Movies serve as a pivotal mode of entertainment globally, a sentiment echoed by the Statistical Handbook of Japan (2016) [1], which identified cinema as the third most popular leisure activity in Japan. The emergence and proliferation of video-on-demand platforms, including Netflix, YouTube, and Amazon Prime Video, have augmented movie accessibility. Moreover, as shown by Statista (2021) [2], Netflix acquired 36.57 million new subscribers worldwide in 2020 due to restrictions on going out and public gatherings during the COVID-19 pandemic, which accelerated the change in ways of watching movies. As a result, movie preferences have become more diverse and personalized. Accordingly, inferring that one’s personality may influence one's movie preferences is reasonable.

The Dark Triad, encompassing Machiavellianism, Psychopathy, and Narcissism, outlines antisocial personality (Jonason and Webster, 2010 [3]; Paulhus and Williams, 2002 [4]; Tamura, Oshio, Tanaka, Masui, and Jonason, 2015 [5]). Machiavellianism is the trait of manipulating and lying to others for personal gain; Psychopathy is the trait of lacking empathy for others, having no concern for the right or wrong of actions, and being cold and fearless; and Narcissism is a self-centered trait of feeling that one is entitled to special recognition from others, such as fame and praise. The Big-Five is a framework that attempts to view the entirety of human personality in terms of five aspects: Extraversion, Agreeableness, Conscientiousness, Neuroticism, and Openness (Gosling, Rentfrow, and Swann, 2003 [6]; Oshio, Abe, and Cutrone, 2012 [7]). Extraversion describes characteristics such as being outgoing and energetic; Agreeableness denotes being friendly and compassionate; Conscientiousness measures how efficient and organized an individual is; Neuroticism means an individual’s sensitivity and nerve; Openness expresses curiosity and inventiveness. In addition, it has been consistently proven that Dark Triad traits are related to low Agreeableness.

Previous studies have extensively explored the impact of personality on movie preferences. Tamborini, Stiff, and Zillman (1987) [8] found that Machiavellianism was a good predictor of preferences for graphic horror. Hall (2005) [9] reported a negative correlation between psychoticism and watching romance and comedy movies, which Neuroticism moderated. Regarding Big-Five personalities, Karumur, Nguyen, and Konstan (2016) [10] provided a set of data showing that individuals who scored high on Openness fancied drama and romance movies, conscientious individuals were more inclined to enjoy romance genres, and neurotic individuals tended to prefer comedy and romantic movies.

Moreover, Kraaykamp and Eijck (2005) [11] studied three types of cultural practices – book reading (literary novels, literature in foreign languages, suspense, romantic novels), cultural behavior (visiting classical concerts, visiting art museums, visiting historical museums, visiting pop concerts), and television preferences (cultural programs, informative programs, soap programs, erotic programs). They found that people high in Openness enjoyed all those practices, whereas none appealed to extroverted people. Conscientious individuals preferred book reading but had no liking for watching TV programs. Emotionally stable (low in Neuroticism) people disliked soap or erotic TV programs. Moreover, people who scored high in Agreeableness showed particular interest in soap programs compared to other TV programs and fancied romantic novels.

However, the conventional approach of assessing movie preferences by genre is becoming increasingly challenging, given the complexity and overlap in modern movie categorizations. For instance, some comedies can be violent (e.g., Hot Fuzz), and some romance movies are full of frightening scenes (e.g., The Lodgers). Therefore, we assessed movie preferences according to specific content instead of genres in this study. The particular contents were evaluated based on the ratings provided by the IMDb (Internet Movie Database) parental guidance. Parental guidance is an indicator for parents to determine the appropriateness of content when showing their children movies. It is divided into five categories – Sex/ Nudity, Violence, Profanity, Alcohol/ Drug, and Frightening. We aim to unveil underlying patterns and motivations governing movie selection among distinct demographic groups by analyzing the relationships between these content preferences and personality traits.

The internal consistency, as measured by Cronbach’s Alpha, yielded reliability coefficients of 0.76 for Machiavellianism and Narcissism and 0.58 for Psychopathy. For the Big-Five personality traits measured by TIPI-J, the correlations between the paired items were as follows: Extraversion (r=.60), Agreeableness (r=.20), Conscien- tiousness (r=.37), Neuroticism (r=.44), and Openness (r=.53).

In examining the correlations between Dark Triad traits and movie content preferences, Machiavellianism positively correlated with Sex/ Nudity and Alcohol/ Drug, Psychopathy positively associated with Profanity and Alcohol/ Drug, and Narcissism positively related to Sex/ Nudity, Profanity and Alcohol/ Drug. Regarding Big-Five personality dimensions, Extraversion and Agreeableness negatively correlated with Profanity. Conscientiousness had a negative correlation with Violence and Frightening. Openness is positively associated with Sex/ Nudity, Profanity, Alcohol/ Drug, and Frightening. The descriptive statistics and the matrix of correlations among the personality traits and content preference ratings are encapsulated in Table 2.

This study aimed to verify the relationship between personality traits and movie preferences. We assessed participants’ personalities by DTDD-J and TIPI-J, quantified the descriptions of each participant’s favorite movies, and then examined the associations.

Regarding Dark Triad personality traits, except for an insignificant negative correlation between Psychopathy and Frightening content, all showed positive associations with the content categories. It was proved that Narcissism and Psychopathy were predictors of sexual partners (Borráz-León and Rantala, 2021) [12], and narcissistic people were more highly aroused in sex (Zeigler-Hill and Andrews, 2021) [13], which may lead to preferences for Sex/ Nudity. Additionally, the positive correlation between the Dark Triad traits and sexual coercion (Lyons, Houghton, Brewer, and O’Brien, 2020) [14] might further substantiate this finding. The positive correlation between substance use and both Narcissism and Psychopathy (Stenason and Vernon, 2016) [15] illuminates the lack of resistance these individuals might have towards Alcohol/ Drug content. Masui and Ura (2016) [16] found that individuals who were high in Psychopathy used aggressive humor more frequently. Aggressive humor is antisocial humor that attempts to be funny by being sarcastic or humiliating to others (Martin, 2007) [17]. This may explain the connection between Psychopathy and the preference for Profanity.

In the realm of Big-Five personality traits, Openness showed positive associations with all five categories, barring an insignificant correlation with Violence. One of the most important features of people high in Openness is that they proactively seek new experiences, which may result in their preferences for stimulating scenes. Agreeableness was generally negatively correlated with all content categories, given the aggressive and unpleasant nature of these contents, which contrasts with the disposition of agreeable individuals. Similarly, Conscientiousness was negatively correlated with all five categories at large. The tendency toward being diligent and careful may keep conscientious people away from dangerous and eccentric behaviors, and this feature may also be expressed in visual preferences. These results are consistent with Kraaykamp and Eijck’s study (2005) [11], which suggested that high Openness leads to diversity in entertainment preferences, while those high in Conscientiousness are probably not fond of visual entertainment.

Exploring the relationships among behaviors, personalities, and content preferences has become common, especially in large enterprises. For instance, Cambridge Analytica Ltd (CA), a British consulting company, became famous for the Facebook-Cambridge Analytica data scandal in the 2010s. CA collected data from millions of Facebook accounts without consent and then used the data hoard for political activities by personality-profiling the users (Gibney, 2018) [18]. Video-on-demand platforms harness user data, such as search history and playback interactions, to predict and cater to viewers’ content preferences. This study could augment our comprehension of the intricate dynamics between personality and behavior, promising academic insights and commercial utilities.

Three sampling biases can be raised in this study. First, the gender distribution of the participants was skewed, with males comprising only 25% of the sample. Given that gender differences in movie preferences are well-documented (Wühr, Lange, and Schwarz, 2017) [19], this imbalance might have influenced the results. Second, the narrow age range of the participants presents another limitation. As movie preferences evolve with age (Infortuna et al., 2021) [20], a more diverse participant age range would provide a more comprehensive insight. Third, as stated in Glevarec, and Cibois, (2021) [21], educational level is critical in shaping an individual’s movie preferences. Our study predominantly involved university students, which raises concerns about the lack of diversity in educational backgrounds among participants.

Another critical aspect is the cultural and geographical context of the IMDb parental guidance. The ratings are predominantly based on US cultural norms and values, but our study focused on a group of university students in Japan. This geographical and cultural discrepancy could introduce a bias or misalignment in interpreting content preferences.

Moreover, the reliance on questionnaire-based data collection might need to fully encapsulate the complexity of movie preferences, which numerous external factors beyond personality traits can influence. For example, blockbuster movies may leave a stronger impression than normal ones, and recent films are likely to be recalled more easily than older ones. In addressing these limitations, future research should consider integrating experimental methods to enhance the reliability and validity of the findings. One potential approach could involve participants viewing movies with specific content in a controlled setting. Subsequently, assessing the correlation between their evaluations and their personality traits could offer more nuanced and reliable insights. This mixed-methods approach would mitigate the impact of recall bias and the influence of high-profile movies, providing a more authentic representation of true preferences.

The research aims to examine the relationships between the Dark Triad personality traits and specific movie content preferences, utilizing IMDb’s parental guidance ratings as a nuanced alternative to traditional genre classifications and the proclivity of individuals high in Dark Triad traits towards more explicit or stimulating movie contents is elucidated. Further research is expected to integrate broader participant demographics and test methodologies.