effects of objective, demographic similarity (gender, age, or ethnic group). Moreover, other recent studies clearly demonstrate that similarity has significant effects on identification and attitudes (Chen et al., 2016; Hoeken et al., 2016; Kim et al., 2016; but see Cohen et al., 2018).

These inconsistent results may be due to the fact that similarity is a complex, multi-dimensional construct and that different experiments manipulate similarity differently. Also, and most germane here, the effect of similarity can be intensified by simultaneously altering other elements of the message (Kaufman & Libby, 2012). This implies that it is possible that similarity enhances identification only under certain reception conditions. It is also possible that certain types of similarity are more effective than others in provoking identification and transportation.

In this project, we introduce a new type of similarity: similarity in terms of social identity, namely emphasizing that a protagonist of a testimonial message shares some attitudinal, psychological, and behavioral characteristics that are uniquely relevant to the specific social identity that forms the basis for the in- and outgroup distinction. For instance, a message could present a protagonist’s similarity not only in terms of demographic traits (where you are from), but also in terms of subjective aspects such as national feelings (what group you identify with) or cultural preferences (such as a language or culinary preferences). We consider that similarity in terms of social identity brings to the forefront what people have in common culturally and the feeling of shared social identity.

This should have important effects. Social identity and self-categorization theories (Tajfel, 1982; Turner, 1985) suggest that people tend to categorize those who are similar to them in some ways as ingroup members and those who are different to them in some characteristics as the outgroup. These categorizations lead people to react with more positive affect, think more favorably about, and be more willing to cooperate with the ingroup than the outgroup (Gaertner, Dovidio, Anastasio, Bachman, & Rust, 1993). Relevant here is the notion that in- or outgroup categorization is context dependent, such that -depending on some cues- others can be seen as one’s ingroup or outgroup (Turner, Hogg, Oakes, Reicher, & Wetherell, 1987).

In addition, in the context of prejudice reduction, it is a complex task to foster identification and transportation when messages feature stigmatized immigrants and the audience does not obtain any benefits from changing their attitudes, unlike in health campaigns (Walter, Murphy, & Gillig, 2018). Stigmatized characters induce to less identification by default (Chung & Slater, 2013) and people with negative outgroup attitudes find it more difficult to identify with an outgroup character (i.e. imagining that they are the character and taking their point of view) (Igartua & Frutos, 2017). In this context, we propose that similarity will be particularly effective for stimulating identification and transportation if it is combined with an additional strategy: training the imagination through instructions of imagined intergroup contact before exposure to a personal testimonial.

Imagined intergroup contact and prejudice reduction

Social psychology consistently shows that one of the most effective strategies for reducing prejudice is to facilitate direct contact with outgroup members (Pettigrew & Tropp, 2006). Positive effects of contact can take place even if it occurs indirectly (Harwood, 2010), such as through ingroup friends (Wright, Aron, McLaughlin, & Ropp, 1997) or through mass media (Park, 2012; Wojcieszak & Azrout, 2016). This project focuses on imagined intergroup contact (Crisp & Turner, 2009). Although the relevance of imagined contact to communication and media psychology is being increasingly recognized (Harwood, Joyce, Chen, Paolini, Xiang, & Rubin, 2017), it is not often studied in research on message effects.

Imagined contact is defined as “the mental simulation of a social interaction with a member or members of an outgroup category” (Miles & Crisp, 2013, p. 4). Experiments manipulate imagined contact through instructions that ask participants to think about a positive encounter with an outgroup member (versus imagining an experience unrelated to an intergroup interaction) and then observe the effect of this instruction on attitudes, emotions, and behavioral intentions towards the outgroup (Husnu & Crisp, 2015; Turner, Crisp, & Lambert, 2007). Such mere imagining of a positive encounter with an outgroup member (e.g., an immigrant) is proposed to reduce prejudice.

Research indeed shows that imagined contact not only influences people’s outgroup attitudes, but also “may be valuable as a means of preparing people for future contact” (Miles & Crisp, 2013, p. 4). Imagined contact can make people more receptive to seeking future contact opportunities and prepare them for engaging in such contact with an open mind, lower anxiety, and positive attitude. Furthermore, studies show that using imagined contact as a preparatory stage, or the first warm-up step, for future contact is especially beneficial with respect to stigmatized outgroups (Crisp & Husnu, 2011).

We go beyond testing the direct effect of imagined contact, because -in our view- imagined contact can be conceived as an exposure condition (or a particular psychological state in which people receive a message; Tukachinsky, 2014), facilitating transportation and identification in the context of media exposure. We posit that the combination of imagined intergroup contact and similarity in terms of social identity could be an optimal reception condition for short testimonials about outgroups to effectively counter prejudice. Given that the ability to imagine being the character (identification) and the ability to form mental images of the narrated story (transportation) are two catalysts thorough which exposure to narrative messages influences attitudes, training the imagination before a person reads a short testimonial should increase the effect of the similarity on identification and transportation and, indirectly, on intergroup attitudes and behavioral intentions.

Overview, objectives and hypotheses

To test our theoretical model, according to which the optimal reception condition (imagined contact and high similarity) enhances both identification with the protagonist and transportation and, thereby, influences outgroup attitudes and behavioral intentions (see figure 1), we conducted two experiments in two different contexts (Spain and the Netherlands). Both experiments used short first-person testimonial messages delivered by a stigmatized immigrant of different origins (Moroccan in Spain and Polish in the Netherlands) as stimuli, similar to those used in studies on narrative health communication (Boeijinga et al., 2017; Kim & Shapiro, 2016), narrative news (Kim, Bigman, Leader, Lerman, & Cappella, 2012; Oliver, Dillard, Bae, & Tamul, 2012) or online stories (Kim & Wojcieszak, 2018; Wojcieszak et al., 2017).

Both studies followed the same experimental design that manipulated imagined intergroup contact before exposure to a first-person testimonial written by an immigrant. We define the optimal reception condition as one that trains the imagination of a recipient before she reads a testimonial whose immigrant protagonist is highly similar in terms of social identity. The term optimal reception condition is used to express the idea that under certain exposure conditions (Tukachinsky, 2014) some features of the message (in this case, the similarity with the protagonist) can have a greater effect on reception processes such as identification and narrative transportation. Thus, it is stated that briefly “training” the imagination, immediately before exposition to a testimonial message that presents a protagonist similar to the audience in terms of social identity, constitutes an optimum strategy to induce a greater level of identification and narrative transportation than in a situation that does not present that previous activation of the imagination and delivers a message of low similarity. That is, certain attributes of the message aimed at increasing the similarity with the protagonist will only lead to a higher identification and narrative transportation when people previously experience a favorable state of exposure. In short, the study proposes a main effect of the experimental condition on the mediating processes (identification and narrative transportation) and an indirect effect on the dependent variables. Within this context, we advance the following theoretical expectations:

Hypothesis 1. The optimal reception condition, as opposed to the control condition, will induce greater identification with the protagonist (H1a) and greater transportation (H1b).

Hypothesis 2. The optimal reception condition will induce greater identification and transportation, which in turn will be associated with a more positive attitude towards immigration (H2a) and a higher intention of intergroup contact (H2b).

Source: Own elaboration.

Figure 1. Hypothesized mediation model (H2) Note. Continuous lines show the hypothesized effect of the optimal reception condition (D3) on reception processes (H1) and the indirect effect on attitudes and intention of behavior (H2). 

Experiment 11


Participants and design

The first experiment relied on the Qualtrics panel to recruit a sample of 400 individuals of Spanish origin, whose parents were also Spanish2. Sex and age quotas were established in order for the sample to represent the Spanish population on these characteristics. Fifty percent of the participants were women, and the mean age was 40.41 (SD = 12.16, range: 18-65).

The design consisted of a pre-test questionnaire, experimental manipulation, and post-test questionnaire. The pre-test measured the sociodemographic variables and political self-positioning (from 0 = left to 10 = right, M = 4.35, SD = 2.53). The participants were then randomly assigned to one of four experimental conditions using a 2 x 2 factorial design. First, the participants were randomly assigned to one of two imagined contact conditions: half received instructions of imagined contact and the other half received instructions to imagine a landscape (control group). These instructions appeared on a screen for two minutes. Participants then were randomly assigned to one of two similarity conditions: half read a first-person testimonial written by a Moroccan immigrant who was similar in terms of social identity and half read a nearly identical testimonial whose protagonist was not similar (see below). Later, the post-test measures were administered.

Stimulus materials (independent variable manipulations)

Imagined intergroup contact was manipulated using the standard instructions from prior work (e.g., experiment 2; Husnu & Crisp, 2010). Participants were told: “You will read a short story, in which a person shares his experiences related to living in Spain. Before reading the story, we would like you to spend the next 2 min imagining yourself meeting a Moroccan immigrant for the first time. While imagining this think specifically of when (e.g. next Thursday) and where (e.g. the bus stop) this conversation might occur. Imagine that the interaction is positive, relaxed, and comfortable. It may help to close your eyes while imagining the conversation”. The control group received the following instructions: “You will read a short story, in which a person shares his experiences related to living in Spain. Before reading the story, we would like you to spend the next 2 min imagining an outdoor scene. Try to imagine aspects of the scene about you (e.g., is it a beach, a forest, are there trees, hills, what’s on the horizon). It may help to close your eyes while imagining the outdoor scene”.

The first-person testimonials featuring a Moroccan immigrant sharing his experiences since his arrival in Spain were constructed taking as reference examples published on websites of immigrant support organizations and articles about the experiences of immigrants. The testimonial appeared as unformatted text, without additional features that could introduce uncontrolled variance. A Moroccan immigrant was chosen as the protagonist because Moroccans are one of the most stigmatized immigrant groups in Spain (Cea D’Ancona & Valles, 2014). In the story, Saîd (a popular name in Morocco) talks about different events, circumstances, and feelings related to his arrival in Spain, first jobs, current employment, social and family life, the language, and feelings of belonging. He also mentions that many people think that Moroccans are taking away jobs from Spanish citizens and that there is a link between immigration and criminality, and asks for greater tolerance.

To experimentally manipulate high (506 words) or low (523 words) similarity in terms of social identity, several parts of the message were systematically altered. In the high (low) similarity condition the protagonist felt Spanish (versus Moroccan), had mostly Spanish (Moroccan) friends, mentioned that his favorite food is Spanish (Moroccan), that his children usually speak Spanish (Arabic), that he reads mainly Spanish (Moroccan) newspapers, and that he wishes to stay in Spain (return to Morocco), and identifies with the Spanish (Moroccan) culture. To reinforce the text-based manipulation, the testimonial was accompanied by a photograph of Saîd in his room with a Spanish (Moroccan) flag on a wall.

Two pilot studies were carried out. The first was conducted to select the photograph of the protagonist (N = 21). We chose the photograph that received the highest score on three positive traits (attractive, agreeable, honest) and a low score on “threatening”. The second pilot study tested whether the manipulation of similarity was effective (N = 54). As intended, participants who read the high similarity testimonial reported significantly higher perceived similarity with the immigrant than those who read the low similarity testimonial. Both versions were rated as equally understandable, believable, and interesting.


Perceived similarity. To assure that the manipulation was effective, after reading the narrative participants answered the following questions: “To what extent do you consider you have some things in common with Saîd?” and “How much do you think he is like you?” (1 = not at all, 7 = very much). The two items were averaged (r = .76, p < .001; M = 3.96, SD = 1.69).

Identification with the protagonist. Identification was assessed using an 11-item scale, previously established as reliable and valid (Igartua & Barrios, 2012), that measures degree of identification with a specific character (e.g., “I felt as if I were Saîd”; 1 = not at all, 5 = very much). These items were averaged into a reliable scale (( = .95, M = 2.97, SD = 1.22).

Transportation. Transportation was measured using five items from previous studies (Green & Brock, 2000; Wojcieszak & Kim, 2016), which tapped into the degree to which the participants were absorbed or immersed in the story (e.g., “I was mentally involved in the text while reading it”, from 1 = strongly disagree to 7 = strongly agree). These items were averaged (α = .89, M = 4.58, SD = 0.95).

Attitudes towards immigration. Attitudes towards immigration were evaluated using a 4-item scale, which had already been used in previous studies (e.g., “how strongly do you favor or oppose that the government finances information campaigns about immigrants”; 1 = strongly oppose, 7 = strongly favor). The four items were averaged into a valid and reliable scale (( = .83, M = 4.64, SD = 1.09; Igartua, 2010, Igartua & Cheng, 2009, Igartua, Guerrero-Martín, Cachón-Ramón, & Rodríguez de Dios, 2018).

Behavioral intentions. Participants’ desire to engage in future contact with Moroccan immigrants was assessed using three items (e.g., “Thinking about the next time you find yourself in a situation where you could interact with a Moroccan immigrant -waiting for a bus, with friends in a café, etc.-, how interested would you be in striking up a conversation?” (1 = not at all interested, 7 = very interested). These items were averaged (( = .87; M = 4.45, SD = 1.51; Crisp & Husnu, 2011).


Preliminary analysis

Random assignment to the four conditions was successful. There were no statistically significant differences among the conditions in terms of socio-demographics (gender ((2(3, N = 400) = 1.55, p = .669; age (F(3, 396) = 0.36, p = .780) and political self-positioning (F(3, 386) = 0.61, p = .607).

The manipulation of similarity also worked as intended. An independent sample t-test found that participants who read the testimonial in which the protagonist was similar in terms of social identity considered that Saîd was similar to them (M = 4.47, SD = 1.64) to a significantly greater extent than participants who read the low similarity testimonial (M = 3.35, SD = 1.56; t(398) = -6.98, p < .001).

Joint effect of imagined contact and similarity on identification and transportation (H1)

We expected that participants would experience greater identification with the protagonist (H1a) and greater transportation (H1b) in the optimal reception condition as compared to the reference condition. To test this hypothesis, two analyses of variance (ANOVA) were run with the experimental condition as the independent variable and identification and transportation as the dependent variables (see Walter et al., 2018, for a similar analytical approach in their Study 3, with a 2 x 2 factorial between-subjects design). A planned contrast analysis (-1, 0, 0, 1) was also carried out, which compared condition 4 or the optimal reception condition (i.e., imagined contact and high similarity, coefficient 1) to the reference condition 1 (i.e., no imagined contact and low similarity, coefficient -1).

There were statistically significant differences in identification (F(3, 396) = 12.19, p < .001, (p 2 = 0.085) and transportation (F(3, 396) = 9.34, p < .001, (p 2 = 0.066) as a function of experimental condition. Moreover, both planned contrasts were statistically significant (identification: t(396) = 4.68, p < .001, r = .22; transportation: t(396) = 3.55, p < .001, r = .17). Participants in the optimal reception condition showed higher identification (Mdif = 0.59, SE = 0.12) and transportation (Mdif = 0.66, SE = 0.18) than those in the reference condition. A second planned contrast analysis compared the average of all three conditions to the optimal condition (contrast codes: -1 = control, -1 = similarity only, -1 = imagined contact only, 3 = optimal condition) and similar results were found for identification (t(396) = 3.31, p < .001, r = .16) and transportation (t(396) = 2.89, p < .01, r = .14). H1 is therefore supported.

Mediation models (H2)

The second hypothesis predicted that the effect of the optimal reception condition on attitudes (H2a) and intended intergroup contact (H2b) would be mediated by identification and transportation. To examine whether the joint effect of imagined contact and similarity had a significant indirect effect on the dependent variables, we used the PROCESS macro for SPSS (Model 4; 10,000 bootstrapped samples with 95% bias-corrected confidence intervals; Hayes, 2013) set to a multicategorical model, treating the control condition (no imagined contact and low similarity), as a reference category (for a similar analytical approach, see: Bolkan, Goodboy, & Myers, 2017; Walter et al., 2018).

The multicategorical independent variable was entered into a mediation model to quantify the relative indirect effects of being in one experimental group (condition 2, D1: no imagined contact and high similarity; condition 3, D2: imagined contact and low similarity; condition 4, D3: imagined contact and high similarity) compared with the reference group (condition 1: no imagined contact and low similarity). This procedure required the estimation of three dummy codes (D1, D2, D3) acting separately as the independent variable and allowing the remaining two dummy codes to serve as covariates (Hayes & Preacher, 2014). This model estimated the relative indirect effect of the optimal reception condition (D3) on attitudes and intended intergroup contact, as mediated by identification and transportation.

Regarding attitudes toward immigration, the optimal reception condition (D3), relative to the reference condition, significantly increased identification (a 31 = .59, SE = .12, p < .001) and transportation (a 32 = .66, SE = .18, p < .001). In addition, identification (b 1 = .61, SE = .10, p < .001) and transportation (b 2 = .23, SE = .07, p < .001) significantly predicted attitudes toward immigration (see figure 2a). The relative indirect effects of the optimal reception condition through identification (a 31 b 1 = .36, SE = .10, 95% CI [.18, .61]) and transportation (a 32 b 2 = .15, SE = .07, 95% CI [.05, .33]) were therefore significant. Although not hypothesized, compared to the reference condition, the similarity condition (D1) produced a significant relative indirect effect on attitudes toward immigration through identification (a 11 b 1 = .38, SE = .10, 95% CI [.20, .63]) and transportation (a 12 b 2 = .16, SE = .07, 95% CI [.05, .35]). Imagined contact (D2) did not produce any significant relative indirect effects on attitudes, neither through identification (a 21 b 1 = .08, SE = .08, 95% CI [-.06, .25]) nor transportation (a 22 b 2 = -.01, SE = .05, 95% CI [-.12, .08]).

Similar results emerged for intended contact with Moroccan immigrants (see figure 2b). The relative indirect effects of the optimal reception condition (D3) through identification (a 31 b 1 = .30, SE = .09, 95% CI [.14, .52]) and transportation (a 32 b 2 = .27, SE = .09, 95% CI [.11, .51]) were statistically significant. Again, compared with the reference condition, similarity (D1) produced a significant relative indirect effect on intended intergroup contact (identification: a 11 b 1 = .31, SE = .09, 95% CI [.15, .54]; transportation: a 12 b 2 = .29, SE = .09, 95% CI [.12, .51]). Imagined contact (D2) did not produce any significant relative indirect effects on behavioral intentions (identification: a 21 b 1 = .06, SE = .05, 95% CI [-.05, .21]; transportation: a 22 b 2 = -.02, SE = .08, 95% CI [-.18, .14]). This consistent pattern supports H2.

Source: Own elaboration.

Figure 2 Results of the mediation analysis (Study 1, Spain, N = 400): relative indirect effect of the optimal reception condition (imagined contact and high similarity narrative, D3) on attitudes towards immigration (a) and behavioral intentions (b) through identification and transportation. Note. The Figure shows the non-standardized regression coefficients, B. The coefficients of the relative direct effects appear in parentheses. The dashed line represents nonsignificant coefficients. + p < .10, * p < .05, ** p < .01, *** p < .001. 


Study 1 confirms our theoretical model. The optimal reception condition induced greater identification and transportation (compared to the reference condition), which had predicted effects on the two dependent variables. However, these results are based on a study using only one testimonial message. Additionally, the testimonial featured a Moroccan immigrant, so the effects may be distorted by the relationship between Moroccan immigrants and Islam (the dominant religion in Morocco at 98%) and, therefore, with Islamophobia (Saleem, Yang, & Ramasubramanian, 2016).

Experiment 2: A Replication

To address these issues, replicate the results, and validate the proposed model we used a similar design in the Netherlands, a distinct sociopolitical context, with a testimonial featuring an immigrant from Poland. Polish immigrants are a large and typically disliked group in the Netherlands, and Poland is a country with a profound Catholic tradition and only a minor Muslim presence at less than 1%.


Participants and design

A sample of 392 people of Dutch origin whose parents were also Dutch were recruited from a Qualtrics panel. Quotas were set for gender and age in order for the sample to reflect the Dutch population. Thus, 50.3% of the participants were women and the mean age was 40.96 (SD = 11.11, range: 18-65).

A 2 x 2 factor design was used and imagined contact and similarity were manipulated in the same way as in Study 1. We again took into account different aspects of social identity relevant to immigrant groups (e.g., feeling more Dutch than Polish), and -to reinforce the similarity manipulation- accompanied the narrative by a photograph of the protagonist (Leszek) near a Dutch (versus Polish) flag.


The measures from Study 1 were adapted to Polish immigration. The pre-test measured the sociodemographic variables and political self-positioning (from 0 = left to 10 = right; M = 5.47, SD = 2.20). Immediately after exposure to the testimonial, the post-test measured perceived similarity (M = 3.51, SD = 1.61), identification with the protagonist (( = .94, M = 2.64, SD = 0.93), transportation (α = .88, M = 4.19, SD = 1.33), attitudes toward immigration (( = .85, M = 4.12, SD = 1.33), and intention to engage in contact with Polish immigrants (( = .85; M = 4.04, SD = 1.37).


Random assignment to experimental conditions was successful: no differences were observed in terms of socio-demographics or political self-positioning. Also, the manipulation of similarity was effective (t(390) = -3.40, p < .001).

Hypothesis 1 posited that participants would feel greater identification with the protagonist (H1a) and transportation (H1b) in the optimal message reception condition, as compared to the reference condition. Analysis of variance (ANOVA) found statistically significant differences in identification (F(3, 388) = 3.31, p < .05, (p 2 = 0.025), and the planned contrast were also statistically significant (t(388) = 2.60, p < .01, r = .13). Participants in the optimal reception condition showed greater identification (Mdif = 0.34, SE = 0.13) than the reference condition. The effect of that condition on transportation was marginally significant (F(3, 388) = 2.37, p = .070, (p 2 = 0.018) and the planned contrast also yielded marginally significant results (Mdif = 0.35, SE = 0.18; t(388) = 1.88, p = .060, r = .09). Like in Study 1, a second planned contrast analysis compared the average of all three conditions to the optimal condition, obtaining similar results for identification (t(388) = 2.85, p < .01, r = .14) and a statistically significant effect on transportation (t(388) = 2.06, p < .05, r = .10). Hence, H1a is supported, and H1b partially supported.

H2 predicted that the effects of the optimal reception condition on attitudes towards immigration (H2a) and intended intergroup contact (H2b) would be mediated by identification and transportation. The PROCESS macro was again used to test this hypothesis. Regarding attitudes towards immigration (see figure 3a), the optimal reception condition (D3) significantly increased identification (a 31 = .34, SE = .13, p < .01) relative to the reference condition. However, identification was not associated with attitudes toward immigration (b 1 = -.02, SE = .10, p = .801), yielding an insignificant indirect effect (a 31 b 1 = -.009, SE = .04, 95% CI [-.10, .06]). The optimal reception condition, relative to the reference condition, increased transportation only marginally (a 32 = .35, SE = .18, p = .060). Because transportation was a significant predictor of attitudes toward immigration (b 2 = .60, SE = .07, p < .001), the relative indirect effect through transportation was marginally significant (a 32 b 2 = .21, SE = .12, 90% CI [.03, .43]). With respect to intended contact with Polish immigrants (see figure 3b), the relative indirect effect of the optimal reception condition through identification was statistically significant (a 31 b 1 = .11, SE = .06, 95% CI [.02, .27]) and marginally significant through transportation (a 32 b 2 = .16, SE = .09, 90% CI [.02, .33]). These results partially confirm H2.

Unlike in Study 1, D1 (similarity) did not have a significant relative indirect effect through identification or transportation on attitudes and intended intergroup contact. Similarly to Study 1, D2 (imagined contact) did not produce any significant relative indirect effects on the two tested outcomes.

Source: Own elaboration.

Figure 3 Results of the mediation analysis (Study 2, Netherlands, N = 392): relative indirect effect of the optimal reception condition (imagined contact and high similarity narrative, D3) on attitudes towards immigration (a) and behavioral intentions (b) through identification and transportation Note. The Figure shows the non-standardized regression coefficients, B. The coefficients of the relative direct effects appear in parentheses. The dashed line represents nonsignificant coefficients. + p < .10, ** p < .01, *** p < .001. 


The results of Study 2 are generally consistent with those obtained in Study 1. Working with a different immigrant group in a different European context, we again found that the optimal reception condition stimulates identification with the protagonist of the story and -to some extent- transportation into the message, which in turn induce intention to engage in intergroup contact.

General Discussion And Conclusions

An important question in much research on narrative persuasion is how to enhance identification of the audiences with message protagonists, and also how to increase transportation -or immersion- of the audience into the message itself. This question also presents a practical challenge for those interested in applying narratives in general and short testimonials in particular to pressing social problems, ranging from health to intergroup hostilities. Inasmuch as identification and transportation facilitate those effects (Tal-Or & Cohen, 2016; Tukachinsky & Tokunaga, 2013), addressing this question is of theoretical and social importance.

This project extended prior work on the factors increasing identification and transportation. Our theoretical model addressed both imagined intergroup contact -manipulated prior to reading a testimonial- as well as similarity of the protagonist to the audience. We proposed a notion of the optimal reception condition, according to which imagining a positive interaction with an outgroup protagonist before reading a testimonial written by an outgroup member similar to the audience in terms of social identity should be especially effective in enhancing identification and transportation, thereby generating more favorable outgroup attitudes and intentions to engage in intergroup contact.

The results from two original experiments provide support for the proposed model. Both the study carried out in Spain (with a Moroccan immigrant as the protagonist) and in the Netherlands (with a Polish immigrant) yielded similar results. Combining instructions of imagined contact with manipulation of similarity indeed made it possible to increase identification and transportation. Hence, we show that varying the conditions of exposure can facilitate the effect of similarity in the reception processes. This result is consistent with and extends previous research, which documents that imagined intergroup contact activates one’s imagination and constitutes a cognitive training for a future encounter with an outgroup member (Crisp & Husnu, 2011). Because both identification and transportation are processes linked to imagination (imagining being the character or the details of the story), it is logical to think that such a training will influence both processes when the participants read a testimonial with a protagonist similar to themselves.

At this point, it is necessary to emphasize that imagined contact manipulation did not seek to increase empathy toward the protagonist of the narrative (an affective process), but to “train” the imagination of participants (a cognitive process), in order to facilitate the processes of narrative transportation and identification with a protagonist who showed a high similarity with the audience. There is a whole line of research, initiated by Batson and collaborators (e.g., Batson et al., 1997; Tarrant & Hadert, 2010; Wojcieszak & Kim, 2016), which has analyzed the influence of empathy on intergroup attitudes towards stigmatized groups. It is quite possible that this may be a promising area of research for future studies. Given that identification is a construct linked to empathy, future studies could test the influence of empathy (as a differentiated exposure condition) on the identification process.

Although previous work on the effect of similarity on identification and transportation has yielded mixed results, Study 1 found that similarity in terms of social identity enhanced both mediators (a non-hypothesized effect). This finding, consistent with the latest studies (Chen et al., 2016; Hoeken et al., 2016; Kaufman & Libby, 2012), underscores the potential of similarity in terms of social identity, a proposed concept. To manipulate demographic similarity it is sufficient to give information on gender or nationality of the protagonist just once in the narrative, attributes that may be overlooked by the audience (Igartua & Fiuza, 2018). In contrast, manipulating similarity in terms of social identity entails including information about different factors (feelings of belonging, language, favorite food, etc.) throughout the text. This may make the character’s social identity more difficult to overlook and potentially more effective than demographic similarity (see, Kim et al., 2016).

The crucial and novel addition of our study regards the proposition that the optimal reception condition increases identification and transportation and this, in turn, generates more positive attitudes and behavioral intentions. Across both studies, identification and transportation were the processes explaining the effects of the optimal reception condition on all the outcomes tested. Transportation played a weaker role in Study 2, perhaps because, over all, Dutch participants experienced less transportation than Spaniards. A second explanation of this result is that testimonial messages (which present a single protagonist) may contain a lower level of suspense than a more complex narrative with several characters and, therefore, induce less narrative transportation (Cohen & Tal-Or, 2017; Tal-Or & Cohen, 2010). Finally, perhaps in the context of testimonial messages, where there is a clear protagonist that captures the audience’s attention, identification could be a more relevant process or explanatory mechanism than narrative transportation (see, Tal-Or & Cohen, 2016). In any case, previous research has demonstrated that identification and transportation underlie and explain narrative persuasion processes (e.g., Walter et al., 2018).

Some limitations need to be kept in mind when interpreting the results. We cannot be confident that the participants followed our instructions and used the allotted time to imagine a positive interaction with an immigrant. Similar to most work in the area (e.g., Husnu & Crisp, 2010), we did not include manipulation checks of imagined contact, and so the limited effects from imagined contact may be due to the lack of compliance on the part of our subjects.

A second limitation refers to the fact that both studies had only a single message per condition, which means it cannot be ruled out that there is something unique about that message that is causing the observed effects. In this sense, Slater, Peter and Valkenburg (2015) have pointed out that “using a single message as a base stimulus for manipulation inherently leaves a greater room for the possibility that study results depend on idiosyncrasies of the message studied and how the manipulation is carried out” (p. 23). Therefore, in future studies repeated measures designs or mixed designs that contemplate the use of several messages for each experimental condition should be employed in order to obtain more robust conclusions. Although this practice is not usual in research on narrative persuasion, in recent years experimental studies have been developed that address the problem of “message variability”, combining the use of repeated measurement designs (that is, each participant is exposed to different messages of the same class during the same experimental session) with multilevel modeling analysis (e.g., Kim, 2019; Schreiner, Appel, Isberner, & Richter, 2018; Shen, 2011, 2018). However, this experimental procedure also has its own limitations. In the first place, there is no concrete guideline on the number of messages needed for each condition so that the study has adequate external validity. In the experiment developed by Kim (2019) only two messages were used, and in the studies carried out by Schreiner et al. (2018) and Shen (2018) four or five messages were used, respectively. In fact, Shen (2011) recognized as a limitation of his study that “(…) only four messages within each type were used in this study, and the four messages were presented in a row, instead of within some form of regular television programming. This means that the external validity could be rather limited” (p. 411). Secondly, it is worth considering to what extent it is realistic to expose participants to four or five persuasive narratives (in a row) on the same topic and evaluate the attitudinal measures at the end of the message exposure. This is a problem of ecological validity, as it has been pointed out by Shen (2018) as a limitation of his work: “(…) each participant was assigned to watch five stimuli messages in a row. (…) This is not how individuals typically encounter advertisement messages in real life” (p. 7).

While it is true that research in Media Psychology has emphasized that it is necessary to use more than one message per experimental condition, this practice is still very infrequent. Reeves, Yeykelis and Cummings (2016) reviewed more than 300 experiments conducted in the last 10 years in this field and concluded that “the majority used only a single stimulus per condition (64.4%)” (p. 57). The use of more than one message per condition is common in research focused on the study of cognitive and affective processing of media messages, particularly when psycho-physiological measures are used (see, Reeves & Geiger, 1994). However, this practice is not usual in narrative persuasion research (Braddock & Dillard, 2016; Tukachinsky, 2014). In fact, the main reason for conducting study 2 was precisely to verify that the independent variables considered exerted the effects found in study 1, slightly varying some features of the message (e.g., the protagonist of the message belongs to a different immigrant group) and the context (the second study was carried out in another European country). In that sense, we used an approximation based on replication with minimal variations or “close replication” (Lindsay & Ehrenberg, 1993). This is a common practice in experimental studies in Communication Science (Boyle & Schmierbach, 2015, p. 225) and, in particular, in narrative persuasion research (e.g., Walter & Cohen, 2019).

Also, we only considered the effects of written messages because those are frequent in health campaigns, websites of social organizations or on social media, and their consumption is made easier by mobile devices. Furthermore, relative to longer video narratives, such as movies or entertainment shows, it is more feasible to systematically manipulate such messages for the purposes of experimental investigation. Nevertheless, visual or audio narratives may have distinct effects on the tested processes. We suspect that these effects would be yet stronger - inasmuch as audiovisual narratives may generate greater identification and transportation (Walter, Murphy, Frank, & Baezconde-Garbanati, 2017). If so, the consistent results presented can offer a conservative estimate of the effects of personal testimonials (and especially of the optimal reception condition).

In a similar vein, our research focused only on one type of stigmatized group: immigrants in two European countries. This focus was guided by the fact that immigration has become a central item on the political agendas of the EU (Meleady, Seger, & Vermue, 2017) and public attitudes toward immigrants in Europe are rather negative (e.g., Krogstad, 2015). We encourage future research to examine whether the effects differ for other stigmatized groups (e.g., the homeless, the HIV positive) and for less disliked immigrants (e.g., Germans in Spain). Testing these potentially differential effects would help to establish the boundary conditions for similarity and imagined contact work in the context of messages about or delivered by various outgroups.

Despite these limitations, our work improves our understanding of the mechanisms that explain the impact of narrative stories. The proposed concept of optimal reception condition facilitates identification and transportation, improves outgroup attitudes, and stimulates intended intergroup contact. Theoretically, these results do not only advance a new concept in the narrative persuasion literature, but they also help to clarify the relatively inconclusive evidence regarding the effects of similarity in past work. These results also have concrete practical implications, suggesting that campaign appeals or public service announcements should attempt to simultaneously encourage recipients to generate certain mental states prior to exposure to messages featuring outgroup members (e.g., by including a simple statement such as “imagine we are having a good time together”) and to present the outgroups as sharing some similarities with the audience members (e.g., emphasizing that everyone has a family). How to do this in practice will be an important challenge for practitioners and scholars interested in minimizing intergroup hostilities. Offering some tools (as we did in our project) and showing that they are effective provides a first step in addressing this challenge.


Batson, C. D., Polycarpou, M. P., Harmon-Jones, E., Imhoff, H. J., Mitchener, E. C., Bednar, L. L., & Highberger, L. (1997). Empathy and attitudes: can feeling for a member of a stigmatized group improve feelings toward the group? Journal of Personality and Social Psychology, 72(1), 105-118. ]

Boeijinga, A., Hoeken, H., & Sanders, J. (2017). Risk versus planning health narratives targeting Dutch truck drivers: obtaining impact via different routes? International Journal of Communication, 11, 5007-5026. Retrieved from ]

Bolkan, S., Goodboy, A. K., & Myers, S. A. (2017). Conditional processes of effective instructor communication and increases in students’ cognitive learning . Communication Education, 66(2), 129-147. ]

Boyle, M. P. & Schmierbach, M. (2015). Applied communication research methods. New York, NY: Routledge. [ Links ]

Braddock, K. & Dillard, J. P. (2016). Meta-analytic evidence for the persuasive effect of narratives on beliefs, attitudes, intentions, and behaviors. Communication Monographs, 83(4), 446-467. ]

Braverman, J. (2008). Testimonials versus informational persuasive messages: the moderating effect of delivery mode and personal involvement. Communication Research, 35(5), 666-694. ]

Cea D’Ancona, M. A., & Valles, M. S. (2014). Evolución del racismo, la xenofobia y otras formas conexas de intolerancia en España (Informe-Encuesta 2014) (Evolution of racism, xenophobia and other related intolerance in Spain (Report-Survey 2014)). Madrid: Ministerio de Empleo y Seguridad Social. [ Links ]

Chen, M., Bell, R. A., & Taylor, L. D. (2016). Narrator point of view and persuasion in health narratives: the role of protagonist-reader similarity, identification, and self-referencing. Journal of Health Communication, 21(8), 908-918. ]

Chung, A. H. & Slater, M. D. (2013). Reducing stigma and out-group distinctions through perspective-taking in narratives. Journal of Communication, 63(5), 894-911. ]

Cohen, J. (2001). Defining identification: a theoretical look at the identification of audiences with media characters. Mass Communication & Society, 4(3), 245-264. ]

Cohen, J. (2014). Mediated relationships and social life: current research on fandom, parasocial relationships, and identification. In M. B. Oliver & A. A. Raney (Eds.), Media and social life (pp. 142-156). New York, NY: Routledge . [ Links ]

Cohen, J. & Tal-Or, N. (2017). Antecedents of identification: character, text, and audiences. In F. Hakemulder, M. M. Kuijpers, E. S. Tan, K. Bálint, & M. M. Doicaru (Eds.), Narrative absorption (pp. 133-153). Amsterdam: John Benjamins Publishing Company. [ Links ]

Cohen, J., Tal-Or, N., & Mazor-Tregerman, M. (2015). The tempering effect of transportation: Exploring the effects of transportation and identification during exposure to controversial two-sided narratives. Journal of Communication, 65(2), 237-258. ]

Cohen, J., Weimann-Saks, D., & Mazor-Tregerman, M. (2018). Does character similarity increase identification and persuasion? Media Psychology, 21(3), 506-528. ]

Crisp, R. J. & Husnu, S. (2011). Attributional processes underlying imagined contact effects. Group Processes & Intergroup Relations, 14(2), 275-287. ]

Crisp, R. J. & Turner, R. N. (2009). Can imagined interactions produce positive perceptions? Reducing prejudice through simulated social contact. The American Psychologist, 64(4), 231-240. ]

de Graaf, A. (2014). The effectiveness of adaptation of the protagonist in narrative impact: similarity influences health beliefs through self-referencing. Human Communication Research, 40(1), 73-90. ]

de Graaf, A., Hoeken, H., Sanders, J., & Beentjes, J. W. J. (2012). Identification as a mechanism of narrative persuasion. Communication Research, 39(6), 802-823. ]

de Graaf, A., Sanders, J., & Hoeken, H. (2016). Characteristics of narrative interventions and health effects: a review of the content, form, and context of narratives in health-related narrative persuasion research. Review of Communication Research, 4, 88-131. ]

De Wit, J. B., Das, E., & Vet, R. (2008). What works best: objective statistics or a personal testimonial? An assessment of the persuasive effects of different types of message evidence on risk perception. Health Psychology, 27(1), 110-115. ]

Faul, F., Erdfelder, E., Buchner, A., & Lang, A.-G. (2009). Statistical power analyses using G*Power 3.1: tests for correlation and regression analyses. Behavior Research Methods, 41(4), 1149-1160. ]

Gaertner, S. L., Dovidio, J. F., Anastasio, P. A., Bachman, B. A., & Rust, M. C. (1993). The common ingroup identity model. Recategorization and the reduction of intergroup bias. European Review of Social Psychology, 4(1), 1-26. ]

Green, M. C. & Brock, T. C. (2000). The role of transportation in the persuasiveness of public narratives. Journal of Personality and Social Psychology, 79(5), 701-721. ]

Harwood, J. (2010). The contact space: a novel framework for intergroup contact research. Journal of Language and Social Psychology, 29(2), 147-177. ]

Harwood, J., Joyce, N., Chen, C.-Y., Paolini, S., Xiang, J., & Rubin, M. (2017). Effects of past and present intergroup communication on perceived fit of an outgroup member and desire for future intergroup contact. Communication Research, 44(4), 530-555. ]

Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process analysis. New York, NY: The Guilford Press. [ Links ]

Hayes, A. F. & Preacher, K. J. (2014). Statistical mediation analysis with a multicategorical independent variable. British Journal of Mathematical and Statistical Psychology, 67, 451-470. ]

Hoeken, H., Kolthoff, M., & Sanders, J. (2016). Story perspective and character similarity as drivers of identification and narrative persuasion. Human Communication Research, 42(2), 292-311. ]

Husnu, S. & Crisp, R. J. (2010). Elaboration enhances the imagined contact effect. Journal of Experimental Social Psychology, 46(6), 943-950. ]

Husnu, S. & Crisp, R. J. (2015). Perspective-taking mediates the imagined contact effect. International Journal of Intercultural Relations, 44, 29-34. ]

Igartua, J. J. (2010). Identification with characters and narrative persuasion through fictional feature films. Communications. The European Journal of Communication Research, 35(4), 347-373. ]

Igartua, J. J. & Barrios, I. (2012). Changing real-world beliefs with controversial movies: processes and mechanisms of narrative persuasion. Journal of Communication, 62(3), 514-531. ]

Igartua, J. J. & Cheng, L. (2009). Moderating effect of group cue while processing news on immigration. Is framing effect a heuristic process? Journal of Communication, 59(4), 726-749. ]

Igartua, J. J. & Fiuza, D. (2018). Persuading with narratives against gender violence. Effect of similarity with the protagonist on identification and risk-perception. Palabra Clave, 21(2), 499-523. ]

Igartua, J. J. & Frutos, F. J. (2017). Enhancing attitudes toward stigmatized groups with movies: Mediating and moderating processes of narrative persuasion. International Journal of Communication, 11, 158-177. Retrieved from ]

Igartua, J. J., Guerrero-Martín, I., Cachón-Ramón, D., & Rodríguez de Dios, I. (2018). Efecto de la similitud con el protagonista de narraciones contra el racismo en las actitudes hacia la inmigración. El rol mediador de la identificación con el protagonista (Effect of Similarity with the Protagonist of Narratives Against Racism in Attitudes Toward Immigration. The Mediating Role of Identification with the Protagonist). Disertaciones. Anuario Electrónico de Estudios de Comunicación Social, 11(1), 56-75. ]

Kaufman, G. F. & Libby, L. K. (2012). Changing beliefs and behavior through experience-taking. Journal of Personality and Social Psychology, 103(1), 1-19. ]

Kim, M. (2019). When similarity strikes back: conditional persuasive effects of character-audience similarity in anti-smoking campaign. Human Communication Research, 45(1), 52-77. ]

Kim, H. S., Bigman, C. A., Leader, A. E., Lerman, C., & Cappella, J. N. (2012). Narrative health communication and behavior change: the influence of exemplars in the news on intention to quit smoking. Journal of Communication, 62(3), 473-492. ]

Kim, H. K. & Lee, T. K. (2017). Conditional effects of gain-loss-framed narratives among current smokers at different stages of change. Journal of Health Communication, 22(12), 990-998. ]

Kim, H. K. & Shapiro, M. A. (2016). When bad things happen to a protagonist like you: the role of self in resistance to negatively framed health narratives. Journal of Health Communication, 21(12), 1227-1235. ]

Kim, M., Shi, R., & Cappella, J. N. (2016). Effect of character-audience similarity on the perceived effectiveness of antismoking PSAs via engagement. Health Communication, 31(10), 1193-1204. ]

Kim, N. & Wojcieszak, M. (2018). Intergroup contact through online comments: effects of direct and extended contact on outgroup attitudes. Computers and Human Behavior, 81, 63-72. ]

Krogstad, J. M. (2015, September 24). What Americans, Europeans think of immigrants. Pew Research Center. Retrieved from https://www.pewresearch.orgLinks ]

Lindsay, R. M. & Ehrenberg, A. S. (1993). The design of replicated studies. The American Statistician, 47(3), 217-228. ]

Kreuter, M. W., Green, M. C., Cappella, J. N., Slater, M. D., Wise, M. E., Storey, D., …Woolley, S. (2007). Narrative communication in cancer prevention and control: a framework to guide research and application. Annals of Behavioral Medicine, 33(3), 221-235. ]

Meleady, R., Seger, C. R., & Vermue, M. (2017). Examining the role of positive and negative intergroup contact and anti-immigrant prejudice in Brexit. British Journal of Social Psychology, 56(4), 799-808. ]

Miles, E. & Crisp, R. J. (2013). A meta-analytic test of the imagined contact hypothesis. Group Processes & Intergroup Relations, 17(1), 3-26. ]

Moyer-Gusé, E. (2008). Toward a theory of entertainment persuasion: explaining the persuasive effects of entertainment-education messages. Communication Theory, 18(3), 407-425. ]

Moyer-Gusé, 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. ]

Oliver, M. B., Dillard, J. P., Bae, K., & Tamul, D. J. (2012). The effect of narrative news format on empathy for stigmatized groups. Journalism & Mass Communication Quarterly, 89(2), 205-224. ]

Park, S.-Y. (2012). Mediated intergroup contact: concept explication, synthesis, and application. Mass Communication and Society, 15(1), 136-159. ]

Pettigrew, T. F. & Tropp, L. R. (2006). A meta-analytic test of intergroup contact theory. Journal of Personality and Social Psychology, 90(5), 751-783. ]

Reeves, B. & Geiger, S. (1994). Designing experiments that assess psychological responses to media messages. In A. Lang (Ed.), Measuring psychological responses to media messages (pp. 165-180). Hillsdale, NJ: Lawrence Erlbaum Associates. [ Links ]

Reeves, B., Yeykelis, L., & Cummings, J. J. (2016). The use of media in media psychology. Media Psychology, 19(1), 49-71. ]

Saleem, M., Yang, G. S., & Ramasubramanian, S. (2016). Reliance on direct and mediated contact and public policies supporting outgroup harm. Journal of Communication, 66(4), 604-624. ]

Schreiner, C., Appel, M., Isberner, M. B., & Richter, T. (2018). Argument strength and the persuasiveness of stories. Discourse Processes, 55(4), 371-386. ]

Shen, L. (2011). The effectiveness of empathy-versus fear-arousing antismoking PSAs. Health Communication, 26(5), 404-415. ]

Shen, L. (2018). Features of empathy-arousing strategic messages. Health Communication, 34(11), 1329-1339. ]

Slater, M., Peter, J., & Valkenburg, P. M. (2015). Message variability and heterogeneity: a core challenge for communication research. Annals of the International Communication Association, 39(1), 3-31. ]

Slater, M. D. & Rouner, D. (2002). Entertainment-education and elaboration likelihood: understanding the processing of narrative persuasion. Communication Theory, 12(2), 173-191. ]

Tajfel, H. (1982). Social identity and intergroup relations. Cambridge: Cambridge University Press. [ Links ]

Tal-Or, N. & Cohen, J. (2010). Understanding audience involvement: conceptualizing and manipulating identification and transportation. Poetics, 38(4), 402-418. ]

Tal-Or, N. & Cohen, J. (2016). Unpacking engagement: convergence and divergence in transportation and identification. Annals of the International Communication Association, 40(1), 33-66. ]

Tarrant, M. & Hadert, A. (2010). Empathic experience and attitudes toward stigmatized groups: evidence for attitude generalization. Journal of Applied Social Psychology, 40(7), 1635-1656. ]

Tukachinsky, R. (2014). Experimental manipulation of psychological involvement with media. Communication Methods and Measures, 8(1), 1-33. ]

Tukachinsky, R. & Tokunaga, R. S. (2013). The effects of engagement with entertainment. Communication Yearbook, 37(1), 287-321. ]

Turner, J. C. (1985). Social categorization and the self-concept: A social cognitive theory of group behavior. In E. J. Lawler (Ed.), Advances in group processes: Theory and research (Vol. 2) (pp. 77-121). Greenwich, CT: JAI Press. [ Links ]

Turner, J. C., Hogg, M. A., Oakes, P. J., Reicher, S. D., & Wetherell, M. S. (1987). Rediscovering the social group: A self-categorization theory. Cambridge, MA: Basil Blackwell. [ Links ]

Turner, R. N., Crisp, R. J., & Lambert, E. (2007). Imagining intergroup contact can improve intergroup attitudes. Group Processes & Intergroup Relations, 10(4), 427-441. ]

Walter, N. & Cohen, J. (2019). When less is more and more is less: the paradoxical metacognitive effects of counterarguing. Communication Monographs, 86(3), 377-397. ]

Walter, N., Murphy, S. T., Frank, L. B., & Baezconde-Garbanati, L. (2017). Each medium tells a different story: the effect of message channel on narrative persuasion. Communication Research Reports, 34(2), 161-170. ]

Walter, N., Murphy, S. T., & Gillig, T. K. (2018). To walk a mile in someone else’ s shoes: how narratives can change causal attribution through perceived interactivity. Human Communication Research, 44(1), 31-57. ]

Wojcieszak, M. & Azrout, R. (2016). I saw you in the news: mediated and direct intergroup contact improve outgroup attitudes. Journal of Communication, 66(6), 1032-1060. ]

Wojcieszak, M., Azrout, R., Boomgaarden, H., Alencar, A. P., & Sheets, P. (2017). Integrating Muslim immigrant minorities: the effects of narrative and statistical messages. Communication Research, 44(4), 582-607. ]

Wojcieszak, M. & Kim, N. (2016). How to improve attitudes toward disliked groups: the effects of narrative versus numerical evidence on political persuasion. Communication Research, 43(6), 785-809. ]

Wright, S. C., Aron, A., McLaughlin, T., & Ropp, S. A. (1997). The extended contact effect: knowledge of cross-group friendships and prejudice. Journal of Personality and Social Psychology, 73(1), 73-90. Retrieved from ]

1All the materials (datasets and questionnaires) related to Experiment 1 and Experiment 2 are available in Open Science Framework (

2To determine the sample size, a previous analysis was carried out with the G*Power program (Faul, Erdfelder, Buchner, & Lang, 2009). The calculation of the sample size depends on several factors such as the type of design, the effect size observed in previous studies (or in meta-analysis reviews), type I error (() and statistical power (1 - (). Braddock & Dillard (2016)’s meta-analysis study was taken into account when obtaining a measure of effect size. Thus, assuming an effect size of .17, a ( value of .05, a power of .80, and a four- groups design, G*Power program indicated that a sample size of 384 participants would be necessary. For that reason, both studies were designed to include a sample of this size or slightly higher.

Received: November 19, 2018; Accepted: August 07, 2019

Juan-José Igartua (Ph.D., University of the Basque Country, 1996) is a Professor of Media Psychology at the Faculty of Social Sciences at the University of Salamanca in Spain. In addition, from 2006, he is the director of the Observatorio de los Contenidos Audiovisuales (, a Recognized Research Group at the University of Salamanca. His research interests focus on the following topics: processes and media effects, narrative persuasion, media entertainment, health communication, news framing, and media, prejudice and immigration.

Magdalena Wojcieszak (Ph.D. Annenberg School for Communication, University of Pennsylvania) is a Professor of Communication at University of California, Davis, and an Affiliate Researcher (PI the ERC Starting Grant) at the Amsterdam School of Communication Research, University of Amsterdam. Her research interests include media effects on polarization and intergroup conflict, factors influencing the selection of political information online, and the effects of exposure to dissimilar political views.

Nuri Kim (Ph.D., Stanford University, 2013) is Assistant Professor at the Wee Kim Wee School of Communication and Information at the Nanyang Technological University in Singapore. Her research interests include intergroup contact theory, deliberation in the context of difference, and narrative persuasion.

Creative Commons License This is an open-access article distributed under the terms of the Creative Commons Attribution License