Perceptions of Moral Violations and Personality Traits Among Heroes and Villains

Abstract

This study investigates disposition-formation processes in entertainment by predicting perceptions of media heroes and villains by their behavior in specific moral domains. Participants rated self-selected heroes and villains from television and film along the moral domains of care, fairness, loyalty, authority, and purity (Haidt & Joseph, 2007) as well as along dimensions of warmth, competence, and duplicity used in impression-formation research (Fiske, Cuddy, Glick, & Xu, 2002). Results show that heroes violate moral norms in domains of authority and purity, whereas villains violated moral norms in the domains of caring and group loyalty. Furthermore, these moral violations are associated with personality dimensions of warmth and competence differently for each character type, such that impressions of heroes are driven by their work in the care domain (i.e., saving or protecting people), whereas for villains, violation of purity norms is most strongly associated with subsequent impression formation processes.

Notes

¹The film versus television distinction was part of a secondary research question investigating if heroes and villains would be evaluated differently for television versus film. There was a small three-way interaction among character role, media, and moral domain, F(4, 286) = 3.31, p = .01, partial ή² = .01, which is why it is included in the present analyses. Notably, the main effect for media failed to reach significance, F(1, 289) = 1.94, p = .33, as did the two-way interactions of media with character, F(1, 289) = 2.17, p = .14, and media with domain, F(4, 286) = .64, p = .63. Visual inspection of the mean patterns in the three-way interaction suggests that film heroes violate the fairness, care, and purity domains less than TV heroes. By contrast, film villains are perceived to violate all domains more than are TV villains (although villains and heroes are very close on the loyalty domains). However, these differences are marginal. Examining the trait variables, the main effect for media failed to reach significance, as did the interactions of media with character, or media with trait (all F < 1).

²The 10 most frequently chosen heroes, with number of participants selecting character in parentheses, were Batman (60), Superman (51), Spiderman (22), Jack Bauer (13), Hercules (9), Harry Potter (7), House (6), Jack Shepherd (8), Iron Man (4), and Meredith Grey (3). Villains selected were The Joker (105), Lex Luther (12), Cruella DeVille (11), Ursula the Sea Witch (10), Spencer Pratt (8), Jafar (6), Plankton (6), Scar (6), Dan Scott (5), and Darth Vader (5).

³Seventy-six participants (26%) used “paired” hero/villains from the same content source. We ran an exploratory analysis using content source (same or different) as a between-subjects factor and domain and character as within subject. For moral domains, the only significant effect was in a three-way interaction with Character × Domain × Same Content Source, F(4, 287) = 2.81, p = .02,  = .06, which is seen in a shift in domain ranking among villains such that authority violations were slightly lower and purity violations were slightly higher among villains from the same versus different content sources. There were no differences between person-perception variables in heroes and villains from same or different content, and the pattern of relationships between the personality trait variables and moral domains was similar independent of content source (all Fs < 1).

⁴The following items were removed from the character trait scale due to inconsistencies between factor loadings for heroes and villains. For heroes the following items loaded with the competence factor: a good person, helpful, devoted, evil, and a bad person. Vicious loaded with the duplicitous factor. For villains the following items loaded with the warmth factor: a good person helpful, good-natured, and sincere; the following items loaded with the duplicitous factor: crazy, dangerous, wicked, and aggressive. All analyses reported in text were replicated with composites consisting of all items loading on each factor. None of the replicated results were different in any way (direction, magnitude, or interpretation), nor added any meaningful information from the results presented in text; therefore they were omitted from this report. They are available upon request from the first author.

⁵The selection of such impure-but-competent characters in this study raises questions regarding the extent to which our results represent perceptions of pure heroes such as Superman as opposed to antihero characters such as House and Don Draper. Recent research (Tamborini, Grizzard, Eden, & Lewis, 2011) suggests that that antiheros can be distinguished from pure heroes and villains by the extent to which they uphold and violate moral domains. In line with the current study, pure heroes were perceived to score significantly higher than villains on upholding all moral domains. By contrast, antiheroes fell somewhere in between heroes and villains, scoring more below heroes on the upholding of some domains but not others. Although it was not our primary rationale for the present study, analyses were conducted to compare characters in the present study that we might think of as antiheroes with those we might consider pure heroes. The results of these analyses are in line with findings from Tamborini, Grizzard, et al. (2011). That is, antiheros (using Batman, House, and Jack Bauer; n = 79) differed significantly compared to pure (all other heroes in our study; n = 212) on ratings of harm, F(1, 289) = 17.94, p < .00; fairness, F(1, 289) = 11.10, p < .05; and authority, F(1, 289) = 8.90, p < .05. The differences in loyalty and purity were not significant. The means for the significant results were consistent such that the antiheros were always perceived to violate the moral domain more than the pure heroes.

These findings are presented in a footnote given the concerns stemming from the fact, in this analysis, the determination of whether a character was a hero or antihero was a post hoc determination by the researchers. We cannot assume that our categorization of these characters as hero or antihero would be the same as those participants. In addition, the comparisons we have made here go far beyond the purpose of the current study, which was to examine heroes versus villains rather than heroes versus villains versus antiheroes. Clearly, this is an important direction for research to take in the future. This study takes a first step in this direction by asking participants to name for themselves who their heroes are. Perhaps an equally interesting question is why, when asked to identify a hero, so many participants picked these types of conflicted characters. This issue goes beyond questions examined here (i.e., the extent to which these intuitive domains shape audience perceptions of heroes vs. villains) to consider whether these domains shape individuals' perceptions of heroes versus antiheroes. Although we may think this is true, this is a question for future research.

Additional information

Notes on contributors

Allison Eden

Allison Eden (Ph.D., Michigan State University, 2011) is an Assistant Professor in the Department of Communication Science at VU University Amsterdam. Her research interests include media psychology, entertainment effects, and morality.

Mary Beth Oliver

Mary Beth Oliver (Ph.D., University of Wisconsin–Madison, 1991) is a Distinguished Professor in the Department of Film/Video and Media Studies at The Pennsylvania State University. Her research interests include media effects, with a focus on media and emotion, and media and social cognition.

Ron Tamborini

Ron Tamborini (Ph.D., University of Indiana, 1982) is a Full Professor in the Department of Communication at Michigan State University. His research examines both traditional and new media, with a focus on how characteristics of technology alter the psychological experience and influence of media.

Anthony Limperos

Anthony Limperos (Ph.D., The Pennsylvania State University, 2011) is an Assistant Professor in the School of Journalism and Telecommunications at University of Kentucky. His research interests include media uses and effects and new communication technology.

Julia Woolley

Julia Woolley (Ph.D., The Pennsylvania State University, 2012) is an Assistant Professor in the Communication Studies Department at California Polytechnic University, San Luis Obispo. Her research interests include media effects, entertainment psychology, and communication and new media.