Do trolls just want to have fun? Assessing the role of humor-related traits in online trolling behavior

Highlights

• The role of humor-related traits in online trolling was analyzed.

• Trolls tend to score higher on joy in laughing at others and aggressive humor.

• Katagelasticism incrementally predicts online trolling beyond the Dark Tetrad.

• Only among high scorers in katagelasticism does psychopathy relate to greater online trolling.

Abstract

This research has aimed to test whether “darker” humor-related dispositions can contribute to the prediction of online trolling over and above the influence of dark personality traits. A total of 201 participants (50.7% women) gave responses to dark personality, humor, and online trolling measures. Our results corroborated the one-factor structure of the Spanish version of the Global Assessment of Internet Trolling (S-GAIT) and replicated online trolling's robust associations with increased psychopathy, sadism, and Machiavellianism. Online trolling also correlated with an elevated use of aggressive and self-defeating humor styles, as well as with heightened expressions of the joy in laughing at others (i.e. katagelasticism) and the joy of being laughed at (i.e., gelotophilia). When applying hierarchical regressions to eliminate the redundancy among these traits, we found that katagelasticism incrementally explained variance ($\Delta$R² = 10.2%) in online trolling even after accounting for gender and the Dark Tetrad (i.e., male gender and high psychopathy as main predictors: $\Delta$R² = 27.3%). A subsequent moderation analysis indicated that higher levels of psychopathy was related to a greater engagement in trolling behavior among those high in katagelasticism. This research provides empirical evidence that contributes to elucidate the “dark” humorous nature of this pervasive antisocial online behavior.

Introduction

Online trolling is a specific antisocial behavior that involves malicious and deliberate provocation of others, with no apparent instrumental motivation, and that occurs across different locations on the Internet, such as social media or news website discussion forums (Buckels et al., 2014; Coles & West, 2016; Ruiz et al., 2011). Recent research has suggested that those who have a higher engagement in online trolling (hereafter trolls) disturb others in a disrespectful manner as a means of seeking enjoyment and fun (Bishop, 2014). Similarly, it has been proposed that trolls may poke fun at their online victims through provocative comments purely for the sake of their own and of others’ (humorous) entertainment (Dynel, 2016; Sanfilippo et al., 2018). Notwithstanding the above, no empirical research has yet investigated the patterns of association that online trolling behavior may take in relation to a comprehensive selection of humor-related dispositions in general, and in relation to those comprising “darker” humor-related characteristics in particular.

This research is intended to narrow this gap in the trolling literature by testing whether perpetration of trolling behaviors varies as a function of interindividual differences in humor-related traits, as conceptualized by two well-established models: (a) the Humor Styles model (i.e., affiliative, self-enhancing, aggressive, and self-defeating humor styles; Martin et al., 2003) and the model for Dispositions toward Ridicule and Being Laughed at (i.e., gelotophobia, gelotophilia, and katagelasticism; Ruch & Proyer, 2009). Moreover, since there is broad evidence that dark personality traits predominate in the personality effects on online trolling (Buckels et al., 2014), and that these aversive non-clinical dimensions are correlated with humor-related dispositions (Martin et al., 2012; Torres‐Marín et al., 2019a), we will try to isolate the contribution of these humor dimensions to this form of antisocial online behavior while controlling for the overlapping variance with the Dark Tetrad constellation of personality (i.e., narcissism, Machiavellianism, psychopathy, and sadism; Buckels et al., 2014).

Drawing on the theoretical postulates pertaining to online trolling, trolls are online users who seek to take advantage of their anonymity to generate conflicts through destructive comments and to create scenarios where their possible victims may appear somehow foolish (e.g., Binns, 2012; Buckels et al., 2014). It can be argued that these individuals exhibit socially aversive characteristics such as making biting criticism of others’ mistakes, stoking arguments, social manipulation, attention-seeking actions, deficits in empathy, and getting pleasure from upsetting others (Buckels et al., 2014; Hardaker, 2010; Kirman et al., 2012; Phillips, 2015; Sest & March 2017; Synnott et al., 2017). Trolls also have the will to break social norms in order to obtain benefits for themselves (and their followers), regardless of possible adverse consequences for their victims such as, for example, distress (Craker & March 2016; Golf-Papez & Veer, 2017).

Accumulated empirical evidence has found the Dark Tetrad personality traits to be significantly correlated with perpetration of online trolling behaviors. Buckels et al. (2014) were the first to show that all of these adverse non-clinical personality traits (i.e., narcissism, Machiavellianism, psychopathy, and sadism) are associated with a greater tendency to engage in online trolling. This pattern of correlations has subsequently been corroborated by further empirical investigations across various Internet locations such as Facebook (Craker & March 2016) or Location-Based Real-Time Dating applications (i.e., Tinder; March et al., 2017). Importantly, there is also increased convergence that, though all these dark personality factors are positively related to individuals’ engagement in online trolling, the traits of sadism and psychopathy exert the strongest predictive utility regarding the perpetration of online trolling behaviors (e.g., Buckels, et al., 2019; Craker & March 2016; March et al., 2017). Furthermore, trolling behavior has been negatively correlated with agreeableness (Buckels et al., 2014) and, in the same vein, it has been reported that trolls are more prone to show reduced empathy (Sest & March 2017), greater negative social reward motivation (Craker & March 2016), and dysfunctional impulsivity (March et al., 2017).

Altogether, these findings suggest that socially aversive traits, particularly psychopathic and sadistic tendencies, seem to predominate in the personality effects on online trolling. Nonetheless, there are alternative individual differences variables that remain understudied in online trolling studies—even when they are also strongly linked to dark personality and disruptive behavior. In this regard, certain narrow dispositions belonging to the negative, “dark side” of humor, such as aggressive humor or the joy in laughing at others (i.e., katagelasticism), can help to elucidate the personality characteristics of those who are more inclined to engage in troll-like behaviors.

Humor is a pervasive communicative manifestation that can be used to exhibit distinct, even opposite, emotional motivations in interpersonal settings (Martin, 2007). Although the study of humor has traditionally been approached by way of positive psychology's paradigms (e.g., its consideration as a morally valued trait: see Peterson & Seligman, 2004), there exist multiple humor-related expressions, such as ridicule or derisive laughter, that can serve to communicate negative intentions in a more subtle and socially acceptable manner (Kowalski, 2000; Shapiro et al., 1991).

Humor researchers have progressively introduced different comprehensive models for describing inter-individual differences in how individuals deal with humorous situations. For instance, Martin et al. (2003) proposed the existence of four humor styles. Whilst two humor styles encompass socially-oriented tendencies—affiliative humor (i.e., the use of benign jokes to enhance an individual's social bonds) and self-defeating humor (i.e., the use of self-deprecating humoristic comments to obtain others' social approval)—the other two humor styles involve more intrapsychic aspects: self-enhancing (i.e. the maintenance of a humorous outlook on life as a way of building resilience), and aggressive humor (i.e., ridiculing others through hostile jokes and pranks to boost one's own self). These humor styles can in turn be divided into positive (i.e., affiliative and self-enhancing humor styles) or negative (i.e., aggressive and self-defeating humor styles) expressions of humor, attending to their conceptual nature and impact on individuals' social functioning. There are also alternative models, referring to a more restricted range of inward experiences or behaviors, that also subsume aspects relating to the “dark” nature of humor and laughter. For example, Ruch and Proyer (2009) delineate three inter-correlated but sufficiently distinct dispositions toward ridicule and being laughed at namely: gelotophobia (i.e., the fear of being laughed at), gelotophilia (i.e., the joy of being laughed at), and katagelasticism (i.e., the joy in laughing at others). It is worth noting that gelotophobia and gelotophilia do not represent opposite poles of the same latent factor, as these dispositions predict differently external outcomes (e.g., relationship satisfaction; Brauer & Proyer, 2018). The three narrow laughter-related dispositions enable researchers to look into the different roles involved in teasing and ridicule-related experiences, and offer valuable insights into avoiding and initiating laughter directed toward oneself and others.

Broad empirical evidence validates the notion that among higher scorers in aversive non-clinical personality traits an elevated incidence of some tendencies toward “dark” dimensions of humor styles may be relevant. For instance, Veselka et al. (2010) investigated the Dark Triad-humor styles association and showed that inclinations to psychopathy and Machiavellianism yielded positive correlations with the use of aggressive humor and, to a lesser extent, self-defeating humor. This study also revealed a small direct association between narcissism and affiliative humor style. Shortly after, Martin et al. (2012) replicated and extended these findings by deploying a facet approach. They found that aggressive humor positively relates to diverse subcomponents of the Dark Triad such as callous affect, erratic lifestyle, interpersonal manipulation, superiority/arrogance, interpersonal tactics, cynical view of humor nature, and disregard for conventional morality. Interestingly, self-defeating humor had a similar pattern of results but its correlation coefficients reflected smaller effect sizes. Finally, affiliative humor had substantial associations with narcissistic facets, namely: leadership/authority, superiority/arrogance, and self-absorption/self-admiration. These findings suggested that aggressive humor had the most robust relationships with the Dark Triad traits. In line with these results, aggressive humor has been associated with reduced levels of empathy-related traits (Hampes, 2010), in addition to lower levels of honesty-humility and greater ones of trait-anger (Torres-Marín et al., 2018; Torres‐Marín et al., 2019a).

Regarding dispositions toward ridicule and being laughed at, in a recent research paper, Torres‐Marín et al., 2019a found that katagelasticism was strongly associated with greater expressions of psychopathy and Machiavellianism. In contrast, gelotophobia and gelotophilia had a less demonstrable overlap with the Dark Triad, offering only small-to-moderate associations of gelotophilia with higher levels of narcissism and gelotophobia with elevated Machiavellianism and reduced narcissism. These findings were consistent with Proyer and colleagues (2012)’ study, which investigated these ridicule-related dispositions in relation to different psychopathic traits. These authors found that katagelasticism contributed to the prediction of a general factor of psychopathy (i.e., involving manipulative lifestyles, callousness, antisocial behavior, and grandiosity). Although gelotophilia correlated positively with grandiosity and antisocial behavior, and gelotophobia with greater manipulative lifestyles and reduced grandiosity, these narrow dispositions did not incrementally explain the variance in the general factor of psychopathy after the inclusion of katagelasticism. Moreover, katagelasticism has also dominated the effects of these dispositions on other aversive tendencies, being related to diminished agreeableness (Ďurka & Ruch, 2015), honesty-humility (Torres‐Marín et al., 2019a), virtuousness (Proyer et al., 2014), and greater expression of bullying-type behavior (in young children and in adolescents; Proyer, Neukom, et al., 2012).

Taken together, these empirical data suggest that the “darker” humor-related personality traits—particularly aggressive humor and katagelasticism—might represent relevant characteristics to describe the dark personality of an individual (Martin et al., 2012; Torres‐Marín et al., 2019a). Therefore, one could reasonably expect that these linkages may be generalized to more specific socially aversive behavioral manifestations such as online trolling.

Previous work has already suggested that online trolling could be suffused with humorous motivations. For instance, the concept of lulz, which refers to an internet neologism strongly linked to the domain of online trolling, encompasses laughter and fun stemming from other internet users' distress (Buckels et al., 2019; Phillips, 2015). Further, this observation is consistent with Dynel (2016)'s work, which postulated that online trolling could be driven by the search for one's own and for others' humorous amusement at the expense of a particular target. More specifically, and grounded on the principles of the superiority theory of humor (see Hobbes, 1996), trolls and their audience may share a humoristic and wit sense of superiority when the former successfully deceive their online victims (Dynel, 2013, 2016). These notions have been extended by using other methodologies. Sanfilippo et al. (2018) proposed the existence of a type of humorous online trolling, and included satire (i.e., sarcastic and critical comments on others' weakness) as a representative troll-like behavior; this as a result of integrating undergraduate students' perceptions of trolling behavior with media and scholarly conceptualizations.

Despite this growing body of knowledge connecting humor and online trolling, to the best of our knowledge, no studies have yet disentangled the association between humor, as an individual difference variable, and online trolling. Humor-based differences have been proven to be relevant for predicting other derogatory online behaviors such as cyberbullying behavior among adolescents (e.g., aggressive humor as a positive robust predictor: (Sari, 2016)). Hence, one might anticipate that these findings can be generalized to other disruptive online behavior such as trolling (Buckels et al., 2014) in addition to other personality tendencies which relate to the “dark” side of humor (i.e., the joy in laughing at others: Ruch & Proyer, 2009).

This research has four main goals: (a) to investigate the internal consistency and factorial structure of the Spanish version of the Global Assessment of Internet Trolling (S-GAIT); (b) to replicate prior findings on the connections between the Dark Tetrad traits and online trolling; (c) to provide the first empirical evidence for the association of different humor-related dispositions with the perpetration of trolling behaviors; and, simultaneously, (d) to determine the potential contribution of such humor traits to the prediction of online trolling over and above the influence of gender and the Dark Tetrad.

In the first place, we expected to corroborate the one-factor structure of the S-GAIT (H1) and to replicate the moderate-to-strong positive associations of online trolling with sadism, psychopathy and Machiavellianism, as well as its small positive association with narcissism (H2; e.g., Buckels et al., 2014). Moreover, attending to theoretical linkages between online trolling and potential “dark” humorous motivations (i.e., joking at the expense of others; e.g., Dynel, 2013, 2016; Philipps, 2015; Sanfilippo et al., 2018), we expected positive associations between online trolling and injurious humor-related traits such as aggressive humor and katagelasticism (H3). We then tested whether the associations of these “dark” humor-related traits with online trolling would go beyond the Dark Tetrad constellation. Considering the similar though sufficiently distinctive nature of socially aversive personality traits and “dark” humor-related dispositions (i.e., mainly moderate-to-strong correlations: Martin et al., 2012; Torres‐Marín et al., 2019a; Veselka et al., 2010), we anticipated that these humor-related dispositions would contribute to the prediction of online trolling over and above their overlapping variance with the Dark Tetrad constellation (H4). Concurrently, we controlled for gender effects, as dark personality traits, aggressive humor, katagelasticism, and online behavior are more pronounced in men (Buckels et al., 2014; Jones & Paulhus, 2014; Martin et al., 2003; Ruch & Proyer, 2009).

Drawing on earlier results pertaining to personality traits, humor-related dispositions, and malevolent online behavior (e.g., Buckels et al., 2014; (Sari, 2016)), we surmised that gender, the Dark Tetrad and humor traits would—in combination—account for the inter-individual variance in online trolling behavior reflecting a medium-to-large effect size (i.e., total R² ≥ 0.13; Cohen, 1988).