The Effect of Fantasy Context on Moral Action and Judgment

Abstract

Research suggests individuals import real-world facts into fictional worlds based on the type of fact and fictional context. We examined the importation of real-world morality across fictional contexts. Undergraduate (Study 1) and MTurk (Study 2) participants were randomly assigned to read either a realistic or matching fantastical interactive narrative. At seven junctions, participants were presented with a choice between behaving morally and behaving immorally to advance their goals. In Study 3, an MTurk sample judged the actions of a character who behaved immorally. For Study 1, a gender by condition interaction was found, with males electing more immoral actions in the fantasy condition. For Study 2, no such effect was found. Nonetheless, in Study 3, participants judged immoral actions in the realistic condition as more immoral compared to the fantasy context. Across all studies, transportation predicted choosing fewer immoral actions and judging immoral actions more harshly.

Keywords

fantasy reality morality Machiavellianism fiction

Reading is an inherently personal activity. While the words on the page remain consistent, the perspectives and assumptions imported into the narrative belong to the individual reader (Carreiras et al., 1996; Gerrig, 1993; Rosenblatt, 1994). Despite these differences, certain “facts” about fictional worlds seem to be held constant across readers. For example, when reading a fictional story, one does not need to be told that humans breathe oxygen or that gravity exists and exerts a force of 9.807 m/s²; unless told otherwise, readers will assume these facts to be true (Gendler, 2011; Ryan, 1980, 1991). But what about the social and moral rules that govern fictional worlds? The purpose of this paper is to examine whether the context of a fictional world—namely, whether or not that context is fantastical—affects the degree to which readers import real-world morality into make-believe worlds.

There is a wealth of evidence and theory supporting the idea that our real world morality affects the way we approach fictional media and that fictional media can affect real world morality (e.g., Bilandzic et al., 2019; Currie, 1995, 2016; Eden et al., 2015; 2017; Grizzard et al., 2017; Lester & Weber, 2016; Schnell & Bilandzic, 2017; Shafer et al., 2016; Tamborini et al., 2013; Walton & Tanner, 1994). Indeed, Flesch (2007) suggested that a large part of the appeal of fiction is that fictional stories typically offer us the opportunity to see the righteous prevailing while wrongdoers are met with their just desserts. Affective Disposition Theory (ADT; Zillman & Cantor, 1977) states that we generally enjoy seeing liked characters receive good outcomes and disliked characters receive bad outcomes, and the extent to which we like or dislike characters is derived from our perceptions of the character’s morality (Zillmann, 2000). Based on personal conceptualizations of morality, a viewer rejoices when those in line with one’s morality are rewarded and those who violate one’s morality are punished. The viewer is a “moral monitor,” constantly revising the moral correctness of a character’s actions as motivations are revealed and behaviors change (Zillmann, 2000, p. 38). Thus, morality may explain why various characters (Raney, 2011, 2017) and narratives (Raney, 2004) appeal to their audience.

Strikingly, however, evidence suggests that the enjoyment of characters is not wholly restricted to characters that are traditionally “good.” The enjoyment of morally ambiguous characters, or even villains, has also been explored in detail (Black et al., 2018; Eden et al., 2011; Janicke & Raney, 2015; Krakowiak & Oliver, 2012; Krakowiak & Tsay-Vogel, 2013, 2015; Shafer & Raney, 2012; van Ommen et al., 2014). It is easy to find examples of immoral protagonists in popular media, such as Frank Underwood of House of Cards and Walter White of Breaking Bad (e.g., Salgaro & Van Tourhout, 2018; Vaage, 2015). Morally ambiguous characters may be more complex than the traditional black-and-white conception of heroes and villains, often displaying both moral and immoral traits (Eden et al., 2015; Krakowiak & Oliver, 2012). Black et al. (2018) highlight the mechanisms by which the enjoyment of “dark” fictional characters may arise. The greater complexity of these characters and the relatability of the situations in which they find themselves may lead to greater identification on behalf of the viewer, due to increased similarity (Cohen, 2001; Konijn & Hoorn, 2005). We may also take enjoyment from the dramatic situations and emotional turmoil these characters experience and create (Breithaupt, 2015; Salgaro & Van Tourhout, 2018). Notably, prior research has shown that feeling empathy for either moral or immoral characters may increase the effects of playing video games on real-world moral behavior (Happ et al., 2015), and it has been suggested that, in the context of fiction, empathy can be separated from morality (Salgaro & Van Tourhout, 2018). Significantly, identifying with and liking a character may encourage greater forgiveness of immoral actions in order to preserve positive perceptions (Raney, 2004, 2011).

Moral disengagement (Bandura, 1999, 2006), a process by which one avoids moral agency by denying that a harm has been committed or justifying immoral behavior, has been documented as a mechanism that leads to the enjoyment of immoral fictional characters (Janicke & Raney, 2018; Krakowiak & Tsay-Vogel, 2013; Sanders & Tsay-Vogel, 2016). In fiction, no real harms are committed, and there are no real victims; this provides the reader fictional relief, freeing them from considering the consequences of actions unfolding before them (Vaage, 2013) and allowing the audience to focus more on emotion, rather than rational thought in evaluating moral actions (Vaage, 2015). Similarly, moral disengagement may play a key role in engagement with violent video games, where the player is actually acting violently within the game-world themselves (Hartmann & Vorderer, 2010).

Thus, prior research and theory suggests both that real-world morality plays a key role in narrative preferences and that audiences may go through a process of moral disengagement in order to enjoy immoral characters. However, less is known about the properties of a narrative that may make it easier for audiences to leave their real-world morality behind. Information derived from fiction can be compartmentalized (Gerrig & Prentice, 1991), but some stories are more obviously fictional than others. To what degree does the context of a narrative—and particularly, realism—play in how we import real-world morality? Here, we are interested specifically in fantasy content. Do we hold characters in worlds far removed from our own to the same moral standard as those in realistic stories? The purpose of the current study was to examine whether the extent to which a story world differs from reality affects the degree to which the audience imports real-world morality into the story.

Prior research has investigated the role that distance from reality plays in the importation of mathematical, scientific, contingent (e.g., “Washington, DC is the Capitol of the United States”), and conventional (e.g., “It is rude to pick your nose”) facts. Weisberg and Goodstein (2009) assigned readers to read different versions of the same story, which differed in their distance from reality. Distance, as conceptualized by Weisberg & Goodstein, is the extent to which a fictional world differs from our own, as operationalized by the number of reality violations (e.g., impossible or magical content) the story contains. A realistic novel that seems to take place in our world would be considered close; a fantasy story with magic and potions would be very far. In an experimental study, Weisberg and Goodstein found that both the distance of the story world from the real world and the type of fact affected the likelihood that participants would import facts from the real world to the story world. As distance from reality increased, less of the real world was imported into participants’ construction of fantastical worlds; however, some facts—namely mathematical and scientific facts—were more likely to be imported across worlds than others. Notably, as distance from reality increased, conventional facts—those related to social norms—were less likely to be imported than mathematical or scientific facts. Significantly, however, this study did not investigate whether moral facts (e.g., “stealing is wrong”) would behave similarly to conventional facts (e.g., “it is rude to pick your nose”), or whether tenets of morality would be treated more on par with scientific and mathematical facts.

This is particularly noteworthy given that moral concerns are a specific type of social convention that some people view as invariant across situations (Turiel, 1983, 2006). As outlined by Turiel, social conventions (such as forms of greeting) are a matter of consensus and serve as guidelines by which individuals can expect others to act. What is considered an appropriate greeting between two individuals may differ widely between cultures or in the relationship between them. In contrast, moral rules stem from concepts outside of popular opinion and form an obligation within a person, regardless of opinions or individual desires. Though conventions and moral prescriptions may align (killing may be considered both illegal and immoral), relaxing of a social convention does not entail an equal relaxing on morality. Even children are able to recognize this distinction and deliver differential judgment between moral transgressions and conventional transgressions. While conventional transgressions can be forgiven in response to contextual features, transgressions against morality cannot (Killen & Smetana, 2015; Turiel et al., 2014). In regard to fictional scenarios, Fast and Van Reet (2018) demonstrated that the acceptability of prosocial behaviors and the unacceptable evaluation of antisocial behaviors remained constant across real and realistic pretend scenarios. Similarly, previous findings suggest that narratives denoted as “true stories” versus narratives denoted as “fictional stories” prompted equal levels of moral disengagement (Krakowiak & Tsay, 2011).

Elsewhere, it has been argued that fiction writers do not have the authority to change morality (Levy, 2005). Whereas an authority figure (writer) can decree that a conventional rule does not apply within a specific fictional context, authorities do not hold such power over concerns of morality. This distinction has been used to explain the phenomenon of imaginative resistance, where people perceive themselves as being unable to imagine worlds in which morality works differently (e.g., Gendler, 2000). To the extent that certain moral “facts” are universal across contexts and even within the imagination, we might expect that, in constructing fictional worlds, we should import moral principles with equal vigor to that which we ascribe to mathematical facts. Yet empirical research on imaginative resistance suggests that some individuals may either not experience it, or be able to overcome it (e.g., Barnes & Black, 2016; Black & Barnes, 2017).

Strikingly, there is also work that suggests that fictionality—that is, knowing that an action or event is make-believe—may affect our moral intuitions in some cases. By reminding oneself that no real harm is committed while playing a violent video game, individuals reduce potential feelings of moral unease (e.g., Hartmann & Vorderer, 2010). Similarly, in a series of studies conducted by Sabo and Giner-Sorolla (2017), support was found for a “fictive pass” in which the evaluation of fictional or imaginary violations does not always mirror the evaluation of equivalent real-world violations. In these studies, participants rated vignettes that were stated to occur in either reality or in the context of imagination (e.g., someone who imagined doing a bad thing, watched a character doing a bad thing in a movie, or did a bad thing while playing a video game). Participants were then asked to rate the moral wrongness of the violations depicted. Results indicated that participants deemed fictional harm violations to be less immoral and less indicative of bad character than real-world harm violations; however, for purity violations, fiction had no mitigating effect. Thus, it seems that fictionality may influence our judgment of some, but not all, moral violations.

Notably, though Sabo and Giner-Sorolla (2017) looked at moral judgment of third parties who committed violations in either reality or imaginary contexts, this experiment did not assess participants’ willingness to commit these violations themselves. Furthermore, the target of the character judgement in the prior study was the person who was viewing the movie or playing the videogame, not the video game or movie characters themselves. Finally, while the fictive pass study adjusted the medium in which the act in question was housed (imagination, movie, or video game), data in the imagination condition were collapsed across these variations. Thus, the study focused on a binary comparison of real-world versus imagination, with all the imaginary contexts being realistic in nature. It is unclear, therefore, what differences in morality might be found between a movie, book, or video game set in a fantasy world and one set within a world that closely resembles our own. Fast and Van Reet (2018) manipulated fictional distance in a study of children’s pretend play and found that, although children import their real-world morality into pretense, they were more forgiving of antisocial behavior when the pretense scenario was fantastical than when it was realistic. In contrast, Hartmann and Vorderer (2010) found that altering the target of a first-person shooter game from human to zombie-like had no effect on moral disengagement.

In the current studies, we manipulated distance from the real world (i.e. fantasy context) to examine the effects on how willing participants are to choose to behave immorally for personal gain (Studies 1 and 2) and their third-person evaluations of immoral behavior (Study 3). The purpose of the first study was to examine whether participants were more willing to make immoral choices in a fantastical interactive narrative than in a nearly-identical realistic interactive narrative. This first study featured an interactive narrative in which participants progressed through a job interview process in either a fantasy or realistic context. Participants were given the option to act in a Machiavellian (behaving immorally to further their own goals) or non-Machiavellian way at each of seven decision points. If morality is imported with equal strength across fictional contexts with varying degrees of fantasy content, there should be no difference in the frequency with which participants choose the Machiavellian actions across conditions. If, on the other hand, morality is being imported differently in realistic and fantasy stories, then we would expect the more distanced fictional world to elicit less moral importation. In contrast to theories that argue that morality is invariant across contexts and imaginary worlds (e.g., Levy, 2005), we hypothesized that, like conventional facts (see Weisberg & Goodstein, 2009), real-world morality would be less likely to be imported into fantasy contexts than realistic, yet fictional, ones. Thus, we predicted that:

H1: Individuals within a fantasy context will choose more Machiavellian options than individuals within a realistic context.

Immoral behavior in these studies was conceptualized as being manipulative and acting with a win-at-all-costs disposition, in line with the definition of Machiavellianism. Individuals high in trait Machiavellianism are more likely to engage in direct and indirect antisocial behaviors (Klimstra et al., 2014; Lau & Marsee, 2013) as well as norm-violating, transgressive behaviors (for a meta-analysis, see Muris et al., 2017) and are more likely to prefer dark characters in fiction (Black et al., 2018). Thus, trait Machiavellianism was included in the following studies as a potential covariate as it was expected that individuals who are high in trait Machiavellianism would make more Machiavellian choices in the interactive narrative task. Since the experiment used an interactive narrative, transportation into the narrative was also measured. Transportation refers to the level of engagement a reader experiences in response to a narrative, entailing cognitive and affective absorption and attentional focus (Green & Brock, 2000). As research shows that the effects of reading depend on transportation (e.g., Green & Brock, 2000), it was expected that transportation into a narrative might be positively correlated with moral importation (and therefore negatively correlated with Machiavellian behavior). In addition to using transportation as a potential covariate in our analyses, we also intended to compare levels of transportation across conditions to ensure that the realistic and fantastical narratives were equally involving. Finally, because the expression of Machiavellianism can manifest differently and in different contexts for men and women (Czibor et al., 2017; Jonason et al., 2013; Jones & Weiser, 2014; Szabó & Jones, 2019), gender and the gender × condition interaction were included as possible predictors in our model.

Experiment 1

Method

Participants

Participants (N = 208) were recruited from the psychology department subject pool; all participants completed the current study and then filled out survey measures of television-viewing habits for a second, unrelated study using Qualtrics. Participants were excluded if they spent less than 5 minutes (15 participants) or more than one hour (11 participants) on the entire survey (inclusive of both studies). An additional 12 participants were excluded for not finishing the survey and 3 for poor responding (providing the same answer for all items in a scale, including those which were reverse coded) on the measure of transportation, for a final sample of 167 students. Participants were asked to provide demographic information including age, gender, and race/ethnicity (as many as applicable) for a sample consisting of 68.9% female, 28.1% male, 3% gender non-binary; 79.6% White, with ages ranging between 18 and 23 (M = 19.18, SD = 1.05). IRB approval was obtained through the university’s Institutional Review Board.

Procedure

Participants completed the Short Dark Triad (see below) and were then randomly assigned to read and complete either a fantastical or realistic interactive story. The Realistic condition had 89 participants (67.4% female, 30.3% male, 2.2% non-binary) and the Fantasy condition had 78 (70.5% female, 25.6% male, 3.9% non-binary). After completing this experiment, all participants went on to complete an unrelated experiment. Non-binary participants were not included in correlational analyses due to insufficient sample size. Participants were compensated with 0.5 hours of research credit.

Instrumentation

Interactive narrative. An interactive narrative was created for the purpose of this study. The narrative placed participants in the context of obtaining a desired job and presented various choices along the job acquisition process. The interactive narrative began with the following prompt for the realistic condition, “The end of your final year of college is quickly approaching. You’ve been applying for jobs constantly, but it seems like you never even make it to the interview stage. Your grades are decent, you’ve been involved in leadership positions in extracurricular activities, and you communicate clearly in written and spoken form. Unfortunately, you do not have much prior work experience, and it really seems like that’s holding you back.” Participants then read a description of a possible “dream job” that required 1–3 years of prior work experience and had to decide at the first decision point whether or not to lie on their resume, taking advantage of the fact that there is another student at their school with a highly similar name who has completed the internship programs in question.

In the fantasy condition, participants read the same narrative, with fantasy terms sprinkled in (i.e. “college” became “magic college” and “resume” became “wizarding resume”). After the first decision point, participants were faced with 6 additional choice points where they had to choose between two options: a moral violation to improve their chances of obtaining the job or a morally upstanding behavior that might leave them at a disadvantage. The seven target immoral behaviors included: lying on a resume; endorsing a harmful stereotype to ingratiate yourself to a bigoted interviewer; relaying the interviewer’s prejudiced behavior to an applicant of the target group to dissuade them from continuing their application to the company; sabotaging another applicant’s messages, so that they do not receive communications from the company; stealing another person’s project at an audition, so that it appears they did not turn in any work; revealing sensitive information about your fellow applicants that make them less appealing to employers; and manipulating the significant other of the chosen job applicant in order to encourage them to turn down the job. Presentation order of the Machiavellian and non-Machiavellian choices was randomized. Choices were coded with a 0 for a non-Machiavellian behavior and a 1 for a Machiavellian behavior, which was then used to create an additive index for immorality (the number of times, out of 7, that a participant had chosen the Machiavellian option). The choice options, as outlined above, were designed to be in line with the Machiavellian subscale of the Short Dark Triad (Jones & Paulhus, 2014) such as “Whatever it takes, you must get the important people on your side.” Regardless of choice, participants progressed in identical ways through the process.

Covariates

Transportation. Participants’ transportation into the interactive story was measured with a 12-item scale adapted from Green and Brock’s (2000) 15-item transportation scale. All items were on a seven-point scale (1 = strongly disagree, 7 = strongly agree). (r_α = .78).

Short dark triad. The Short Dark Triad (SD3) (Jones & Paulhus, 2014) was used to measure participants’ level of dark triad personality traits. Subscales include Machiavellianism, Narcissism, and Psychopathy. Items are ranked on a 5-point Likert scale from 1 (Strongly disagree) to 5 (Strongly agree). Items include, “I like to use clever manipulation to get my way” and “Many group activities tend to be dull without me.” Machiavellianism subscale internal consistency was: r_α = .75.

Results

Preliminary Analyses

Data analysis was conducted using IBM SPSS Version 25. Pearson correlation coefficients were computed to assess the relationship between the number of Machiavellian choices selected and potential covariates (trait Machiavellianism and state transportation). There was a positive correlation between number of Machiavellian choices selected and trait Machiavellianism, (r(167) = .31, p < .001, 95% CI [.17, .47]). There was a negative correlation between number of Machiavellian choices selected and state transportation into the narrative, (r(167) = –.230, p = .003, 95% CI [–.39, –.08]). Because these were preliminary analyses conducted to determine whether covariates would be included in our subsequent analysis, corrections for multiple comparisons were not applied (see Greenland, 2021).¹ Thus, trait Machiavellianism and transportation into the interactive narrative were included in subsequent models as covariates.

An independent-samples t-test was conducted to compare transportation across our two interactive stories to ensure that the two conditions were equally engaging. There was no significant difference in state transportation for the Realistic condition (M = 4.46, SD = .68) and Fantasy condition (M = 4.29, SD = .93); t(165) = 1.36, p = .176, d = 0.21, 95% CI [–.08, .42]. The unadjusted number of Machiavellian choices selected were: Realistic (M = 0.80, SD = 1.25); Fantasy (M = 1.19, SD = 1.49)

Primary Analyses

A negative binomial regression was run to predict the number of Machiavellian choices selected using condition, participant gender (female coded as 0, male coded as 1), and the gender × condition interaction as predictors, with trait Machiavellianism and state transportation into the narrative as covariates. A negative binomial regression was chosen to address the potential for overdispersion as well as the limited range of count values. The negative binomial regression is robust to violations of dispersion and when there is no overdispersion, the model is equivalent to Poisson regression (Heck et al., 2013). The negative binomial regression model was statistically significant Χ²(3) = 16.580, p = .001 (See Table 1). A gender by condition interaction was found Χ²(1) = 3.854, p = .050. To further examine the gender by condition interaction, separate negative binomial regressions were calculated to compare the conditions within gender; estimated marginal means were calculated holding trait Machiavellianism and state transportation constant (See Figure 1). Males were significantly more likely to commit moral violations in the fantasy condition (M_adj = 2.30, SE =0.45, 95% CI [1.57, 3.37]) than in the realistic condition (M_adj = .89, SE = .26, 95% CI [.50, 1.57]): Χ²(1) = 7.385, p = .007. Females in the fantasy condition (M_adj = 0.78, SE = .13, 95% CI [.57, 1.08]) were not significantly more likely to commit a moral violation than in the realistic condition (M_adj = 0.72, SE = 0.15, 95% CI [.47, 1.08]): Χ²(1) = .106, p = .744.

Table 1.

Parameter Estimates of the Negative Binomial Regressions in Study 1.

Effect	Estimate	SE	exp(B)	Wald	p value
Model 1
Gender	–1.079	.254	.34	18.021	< .001
Condition	–.951	.35	.386	7.385	.007
Gender × Condition	.864	.44	2.372	3.854	.05
Model 2
Gender	–.679	.281	.507	5.831	.016
Condition	–.826	.346	.438	5.692	.017
Gender × Condition	.623	.427	1.87	2.132	.144
Machiavellianism	.418	.121	1.52	11.97	.001
Transportation	–.312	.096	.732	10.46	.001

Figure 1.

Estimated Marginal Means of Condition by Gender in Study 1. Estimated marginal means of condition separated by gender for number of Machiavellian choices. Error bars represent standard error.

While the hypothesized model included state transportation and trait Machiavellianism as covariates rather than predictors, based on a request from reviewers, a subsequent negative binomial regression model that introduced these variables as predictors into the prior model was conducted. See Table 1 for results.

Exploratory Analyses

In order to determine if there were any individual choice points driving the interaction, a post-hoc analysis was performed to compare males in the fantasy condition with males in the realistic condition across each of the seven choice points. A chi-square test of independence was performed for males on each choice point. A Bonferroni adjusted alpha level of .007 was used for each test (.05/7). There was no statistically significant difference on any of the choice points after correcting for multiple comparisons.

Discussion

The purpose of Study 1 was to examine whether fictional distance from the real world (i.e. fantasy context) affected the degree to which morality was imported into an interactive story world, specifically in terms of how willing participants were to choose to behave immorally for personal gain. Males, but not females, were more likely to choose the immoral options in a fantastical interactive story than in a realistic one. These results are in line with research by Fast and van Reet (2018) finding that children view antisocial behaviors done in fantastical pretend contexts as less immoral than the same actions done in realistic pretend scenarios. These results suggest that, at least for the males in our sample, morality may vary with distance in a manner like conventional facts (e.g., Weisberg & Goodstein, 2009), rather than being held constant across all fictional contexts, thus supporting H1 (that individuals within a fantasy context would choose more Machiavellian options than individuals within a realistic context) for male participants only. This hypothesis was not supported for female participants. A subsequent model confirmed preliminary correlational analyses: Machiavellian choices were predicted by both trait Machiavellianism (individuals who are more Machiavellian at the trait level indeed chose to behave in a more Machiavellian fashion) and state transportation into the narrative (individuals who felt more transported into the narrative were less likely to make Machiavellian choices).

The gender effect found in the current experiment could have several explanations. Liberman et al. (2007) note that distanced situations (in terms of time, space, etcetera) encourage people to think in value terms; as a result, individuals with differing values may differ in planned behaviors in distanced situations. Additionally, increased psychological distance has been shown to minimize the perceived harm of a moral violation (McGraw & Warren, 2010). In comparison to men, females possess higher moral identity and lower moral disengagement (Aquino & Reed, 2002; Clark & Dawson, 1996; Detert et al., 2008; Trzebiatowska & Bruce, 2012). Thus, it may be that the distance introduced in the fantastical context encouraged males and females to default to different core values. Specifically, core values that may differ between genders are the emphasis on either agentic (dealing with self-advancement) or communal (dealing with the maintenance of positive relationships) values (e.g., Trapnell & Paulhus, 2012). Similarly, females and males may have made different choices in this task due in part to gender differences in levels of empathizing (vs. systemizing) between males and females (Greenberg et al., 2018) or in emotional factors associated with moral decision-making, such as shame proneness, guilt, and the expectation of positive emotions from immoral decisions (Ward & King, 2018).

Alternatively, it may be that the gender differences found in this experiment are the result of gender differences in the way that morality is being imported across conditions. In other words, it may be that men, but not women, see the immoral behaviors as more acceptable in a fantasy context, and that is why they are more willing to choose the Machiavellian options. Before exploring this possibility, we first wanted to examine whether the gender effect found here was an artifact of the sample used: undergraduate students in an introductory psychology class, which skewed female. Thus, the purpose of Study 2 was to examine whether a similar interaction between gender and condition would be found in a second sample, drawn from a different source of participants, and to provide a second test of the hypothesis tested in our first experiment:

H1: Individuals within a fantasy context will choose more Machiavellian options than individuals within a realistic context.

RQ1: Will the significant interaction between condition and gender found in Experiment 1 replicate with a new sample, with male participants, but not female participants, making more Machiavellian choices in the fantasy condition?

In addition to the covariates used in Study 1 (trait Machiavellianism and state transportation), age was included as a potential covariate in Study 2, given the broader age-range tested and prior research showing age-related differences in Machiavellianism (Mudrack, 1989, 1992).

Experiment 2

Method

Participants

Participants (N = 301) were recruited via Amazon Mechanical Turk (MTurk). MTurk is an online platform where workers complete tasks for compensation. Prior research has demonstrated that participants recruited via MTurk display similar results to and are more representative than convenience samples (Berinsky et al., 2012); indeed, in a recent study, MTurk participants performed better on an attention check than subject pool participants (Hauser & Schwarz, 2016). Participants were excluded from the final sample for not completing the survey (14 participants) and poor responding on covariates (5 participants). Further, in line with Study 1, participants were excluded if they spent more than one hour on the survey (1 participant) or less than five minutes for total survey duration (3 participants, all of whom were already excluded for poor responding—providing the same answer for all items in a scale, including those which were reverse coded—or not finishing). Additionally, to address reviewer concerns about the quality of responding in our MTurk sample, we also applied an exclusion criterion geared toward ensuring that participants spent a sufficient amount of time reading and responding to the interactive narrative specifically, excluding participants who completed the interactive narrative task faster than one standard deviation below the mean (seconds) (M = 223, SD = 165.30) (5 participants). Participants were asked to provide demographic information including age, gender, and race/ethnicity (as many as applicable). The final sample consisted of 276 individuals (39.1% female; 72.8% White), with ages ranging from 20 to 70 (M = 35.14, SD = 10.42). Time range:15.61 minutes. IRB approval was received to conduct this follow-up experiment.

Procedure

Participants were recruited on MTurk and then randomly assigned to either the realistic or fantasy interactive story; the procedure and covariates used were identical to those used in Study 1, Machiavellianism (r_α = .838).,transportation (r_α = .724); age was also included as a potential covariate. MTurk participants were compensated with $1.25.

Results

Preliminary Analyses

Data analysis was conducted using IBM SPSS Version 25. Pearson correlation coefficients were computed to assess the relationship between the number of Machiavellian choices selected and three potential covariates: trait Machiavellianism, state transportation, and age. Because these were preliminary analyses conducted to determine whether covariates would be included in our subsequent analysis, corrections for multiple comparisons were not applied (see Greenland, 2021). As in Study 1, there was a positive correlation between number of Machiavellian choices selected and trait Machiavellianism, (r(276) = .472, p < .001, 95% CI [.37, .58]). There was a negative correlation between number of Machiavellian choices selected and transportation into the narrative, (r(276) = –.118, p = .049, 95% CI [–.24, 0]). There was a negative correlation between number of Machiavellian choices selected and age (r(276) = –.2, p < .001, 95% CI [–.40, –.18]).² Thus, all three covariates were included in the subsequent model.

An independent-samples t-test was also conducted to compare the levels of transportation into our two interactive stories to ensure that they were equally absorbing. There was no significant difference in transportation for the Realistic condition (M = 4.88, SD = .79) and Fantasy condition (M = 4.70, SD = .86); t(274) = 1.78, p = .08, d = 0.21, 95% CI [–.02, .37]. Overall, participants in Study 2 reported greater transportation than those in Study 1; t(441) = 5.09, p < . 001, d = 0.49. The unadjusted number of choices selected in Study 2 were: Realistic (M = 1.66, SD = 1.89); Fantasy (M = 1.70, SD = 1.83) (See Figure 2). Overall, participants in Study 2 selected more immoral choices than those in Study 1; t(441) = 4.40, p < .001.

Figure 2.

Unadjusted Means by Condition in Study 2. Unadjusted means by condition for number of Machiavellian choices. Error bars represent standard deviations.

Primary Analyses

A negative binomial regression was run to predict the number of Machiavellian choices selected with condition, participant gender, and the interaction between them as predictors and age, trait Machiavellianism, and transportation into the narrative as covariates. The negative binomial regression model was not statistically significant Χ²(3) = .903, p = .825. There was no significant interaction between gender and condition Χ²(1) = .133, p = .72. While the hypothesized model included age, transportation, and Machiavellianism as covariates, based on a request from reviewers, a subsequent negative binomial regression model was run in which the variables of transportation and Machiavellianism were instead added as predictors to the prior model. See Table 2 for results.

Table 2.

Parameter Estimates of the Negative Binomial Regressions in Study 2.

Effect	Estimate	SE	exp(B)	Wald	p value
Model 1
Gender	.047	.036	1.048	1.69	.194
Condition	.009	.039	1.009	.054	.816
Gender × Condition	–.014	.053	.986	.133	.72
Model 2
Gender	.007	.03	1.007	.051	.822
Condition	–.018	.033	.982	.303	.582
Gender × Condition	.037	.045	1.038	.665	.415
Machiavellianism	.103	.011	1.11	91.41	<.001
Transportation	–.022	.012	.978	3.59	.058

Discussion

The purpose of Study 2 was to examine whether the effect found in our first study—and particularly the gender difference—replicated outside of a college sample. In contrast to Study 1, participants in Study 2 were no more likely to choose immoral, Machiavellian actions in fantasy contexts; thus H1 (that individuals within a fantasy context would choose more Machiavellian options than individuals within a realistic context) was not confirmed, and the significant interaction between gender and condition found in Experiment 1 did not replicate in a new sample (RQ1). Nonetheless, a finding that was consistent across Study 1 and Study 2, was that transportation into the narrative was correlated with fewer Machiavellian choices, suggesting that individuals who are more absorbed in a fictional world may be more likely to make choices in line with real-world morality. It should be noted, however, that corrections for multiple comparisons were not applied to our preliminary analyses. If an adjusted alpha-level had been applied, the negative correlation between transportation and Machiavellian choices would not reach significance. Thus, although the relationship found between Transportation and our outcome measure in Study 2 is consistent with the significant negative correlation found in Study 1 (which was significant even at an adjusted alpha level), further replication is needed to verify the relationship between Transportation and moral importation.

The differences between Study 1 and Study 2 could be explained by a variety of factors; however, this difference is likely not due to lower quality responding from our MTurk sample. Although we cannot directly compare the time spent on the task in Study 1 and Study 2, due to the fact that Study 1 participants completed additional measures after the completion of this study and time data was only gathered for total survey duration, it is unlikely that our MTurk participants spent less time on the study overall, as the exclusion criteria for Study 2 were slightly stricter than those used in Study 1. Further, participants in Study 2, relative to Study 1, were significantly more transported into the narrative, indicating, if anything, a higher level of engagement with the task. Additionally, the effect size of this difference was substantial at d = 0.49. Similarly, the lack of an effect in Study 2 is not due to a general reluctance to choose the Machiavellian option in Study 2, as participants in Study 2 chose a greater number of immoral choices, irrespective of condition.

What, then, can we make of the difference across these studies? One reason that Study 2 participants may have reacted more similarly across our two conditions involves the overall difference found in transportation into the narrative. As noted, participants in Study 2 were more transported into the narrative and, across both studies, transportation was shown to be related to the tendency to import real-world morality into stories. Thus, it may be that there is some kind of transportation threshold, under which there is an effect of fantasy versus reality, yet upon reaching a sufficient level of transportation, fictional distance has no impact, perhaps because the narrative feels more “real” regardless. This is in line with prior research showing that increased transportation is positively related to perceived realism of a story (Green, 2004).

Another intriguing explanation for the disparity in our results across Studies 1 and 2 involves the age difference between our samples and the specific content of our story stimuli. The sample for Study 1 was made up of college students who are very close, temporally, to the circumstances depicted in the story, which focused on the job search process. The sample from MTurk was made up of individuals who were older. For this older sample, it is possible that the distance between their own lives and a story focused on going through the interview process for a first job feels significant, regardless of whether the scenario is fantastic in nature. As outlined by Maglio et al. (2013), there is a common underlying construct to psychological distance, such that perceived distance across time is no greater or less than distance elicited through how hypothetical an event may be. Additionally, the extent to which participants were motivated to “achieve the goal” or get the job, may have influenced how they chose to respond; younger individuals may have been more motivated to acquire their “first” post-college job, while older individuals may not have shared this motivation.

Study 3, which was run concurrently with Study 2, examined whether MTurk participants would show evidence of differential moral importation across fantastic and realistic stories with respect to third-party judgment, thereby setting aside issues of personal motivation. Thus, the purpose of Study 3 was to explore whether fantasy context affected how harshly the immoral actions of a third-party fictional character were judged. Additionally, examining third-party judgment allowed us to further probe the gender effect found in Study 1, examining whether there are gender differences in moral importation when any gender differences that might exist in willingness to act in an immoral fashion are set aside. Thus, Study 3 was designed to extend the dependent measure from the first two studies from participants’ choices within an interactive narrative to participants’ moral judgments of a character within a traditional narrative, and the hypothesis and research question from Study 2 were modified accordingly:

H2: Individuals will render harsher third-party judgments of Machiavellian behavior in realistic contexts than in fantasy contexts.

As in study three, trait Machiavellianism, state transportation, and age were considered as potential covariates.

RQ2: Will the significant interaction between condition and gender found in Experiment 1 extend to third-party judgment, with male participants, but not female participants, rendering less harsh judgments in fantasy contexts?

Experiment 3

Method

Participants

Participants (N = 301) were recruited via MTurk with the same specifications as in Study 2, during the same timeframe. Individuals who completed Study 2 were not allowed to participate in Study 3 and vice versa. Participants were excluded for not completing the survey (26 participants), poor responding on covariates (providing the same answer for all items in a scale, including those which were reverse coded) (6 participants), and for being under five minutes in total study duration or for completing the evaluation section faster than one standard deviation below the mean (seconds) (M = 234, SD = 188.42) (7 participants). Participants were asked to provide demographic information including age, gender, and race/ethnicity (as many as applicable) resulting in a final sample of 262 (39.3% female; 59.9% male, .8% non-binary; 73.7% White) with ages ranging from 18 to 70 (M = 34.20, SD = 10.37). The range in completion time was 25.86 minutes. IRB approval was received to conduct this follow-up experiment.

Procedure

Participants were recruited on MTurk and then randomly assigned to read either a realistic or fantasy story. The method remained identical to Studies 1 and 2, except that the interactive narrative was adjusted so that rather than choosing actions themselves, participants were evaluating the moral wrongness of the actions taken by a fictional character who chose all of the Machiavellian options from the previous studies. In order to avoid specifying the character’s gender, a gender-neutral name (Taylor) was used for the character, and the character was always referred to by name, ensuring that no gendered pronouns were used. Actions at each of the seven choice points outlined in Study 1 were rated on a 7-point Likert scale from 1 (not at all immoral) to 7 (very immoral); these ratings were then averaged and taken as a measure of moral judgment. MTurk participants were compensated with $1.25.

Following this task, participants filled out a measure of state transportation (r_α = .74), Machiavellianism (r_α = .88), and demographics, as in Study 1. Non-binary participants were not included in correlational analyses due to insufficient sample size.

Results

Preliminary Analyses

Data analysis was conducted using IBM SPSS Version 25. Pearson correlation coefficients were computed to assess the relationship between moral judgment of the third party’s actions and the potential covariates: trait Machiavellianism and state transportation (both significantly correlated with the outcome variable in studies 1 and 2), and age (significantly correlated in study 2). Because these were preliminary analyses conducted to determine whether covariates would be included in our subsequent analysis, corrections for multiple comparisons were not applied (see Greenland, 2021). There was a negative correlation between harshness of moral judgment and trait Machiavellianism (r(262) = –.148, p = .016), with those higher in trait Machiavellianism evaluating the character’s Machiavellian actions less harshly. There was a positive correlation between moral judgement and transportation (r(262) = .231, p < .001; those who experienced greater transportation delivered harsher evaluations. There was no correlation between moral judgement and age r(260) = .064, p = .307, however age was significantly related to Machiavellianism r(260) = –.160, p = .01 and was therefore included in the model.³

There was no significant difference in transportation across the Realistic condition (M = 4.66, SD = .82) and Fantasy condition (M = 4.51, SD = .94); t(260) = 1.31, p = .191, d = .17, 95% CI [–.07, .36] . As in Study 2, participants from Study 3 were more transported into the narrative than participants in Study 1; t(427) = 2.37, p = .018.

Primary Analyses

An ANCOVA was conducted controlling for age, trait Machiavellianism, and state transportation. Variance homogeneity was assumed as the variance ratio was less than 3 (Dean et al., 1999; Keppel et al., 1992, Kirk, 2012). There was a significant effect of condition on moral judgement, such that actions taken in the realistic condition (M_adj = 5.56, SE = 0.078, 95% CI [5.41, 5.72]) were rated as more immoral than actions taken in the fantasy condition (M_adj = 5.33, SE = .079, 95% CI [5.17, 5.48]); F(1,253) = 4.46, p = .036, η_p² = .02. There was no significant effect of gender on immorality judgment, F(1,253) = 1.802, p = .181, η_p² = .01. Linear regressions were conducted to determine predictors of moral judgment. Trait Machiavellianism was a significant predictor of moral judgement (β = –.141, p = .014, 95% CI [−0.25, −0.03, η_p² = .02); those higher in Machiavellianism rated actions as less immoral. Transportation into the narrative was a significant predictor of moral judgment (β = .222, p < .001, 95% CI [.11, .33], η_p² = .06); those who were highly transported delivered harsher moral judgments. Unadjusted means and standard deviations can be found in Table 3. Estimated marginal means are displayed in Figure 3.

Table 3.

Means, Adjusted Means, Standard Deviations and Standard Errors for Immorality Ratings in Study 3.

Condition	M	(SD)	M _adj	(SE)
Realistic
Male	5.45	(.82)	5.46	(.10)
Female	5.73	(.75)	5.66	(.12)
Fantasy
Male	5.26	(.90)	5.28	(.10)
Female	5.35	(1.13)	5.37	(.12)

Note. Adjusted means controlled for age, transportation, and trait Machiavellianism.

Figure 3.

Estimated Marginal Means of Condition by Gender in Study 3. Estimated marginal means of condition separated by gender for immorality ratings. Error bars represent standard error.

In order to more fully examine the role of transportation, a moderation analysis was conducted using Model 1 of the PROCESS macro at 10,000 iterations. Condition (Realistic vs. Fantasy) was the predictor variable, transportation was the moderator, moral judgment was the outcome. There was no significant interaction between condition and transportation (β = .002, SE = .014, p = .91, 95% CI [–.03, .03]).

Exploratory Analyses

In order to determine if there were any individual evaluations driving the effect of condition on third-party moral judgment, a post-hoc analysis was performed to compare the realistic and fantasy conditions across individual evaluations. ANCOVAs were run for each of the seven points in the experiment. A Bonferroni adjusted alpha level of .007 was used for each test (.05/7). There was no statistically significant difference on any of the choice points after correcting for multiple comparisons.

Discussion

The purpose of Study 3 was to extend the findings of Studies 1 and 2, which focused on how participants chose to behave in an interactive story, to third-party judgments of immoral behavior performed within fantastical and realistic stories. The exploration of third-party judgment strove to eliminate any age-differences in motivation to land one’s first job out of college and to illuminate whether the gender difference from Study 1 is best explained via different moral standards or differences in willingness to select immoral behaviors. Participants who rated the behavior of a third-party, immoral actor judged the actions to be more immoral in a realistic story than in a fantastical one, confirming H2, that individuals will render harsher third-party judgments of Machiavellian behavior in realistic contexts than in fantasy contexts. In contrast to Experiment 1 and in line with Experiment 2, no significant interaction was found between gender and condition (RQ2). That there was no effect of gender on the evaluations of immoral behavior suggests that both males and females imported morality differently into fantasy (versus realistic) contexts, with actions taken within fantasy held to a lesser standard regarding morality. A finding that was consistent across all three studies was that transportation was related to participants’ responding: in Study 3, increased transportation predicted harsher moral judgments of the choices made by a third party. Thus, Study 3 serves to confirm the correlation found between transportation and our moral outcome measure in Study 2.

Further, in contrast to the null result in Study 2, where participants did not elect to act more immorally in a fantasy context, participants in Study 3 granted more latitude toward Machiavellian behavior in a fantasy context, suggesting an effect of context on moral importation. The results of Study 3 are in line with the suggestion that the null result for first-person action in Study 2 may be related to personal distance from the circumstances of the text and resulting motivational differences. Future research is needed to examine this possibility more thoroughly.

Since Study 3 shifted the focus from first-person action to third-person evaluation, it is more akin to prior research dealing with evaluations of, or preferences for, morally ambiguous or immoral characters (Black et al., 2018; Eden et al., 2011; Janicke & Raney, 2015; Krakowiak & Oliver, 2012; Krakowiak & Tsay-Vogel, 2013, 2015; Shafer & Raney, 2012; van Ommen et al., 2014). The current results suggest that the context of the story (fantasy versus reality) may be one factor to consider in future research on this topic.

General Discussion

Across three studies, we examined the degree to which morality was imported into realistic and fantastical fictional stories. In Study 1 and Study 2, participants completed an interactive story where they had to choose whether or not to behave in an immoral fashion for personal (in-story) gain. In Study 3, participants read a non-interactive version of the story in which a story character repeatedly behaved immorally for personal gain and were asked to rate the morality of those actions. Participants in Study 1 were college undergraduates, for whom the story was initially written to be quite relevant; participants in Studies 2 and 3 were recruited via MTurk and were older and more temporally distanced from the story’s events. The results of these experiments suggest that morality is imported to a different extent across fictional contexts, at least in some cases, and that moral importation increases with greater transportation into a narrative, regardless of whether that narrative is fantastical or realistic.

In the first study, a gender by condition interaction was present, in which males in a first-person fantasy interactive narrative chose to behave more immorally than those who read a more realistic story. This effect of gender was not replicated in Studies 2 and 3. Exposing an older population, from MTurk, to the same interactive story, which focused on getting your first job out of college, did not yield this interaction or, indeed, any differences across condition. In contrast, older participants who were tasked with evaluating the actions of a character who made the immoral choices at each of the seven decision points from the interactive story revealed an effect of condition, with immoral actions taken in the fantasy context judged less immoral than in the context of a more realistic story. Across studies, those who were more transported into the narrative were more likely to import real-world morality, choosing less immoral actions (interactive story, Studies 1 and 2) and judging a character’s Machiavellian actions as more immoral (Study 3).

As a whole, this pattern of results is striking for a variety of reasons. First, although prior research has examined the role that fictionality plays in moral judgment (e.g., Sabo & Giner-Sorolla, 2017) and moral disengagement (e.g., Krakowiak & Tsay, 2011), as well as the role that fantasy content plays in how different kinds of facts are imported into fictional worlds (Weisberg & Goodstein, 2009) and the role that fantasy context plays in moral judgment in pretend play (Fast & Van Reet, 2018), the current research is the first that we know of to show that the importation of morality into fictional worlds depends not only on story content, but also on the participant’s degree of transportation into the narrative. Our results suggest that those who are more transported into the narrative are applying the moral standards of the real world to the fictional context to a greater degree. Importantly, the amount of transportation elicited by the realistic and fantasy conditions did not differ significantly. Furthermore, transportation did not moderate the relationship between condition and judgment in Study 3. Moving forward, trait level transportability should be examined to determine whether individuals who are more easily transported differ in their evaluations of immoral actions, or whether the results found here indicate that the importation of morality into fiction depends on the particular interaction between text and reader.

Another direction for future research involves the differences found across the interactive story experiments in Study 1 and Study 2. Despite findings which suggest MTurk participants can provide viable data (Berinsky et al., 2012; Hauser & Schwarz, 2016), there are noted differences between student samples and those obtained by MTurk (Goodman et al., 2013), namely that individuals recruited via MTurk tend to be less extraverted and to possess lower self-esteem by comparison. These differences in personality could potentially explain the differences in results between Studies 1 and 2; future research is needed to directly examine the role that personality may play in these effects. Another possible reason that the effect was not replicated within an older population is that our MTurk participants were personally distanced from both the fantasy condition (fictional distance and temporal distance) and the realistic condition (temporal distance). The narrative in question featured the acquisition of a job straight out of college; the intervening years for some of the participants may have elicited distance of a similar magnitude, compared to current college students whose acquisition of a job is a salient concern. Future research is needed to investigate the role that temporal distance—and other forms, such as geographical distance—play in moral judgment.

Another difference found across Study 1 and Study 2 involves the interaction between gender and condition: a significant interaction was found in Study 1, but not in either Study 2 or Study 3, calling into question the Study 1 result. As noted in prior work, females possess higher moral identity and lower moral disengagement (Aquino & Reed, 2002; Clark & Dawson, 1996; Detert et al., 2008; Trzebiatowska & Bruce, 2012), yet in Studies 2 and 3, male participants and female participants did not differ, either in third-party moral judgment or in choice in an interactive story. It should be noted, however, that a key limitation of our study was the reliance of a single narrative (acquiring a job out of college), with only two options at each choice point. A greater assortment of options and a greater variety of moral dilemmas encountered may prove useful in probing any individual or gender differences that may exist.

For example, instead of offering participants a choice between committing a moral action for personal gain and refraining from committing that action, participants could be offered a third option—behaving morally at personal cost—to examine whether there are gender differences in altruistic behavior and whether a tendency to make altruistic in-story choices depends on the realism of the story context. Similarly, with respect to third-party judgment, it would be interesting to know whether participants differ in their judgment of fictional moral behavior, as well as fictional immoral behavior. If immoral actions committed in a realistic fictional context are seen as worse than those committed in a fantastical context, are moral actions committed in realistic contexts likewise seen as more moral than their fantastical counterparts?

A second limitation of the current studies that needs to be addressed is that the “interactive” story was not, in fact, interactive. Participants were given choices about how to behave in-story, but those choices yielded identical consequences, regardless of how the participants proceeded. It is possible that allowing participants to “succeed” while making the more moral choice decreases any motivation that might exist to behave immorally; thus, future research is needed to examine whether more significant differences may be found in cases where immoral behavior is more clearly incentivized in-story.

Another limitation worth consideration is that, in Study 3, there was only one measure in regard to evaluating the fictional character’s behavior. We focused specifically on moral judgment; absent was a measure of liking for the character who was committing said behavior, which would have further elaborated the influence of context on views of characters who behave badly. Future research incorporating the role that fantasy content may play in moral disengagement (e.g., Janicke & Raney, 2018; Krakowiak & Tsay-Vogel, 2013; Sanders & Tsay-Vogel, 2016) and Affective Disposition Theory (e.g., Raney, 2004, 2017; Zillman & Cantor, 1977; Zillmann, 2000) could be fruitful. Future studies that incorporate established fictional characters would also be interesting; much has been discussed about characters such as Walter White, Frank Underwood, and Tony Soprano (e.g., Salgaro & Van Tourhout, 2018; Vaage, 2015), all of whom live in realistic story worlds; however, it would be interesting to see if even more moral latitude is given to characters in fantasy contexts, such as those on Game of Thrones.

Finally, it should be noted that the relationship found between moral importation and transportation, though consistent across three studies, emerged in preliminary analyses that were not subjected to the statistical corrections expected for hypothesis testing (see Greenland, 2021, for discussion of applying corrections to confounding effects). Had we applied corrections for multiple comparisons to our preliminary correlational analyses, the correlation between Transportation and moral importation would remain significant in Studies 1 and 3, but not in Study 2. Thus, future research is needed to confirm the pattern of results found here, extending this work to other written narratives, as well as to film stimuli, to develop a clearer and more comprehensive understanding of the relationship between being transported into a story world and applying the precepts of real-world morality to that world.

Despite these limitations, the results presented here contribute to a growing body of work examining how morality is imported into fictional contexts and suggest that the extent to which a reader is transported may influence not only their actions in a story-world, but also their third-party evaluations of fictional violations. Strikingly, the current results may have applications outside the domain of reading. For example, might actions taken in MMORPGs (massive multiplayer online role-playing games) be held to different standards, depending on the player’s immersion in the game? The improvement in virtual reality (VR) technology in the past decade has enabled greater transportation/immersion than ever before. How might the effects found within these studies be strengthened when the evaluator can bear witness in an immersive, high fidelity environment (e.g., Balakrishnan & Sundar, 2011)? This is an area ripe for future research.

Supplemental Material

sj-pdf-1-ica-10.1177_02762366211019782 - Supplemental material for The Effect of Fantasy Context on Moral Action and Judgment

Supplemental material, sj-pdf-1-ica-10.1177_02762366211019782 for The Effect of Fantasy Context on Moral Action and Judgment by Brian Ruedinger and Jennifer Barnes in Imagination, Cognition and Personality

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Brian Ruedinger

Notes

Author Biographies

Brian Ruedinger is a PhD candidate in the Department of Psychology at the University of Oklahoma.

Jennifer Barnes is an Associate Professor of Psychology and Professional Writing at the University of Oklahoma. In addition to studying the cognitive science of fiction and imagination, she is also a New York Times bestselling novelist. Her research focuses on the impacts of engaging with fiction, the limits of the imagination, and the psychology of fandom.

References

Aquino

Reed

I. I.

(2002). The self-importance of moral identity. Journal of Personality and Social Psychology, 83(6), 1423–1440.

Balakrishnan

Sundar

S. S.

(2011). Where am I? How can I get there? Impact of navigability and narrative transportation on spatial presence. Human–Computer Interaction, 26(3), 161–204.

Bandura

(1999). Moral disengagement in the perpetration of inhumanities. Personality and Social Psychology Review, 3(3), 193–209.

Bandura

(2006). Toward a psychology of human agency. Perspectives on Psychological Science, 1(2), 164–180.

Barnes

Black

(2016). Impossible or improbable: The difficulty of imagining morally deviant worlds. Imagination, Cognition and Personality, 36(1), 27–40.

Berinsky, A. J., Huber, G. A., & Lenz, G. S. (2012). Evaluating online labor markets for experimental research: Amazon.com's Mechanical Turk. Political analysis, 20(3), 351–368.

Bilandzic

Schnell

Sukalla

(2019). The cultivation of idealistic moral expectations: The role of television exposure and narrative engageability. Mass Communication and Society, 2(5), 604–630.

Black

J. E.

Barnes

J. L.

(2017). Measuring the unimaginable: Imaginative resistance to fiction and related constructs. Personality and Individual Differences, 111, 71–79.

Black

J. E.

Helmy

Robson

Barnes

J. L.

(2019). Who can resist a villain? Morality, Machiavellianism, imaginative resistance and liking for dark fictional characters. Poetics, 74, 101344.

10.

Breithaupt, F. (2016). Empathy for Empathy’s Sake: Aesthetics and Everyday Empathic Sadism. In Empathy and its Limits (pp. 151–165). Palgrave Macmillan, London.

11.

Carreiras

Garnham

Oakhill

Cain

(1996). The use of stereotypical gender information in constructing a mental model: Evidence from English and Spanish. The Quarterly Journal of Experimental Psychology Section A, 49(3), 639–663.

12.

Clark

J. W.

Dawson

L. E.

(1996). Personal religiousness and ethical judgements: An empirical analysis. Journal of Business Ethics, 15(3), 359–372.

13.

Cohen

(2001). Defining identification: A theoretical look at the identification of audiences with media characters. Mass Communication and Society, 4(3), 245–264.

14.

Currie

(1995). The moral psychology of fiction. Australasian Journal of Philosophy, 73(2), 250–259.

15.

Currie

(2016). Does fiction make us less empathic? Teorema: Revista Internacional de Filosofía, 35(3), 47–68.

16.

Czibor

Szabo

Z. P.

Jones

D. N.

Zsido

A. N.

Paal

Szijjarto

Carre

J. R.

Bereczkei

(2017). Male and female face of Machiavellianism: Opportunism or anxiety? Personality and Individual Differences, 117, 221–229.

17.

Dean

Voss

Draguljić

(1999). Design and analysis of experiments ( Vol. 1). Springer.

18.

Detert

J. R.

Treviño

L. K.

Sweitzer

V. L.

(2008). Moral disengagement in ethical decision making: A study of antecedents and outcomes. Journal of Applied Psychology, 93(2), 374–391.

19.

Eden

Daalmans

Johnson

B. K.

(2017). Morality predicts enjoyment but not appreciation of morally ambiguous characters. Media Psychology, 20(3), 349–373.

20.

Eden

Grizzard

Lewis

R. J.

(2011). Disposition development in drama: The role of moral, immoral and ambiguously moral characters. International Journal of Arts and Technology, 4(1), 33–47.

21.

Eden

Oliver

M. B.

Tamborini

Limperos

Woolley

(2015). Perceptions of moral violations and personality traits among heroes and villains. Mass Communication and Society, 18(2), 186–208.

22.

Fast

A. A.

Van Reet

(2018). Preschool children transfer real-world moral reasoning into pretense. Cognitive Development, 45, 40–47.

23.

Flesch

(2007). Comeuppance: Costly signaling, altruistic punishment, and other biological components of fiction. Harvard University Press.

24.

Gendler

T. S.

(2000). The puzzle of imaginative resistance. The Journal of Philosophy, 97(2), 55–81.

25.

Gendler

T. S.

(2011). Is Dumbledore gay? Who's to Say?: Truth in Fiction and Authorial Authority. The Philosophers' Magazine, (52), 94–97.

26.

Gerrig

R. J.

(1993). Experiencing narrative worlds: On the psychological activities of reading. Westview Press.

27.

Gerrig

R. J.

Prentice

D. A.

(1991). The representation of fictional information. Psychological Science, 2(5), 336–340.

28.

Goodman

J. K.

Cryder

C. E.

Cheema

(2013). Data collection in a flat world: The strengths and weaknesses of Mechanical Turk samples. Journal of Behavioral Decision Making, 26(3), 213–224.

29.

Green

M. C.

(2004). Transportation into narrative worlds: The role of prior knowledge and perceived realism. Discourse Processes, 38(2), 247–266.

30.

Greenberg

D. M.

Warrier

Allison

Baron-Cohen

(2018). Testing the empathizing–systemizing theory of sex differences and the extreme male brain theory of autism in half a million people. Proceedings of the National Academy of Sciences, 115(48), 12152–12157.

31.

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.

32.

Greenland

(2021). Analysis goals, error‐cost sensitivity, and analysis hacking: Essential considerations in hypothesis testing and multiple comparisons. Paediatric and Perinatal Epidemiology, 35(1), 8–23.

33.

Grizzard

Shaw

A. Z.

Dolan

E. A.

Anderson

K. B.

Hahn

Prabhu

(2017). Does repeated exposure to popular media strengthen moral intuitions?: Exploratory evidence regarding consistent and conflicted moral content. Media Psychology, 20(4), 557–583.

34.

Happ

Melzer

Steffgen

(2015). Like the good or bad guy—Empathy in antisocial and prosocial games. Psychology of Popular Media Culture, 4(2), 80–96.

35.

Hartmann

Vorderer

(2010). It's okay to shoot a character: Moral disengagement in violent video games. Journal of Communication, 60(1), 94–119.

36.

Hauser

D. J.

Schwarz

(2016). Attentive Turkers: MTurk participants perform better on online attention checks than do subject Pool participants. Behavior Research Methods, 48(1), 400–407.

37.

Heck

R. H.

Thomas

Tabata

(2013). Multilevel modeling of categorical outcomes using IBM SPSS. Routledge Academic.

38.

Janicke

S. H.

Raney

A. A.

(2018). Modeling the antihero narrative enjoyment process. Psychology of Popular Media Culture, 7(4), 533–546.

39.

Janicke

Raney

(2015). Exploring the role of identification and moral disengagement in the enjoyment of an antihero television series. Communications, 40(4), 485–495. https://doi.org/10.1515/commun-2015-0022

40.

Jonason

P. K.

Lyons

Bethell

E. J.

Ross

(2013). Different routes to limited empathy in the sexes: Examining the links between the Dark Triad and empathy. Personality and Individual Differences, 54(5), 572–576.

41.

Jones

D. N.

Paulhus

D. L.

(2014). Introducing the short dark triad (SD3) a brief measure of dark personality traits. Assessment, 21(1), 28–41.

42.

Jones

D. N.

Weiser

D. A.

(2014). Differential infidelity patterns among the Dark Triad. Personality and Individual Differences, 57, 20–24.

43.

Keppel

Saufley

W. H.

Tokunaga

(1992). Introduction to design and analysis: A student's handbook. Macmillan.

44.

Killen

Smetana

J. G.

(2015). Origins and development of morality. In M. E. Lamb & R. M. Lerner (Eds.), Handbook of child psychology and developmental science: Socioemotional processes (p. 701–749). John Wiley & Sons, Inc. https://doi.org/10.1002/9781118963418.childpsy317

45.

Kirk

R. E.

(2012). Experimental design. In The SAGE handbook of quantitative methods in psychology (pp. 23–45). SAGE Publications Ltd. https://www.doi.org/10.4135/9780857020994

46.

Klimstra

T. A.

Sijtsema

J. J.

Henrichs

Cima

(2014). The dark triad of personality in adolescence: Psychometric properties of a concise measure and associations with adolescent adjustment from a multi-informant perspective. Journal of Research in Personality, 53, 84–92.

47.

Konijn

E. A.

Hoorn

J. F.

(2005). Some like it bad: Testing a model for perceiving and experiencing fictional characters. Media Psychology, 7(2), 107–144.

48.

Krakowiak

K. M.

Oliver

M. B.

(2012). When good characters do bad things: Examining the effect of moral ambiguity on enjoyment. Journal of Communication, 62(1), 117–135.

49.

Krakowiak

K. M.

Tsay

(2011). The role of moral disengagement in the enjoyment of real and fictional characters. International Journal of Arts and Technology, 4(1), 90–101.

50.

Krakowiak

K. M.

Tsay-Vogel

(2013). What makes characters’ bad behaviors acceptable? The effects of character motivation and outcome on perceptions, character liking, and moral disengagement. Mass Communication and Society, 16(2), 179–199.

51.

Krakowiak

K. M.

Tsay-Vogel

(2015). The dual role of morally ambiguous characters: Examining the effect of morality salience on narrative responses. Human Communication Research, 41(3), 390–411.

52.

Lau

K. S.

Marsee

M. A.

(2013). Exploring narcissism, psychopathy, and machiavellianism in youth: Examination of associations with antisocial behavior and aggression. Journal of Child and Family Studies, 22(3), 355–367.

53.

Lester, N., & Weber, R. (2016, August 10-13). Construal level affects intuitive moral responses to narrative content [Paper presentation]. In

38th Annual Conference of the Cognitive Science Society, Austin, TX, United States.

54.

Levy

(2005). Imaginative resistance and the moral/conventional distinction. Philosophical Psychology, 18(2), 231–241.

55.

Liberman

Trope

Stephan

(2007). Psychological distance. In A. W. Kruglanski & E. T. Higgins (Eds.), Social psychology: Handbook of basic principles (pp. 353–381). The Guilford Press.

56.

Maglio

S. J.

Trope

Liberman

(2013). The common currency of psychological distance. Current Directions in Psychological Science, 22(4), 278–282.

57.

McGraw

A. P.

Warren

(2010). Benign violations: Making immoral behavior funny. Psychological Science, 21(8), 1141–1149.

58.

Mudrack

P. E.

(1989). Age-related differences in Machiavellianism in an adult sample. Psychological Reports, 64(3_suppl), 1047–1050.

59.

Mudrack

P. E.

(1992). Additional evidence on age-related differences in Machiavellianism in an adult sample. Psychological Reports, 70(3 Pt 2), 1210–1210.

60.

Muris

Merckelbach

Otgaar

Meijer

(2017). The malevolent side of human nature: A meta-analysis and critical review of the literature on the dark triad (narcissism, Machiavellianism, and psychopathy). Perspectives on Psychological Science, 12(2), 183–204.

61.

Raney

A. A.

(2004). Expanding disposition theory: Reconsidering character liking, moral evaluations, and enjoyment. Communication Theory, 14(4), 348–369.

62.

Raney

A. A.

(2011). The role of morality in emotional reactions to and enjoyment of media entertainment. Journal of Media Psychology, 23(1), 18–23.

63.

Raney

A. A.

(2017). Affective disposition theory. The International Encyclopedia of Media Effects, 1–11.

64.

Rosenblatt

L. M.

(1994). The reader, the text, the poem: The transactional theory of the literary work. SIU Press.

65.

Ryan

M. L.

(1980). Fiction, non-factuals, and the principle of minimal departure. Poetics, 9(4), 403–422.

66.

Ryan

M. L.

(1991). Possible worlds and accessibility relations: A semantic typology of fiction. Poetics Today, 12(3), 553–576.

67.

Sabo

J. S.

Giner-Sorolla

(2017). Imagining wrong: Fictitious contexts mitigate condemnation of harm more than impurity. Journal of Experimental Psychology: General, 146(1), 134–153.

68.

Salgaro

Van Tourhout

(2018). Why does frank underwood look at us? Contemporary heroes suggest the need of a turn in the conceptualization of fictional empathy. Journal of Literary Theory, 12(2), 345–368.

69.

Sanders

M. S.

Tsay-Vogel

(2016). Beyond heroes and villains: Examining explanatory mechanisms underlying moral disengagement. Mass Communication and Society, 19(3), 230–252.

70.

Schnell

Bilandzic

H. K.

(2017). Television stories and the cultivation of moral reasoning: The role of genre exposure and narrative engageability. Journal of Media Ethics, 32(4), 202–220.

71.

Shafer, D.M., Janicke, S. H. & Seibert, J. (2016). Judgment and choice: Moral complexity, enjoyment and meaningfulness in interactiveand non-interactive narratives. IOSR Journal of Humanities and Social Science, 21(8), 97–106. doi: 10.9790/0837-2106010106J

72.

Shafer

D. M.

Raney

A. A.

(2012). Exploring how we enjoy antihero narratives. Journal of Communication, 62(6), 1028–1046.

73.

Szabó

Jones

D. N.

(2019). Gender differences moderate Machiavellianism and impulsivity: Implications for Dark Triad research. Personality and Individual Differences, 141, 160–165.

74.

Tamborini

Eden

Bowman

N. D.

Grizzard

Weber

Lewis

R. J.

(2013). Predicting media appeal from instinctive moral values. Mass Communication and Society, 16(3), 325–346.

75.

Trapnell

P. D.

Paulhus

D. L.

(2012). Agentic and communal values: Their scope and measurement. Journal of Personality Assessment, 94(1), 39–52.

76.

Trzebiatowska

Bruce

(2012). Why are women more religious than men? Oxford University Press.

77.

Turiel

(1983). The development of social knowledge: Morality and convention. Cambridge University Press.

78.

Turiel

(2006). Thought, emotions, and social interactional processes in moral development. In M. Killen & J. Smetana (Eds.), Handbook of moral development (pp. 7–35). Lawrence Erlbaum Associates.

79.

Turiel

Smetana

J. G.

Killen

(2014). Social contexts in social cognitive development. Handbook of Moral Behavior and Development, 2, 307–332.

80.

Vaage

M. B.

(2013). Fictional reliefs and reality checks. Screen, 54(2), 218–237.

81.

Vaage

M. B.

(2015). The antihero in American television. Routledge.

82.

van Ommen

Daalmans

Weijers

(2014). Who is the doctor in this house? Analyzing the moral evaluations of medical students and physicians of house. AJOB Empirical Bioethics, 5(4), 61–74.

83.

Walton

K. L.

Tanner

(1994). Morals in fiction and fictional morality. 68, 27–66. Proceedings of the Aristotelian Society, Supplementary Volumes,

84.

Ward

S. J.

King

L. A.

(2018). Gender differences in emotion explain women’s lower immoral intentions and harsher moral condemnation. Personality and Social Psychology Bulletin, 44(5), 653–669.

85.

Weisberg

D. S.

Goodstein

(2009). What belongs in a fictional world. Journal of Cognition and Culture, 9(1-2), 69–78.

86.

Zillman

Cantor

J. R.

(1977). Affective responses to the emotions of a protagonist. Journal of Experimental Social Psychology, 13(2), 155–165.

87.

Zillmann

(2000). Basal morality in drama appreciation. In I. Bondebjerg (Ed.), Moving images, culture, and the mind (1st ed., Vol. 1, pp. 53–63). Luton, UK: Luton University Press.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

1.35 MB