Personality and Individual Differences 67 (2014) 97–102

Contents lists available at ScienceDirect

Personality and Individual Differences
journal homepage:

Trolls just want to have fun
Erin E. Buckels a,⇑, Paul D. Trapnell b, Delroy L. Paulhus c
University of Manitoba, Winnipeg, Manitoba, Canada
University of Winnipeg, Winnipeg, Manitoba, Canada
University of British Columbia, Vancouver, British Columbia, Canada

a r t i c l e i n f o a b s t r a c t

Article history: In two online studies (total N = 1215), respondents completed personality inventories and a survey of
Available online 8 February 2014 their Internet commenting styles. Overall, strong positive associations emerged among online comment-
ing frequency, trolling enjoyment, and troll identity, pointing to a common construct underlying the
Keywords: measures. Both studies revealed similar patterns of relations between trolling and the Dark Tetrad of
Sadism personality: trolling correlated positively with sadism, psychopathy, and Machiavellianism, using both
Dark Tetrad enjoyment ratings and identity scores. Of all personality measures, sadism showed the most robust
Dark Triad
associations with trolling and, importantly, the relationship was specific to trolling behavior. Enjoyment
of other online activities, such as chatting and debating, was unrelated to sadism. Thus cyber-trolling
Antisocial Internet behavior appears to be an Internet manifestation of everyday sadism.
Personality Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction disruptive aspects may distinguish trolling from other forms of on-
line antisociality, such as cyber-bullying, where perpetrator identi-
Online trolling is the practice of behaving in a deceptive, ties are usually clear (Lenhardt, 2013) and the intent is more
destructive, or disruptive manner in a social setting on the Internet straightforward.
with no apparent instrumental purpose. From a lay-perspective, Frequency of activity is an important correlate of antisocial uses
Internet trolls share many characteristics of the classic Joker vil- of technology. For instance, cyber-bullying is often perpetrated by
lain: a modern variant of the Trickster archetype from ancient folk- heavy Internet users (Juvonen & Gross, 2008), and disagreeable
lore (Hyde, 1998). Much like the Joker, trolls operate as agents of persons use mobile technologies more than others – not for social-
chaos on the Internet, exploiting ‘‘hot-button issues’’ to make users izing, but for personal entertainment (Phillips & Butt, 2006). Simi-
appear overly emotional or foolish in some manner. If an unfortu- larly, gamers who express non-social motivations for online
nate person falls into their trap, trolling intensifies for further, mer- gaming (e.g., competition, personal achievement) demonstrate
ciless amusement. This is why novice Internet users are routinely lower levels of agreeableness and conscientiousness than others
admonished, ‘‘Do not feed the trolls!’’. (Graham & Gosling, 2013). Still other research has linked low
Despite public awareness of the phenomenon, there is little agreeableness, low conscientiousness, and high extraversion to
empirical research on trolling. Existing literatures are scattered heavy Internet use (Andreassen, Griffiths, & Gjertsen, 2013). These
and multidisciplinary in nature (Hardaker, 2010; Herring, Job- patterns parallel gender differences in online behavior: Men are
Sluder, & Scheckler, 2002; McCosker, in press; Shachaf & Hara, higher in overall Internet use (Joiner, Gavin, & Duffield, 2005)
2010). For instance, Shachaf and Hara (2010) conducted interviews and higher in antisocial behavior online (Zweig, Dank, Yahner, &
of Wikipedia trolls, finding themes of boredom, attention seeking, Lachman, 2013). Overall, the findings suggest that it may be fruit-
revenge, pleasure, and a desire to cause damage to the community ful to examine associations of trolling with the Big Five, gender dif-
among their expressed motivations for trolling. In other research, ferences, and global Internet habits.
Hardaker (2010) conducted a content analysis of Usenet posts that The noxious personality variables known as the Dark Tetrad of
identified four primary characteristics of trolling: aggression, personality – narcissism, Machiavellianism, psychopathy, and
deception, disruption, and success. The deceptive and ‘‘pointless’’ sadistic personality (Buckels, Jones, & Paulhus, 2013; Furnham,
Richards, & Paulhus, 2013) – are yet to be investigated in the troll-
ing literature. Their relevance is suggested by research linking
⇑ Corresponding author. Address: Department of Psychology, University of
these traits to bullying in both adolescents (Fanti & Kimonis,
Manitoba, 190 Dysart Rd., Winnipeg, MB, R3T 2N2, Canada. Tel.: +1 204 977 8445.
E-mail address: (E.E. Buckels).
2013) and adults (Baughman, Dearing, Giammarco, & Vernon,
0191-8869/Ó 2014 Elsevier Ltd. All rights reserved.

50) for their time. embedded in a larger battery of personality questionnaires.. Davies. t(500) = 2.88.g. Linton & Power. Because of low endorsement rates of SD = 11.01.83). we constructed a second brief index: the Global neuroticism (a = . Rab.001. We further expected Commenting time was also negatively associated with age. but sess direct sadism.61) and seven did not differ on conscientiousness. p = .19. 2 The pattern of results was unchanged when all categories were used. and the scores were highly correlated in this sample (r = . 2. p < .65.. A sec- tendency to enjoy hurting others (e. in press). rated on five-point scales from 1 (strongly disagree) to 5 expected.005. O’Meara. r(508) = . d = . sues that are important to you’’. We expected that the Dark Tetrad would be positively r(508) = .21. Participants and procedure and 5. rated on seven-point scales ranging from 1 (not at all) to 7 (very much). ‘‘Pay- out the possibility that overall Internet use explains relations with back needs to be quick and nasty’’. p = . SD = 0.1% said they especially enjoy making friends.04.001..12.09.72 h of news sites.)?’’ A second question probed their preferred commenting per day. the Dark Tetrad scores were highest among those who selected trolling as the most enjoyable activity.1.8% specified another activity.97.03. Hence. In contrast. ond multivariate analysis on the Big Five scores indicated that. psychopathy. activity on ‘‘troll-able’’ websites (defined here as websites that per.8% of participants expressed a preference for 2.16.03. which assessed trolling Finally.E. Jones & Paulhus. p = . the mean number of commenting hours questionnaires to student and community samples in two online per day was 1. and subclinical psychopathy (e. Measures Tetrad measures: direct sadism. Carrier.85).2. we applied a square root per day do you spend posting comments on websites (e. ‘‘trolling other users’’. p’s > . istic blood spurts’’.1 Commenting time was associated with studies. lower conscientiousness scores.37. conscientiousness (a = . d = . 1999) was used to assess extraver- for use as a control variable.26. 2013) and those with antisocial personality disorder (Rosen. & Cheever. transformation to the raw scores. narcissism. p = . r(512) = . ‘‘I enjoy hurting people’’. ipants received monetary compensation ($0. or openness.87). 2..79). p < .1. F(20.002. & t(500) = 3. ‘‘Hurting people is exciting’’. of the first four answer options was randomized. neuroticism. Method debating issues. trolling culture embraces a concept virtually synonymous with sadistic pleasure: in troll-speak. psychop- associated with a tendency to rate trolling as the most favored athy.98 E.2. (e. p’s > . as a = . d = 0. t(500) = 2.. 2013).79).91. 2013) use Facebook more fre. 2013) to prove most germane to trolling.4% female.25.006.23. Partic.24.001. we focused on predicting enjoyment of trolling. a = . we expected everyday sadism (Buckels et al.04. with a SD = 0. vicarious sadism.91) = 6.’’ 2.69). The VAST direct sadism subscale 3. p < . ‘‘I have been compared Hence in Study 2. Online commenting frequency To evaluate the predictors of Internet trolling.009. ‘‘In video games. The 44-item Big Five trolling. to triangulate on trolling with sion (a = . 2010.0) from Amazon’s Mechanical Turk website (http:// the ‘‘making friends’’ option. To rule tell your secrets’’. Finally. The key A multivariate analysis on the Dark Tetrad revealed a significant questions regarding trolling and other online behaviors were effect of activity preference: Wilks’ k = 0.g. forums. agreeableness. we combined2 that category with the www. we administered Across all participants. 5.16. press) was used to assess narcissism (e. r(512) = . Planned orthogonal contrasts indicated that the effect was significant for all Dark 2. that extraversion and agreeableness – the Big Five dimensions r(508) = . p = . SD = 1.04. in egorical index of trolling that likely underestimated the effects. chatting.88).3% of partic- We recruited 418 participants (42. containing 10 items to assess a dispositional d = . and ‘‘other (specify)’’. YouTube. was the Short Sadistic Impulse Scale (SSIS.. t(413) = 2. Locke. p < . neuroticism. narcissism. .2%. p = . d = . / Personality and Individual Differences 67 (2014) 97–102 2012. ipants were non-commenters. such as debating and scores on all Dark Tetrad measures except narcissism: direct sadism. than others. and openness to experience (a = . d = . The order Ross. In Study 1. and men reported greater numbers of hours most relevant to the Dark Tetrad’s social world (Paulhus & Wil.78) Assessment of Internet Trolling (GAIT) scale. p = .com) to complete survey questions online. containing six items to as. Machiavellianism (e.1. t(500) = 3. was restricted to respondents from the United States. First vicarious sadism. p = .78. 2. Results After all. 2.02.64. p = . on these comment sites?’’ with five response options: ‘‘debating is. The sample ‘‘other’’ category in the following analyses.55. r(418) = .1. Study 1 2. t(500) = 2. pants who indicated that they did not spend any time posting com- quently than others. a = . posting comments (M = Orr. Also sugges. Two measures of sadistic personality were administered.3% preferred chatting. Hammond. p = . items to assess vicarious sadism (e. 1646. a = . liams. and Machiavellianism.21.77. ‘‘making tive is research showing that narcissists (Ljepava. we also included a question about total time spent online Inventory (John & Srivastava.20. Those partici- Whaling. thus indicating that dark personalities leave ments were labeled as ‘‘non-commenters.04. M age = 29. & new friends’’. d = . p < . ‘‘chatting with other users’’. Study 2 is conceptually equivalent to the SSIS. participants who chose trolling as their favorite activity (strongly agree). The transformed scores were used in the analyses activity when commenting online: ‘‘What do you enjoy doing most that follow. and Machiavellianism. mit users to interact by posting comments).00) = 1. multiple measures.2. Paulhus & Jones. and narcissism were all non-sig- personality traits.. we assessed enjoyment of each commenting to famous people’’. a = . ‘‘It’s not wise to activity (including trolling) on separate continuous scales.001). their favorite activity from a list of options. Favored activity when commenting A total of 23. 1 confirmed that.g. and higher as opposed to other online social activities. Second was the Varieties of Sadistic Tendencies were higher on extraversion. p < . r(512) = . ‘‘lulz. The remaining 41. nificant predictors of commenting frequency. t(413) = 2. t(505.2.72).27. extraver- preference for trolling indicative of extraverted but disagreeable sion.6% reported enjoying trolling other users. I like the real. Jones & Paulhus. Inspection of the pattern depicted in Fig.80). 21. 2011). as expected.g.g.. etc. the section on Internet behavior asked participants to estimate their overall commenting frequency: ‘‘How many hours 1 Because the frequency scores were positively skewed. This necessitated a cat- The 27-item Short Dark Triad scale (SD3. resulting in a transformed mean of 0. and low- scale (VAST.78) than did women (M = 0.mturk.1.62). d = . 2002) – would evidence associations with trolling. p = .g. a = . SD = 0.g.18.2. agreeableness (a = .73.001.’’ large digital footprints. er on agreeableness. Buckels et al. we standardized A limitation of Study 1 is that we asked participants to select and summed them to create a direct sadism composite score. Of the Dark Tetrad.

seven items. and 52% female. boasting expanded content coverage frequency.89.80). For reasons The section on Internet use asked participants. M age = 36. M age = 35. ‘‘I enjoy griefing other players in multiplayer games’’. ‘‘I like to troll people in forums or the comments section of short trolling scale.g. Coefficient al- ple. enjoyment of each activity used in Study 1 (debating.E.79). we menting frequency. etc. ally responded to a yes/no question. YouTube. The first three items addressed trolling experience and enjoying various forms of trolling.81). a = .2 0 -0.79). Method ior: ‘‘How many hours per day do you spend posting comments on websites (e. forums.98) recruited on Amazon’s Mechan. Error bars represent standard errors.g.g. and sub. the commenters) were asked to provide additional information about their posting behav- 3. To assess sadistic personality. controlling for overall Internet use.1.81). Although all spersed in the other measures: ‘‘I have sent people to shock websites participants completed the dark personality measures and the for the lulz’’. here 5-point (e. Results The 27-item Short Dark Triad scale (Jones & Paulhus. and rated enjoyment of various activities – used the 18-item Comprehensive Assessment of Sadistic Tenden.6 Direct Sadism 0. news sites.. ‘‘I enjoy playing the villain in 4 This statement is from the evolving (and often trolled) Rules of the Internet (see games and torturing other characters’’.. Buckels et al. hours per day do you spend on the Internet?’’ Participants addition- fects on trolling. and Machiavellianism. the mean of the 11 physical and verbal items (a = .63) who completed the questionnaires for extra course cred. YouTube.83). used to assess extraversion (a = . to 7 (very enjoyable). chatting.04. com- clinical psychopathy (a = .4) completed the questionnaire ‘‘The more beautiful and pure a thing is. ‘‘How many articulated earlier. measures of were rated on 5-point scales ranging from 1 (strongly disagree) to the Big Five were included for comparison.g. 2013). Participants and procedure commenting hours on overall Internet hours and saved the residual Two large samples were collected online. only a subset of MTurk participants (N = 207.1. The second consisted of 609 United States residents (43% trolling..1.78). / Personality and Individual Differences 67 (2014) 97–102 99 1 Machiavellianism Mean Personality Score (Standardized) 0. six items.6 Favorite Activity when Posting Comments Online Fig. identification. ‘‘I enjoy making jokes at the expense of others’’. 1. Table 1 displays correlations between the Dark Tetrad.75). agreeableness (a = . the more satisfying it is to cor- on commenting behavior. forums. http://www.e. behavior.80). a = .g.14) did not affect ties of sadistic tendencies: direct physical sadism (e. Global Assessment of Internet Trolling (GAIT). .4 Vicarious Sadism 0. 3. and enjoyment. ‘‘Do you post comments on ism and trolling would remain significant even when controlling websites (e. websites’’. and openness to experience (a = . in press) assessed narcissism (a = . The CAST assesses three distinct varie. including trolling – while posting comments online. E. and making friends) on scales from 1 (not at all enjoyable) female. M age = 21. neuroticism (a = . As in Study 1. SD = 3. Dark Tetrad scores as a function of favorite online activity in Study 1. direct verbal sadism 3 Responses to the CAST are commonly collected on 7-point scales. Buckels & Paulhus.)?’’ We regressed 3.61.. and vicarious sadism (e. a = . 5 (strongly agree)..50 each for their time.3 We computed a composite for direct sadism as Study 2 also featured data from a larger and more diverse sam.4 rated on 5-point scales from 1 (strongly disagree) to 5 the Big Five Inventory.. Finally.4 (Non-Commenter) Debating Issues Chatting Trolling Other -0. that controlling for Internet use (M = 6. ‘‘I enjoy physically hurting people’’. news sites. Thus we predicted that the relations between sad. furnishing us with enough statistical power to test hypotheses pha for CAST total scores was . Mean responses The 44-item Big Five Inventory (John & Srivastava.82) formed the composite score labeled.mturk. commenters rated their SD = 3.2. Those who answered ‘‘yes’’ (i.1.81). con.15. it points. (strongly agree). Items http://rulesoftheinternet. Machiavellianism (a = . We also included four items relevant to trolling that were inter- ticipants received $0. As expected.2. SD = 12. The first consisted of scores to create a second index of commenting frequency when 188 Canadian psychology students (55% female.86). rating scales were used to stay consistent with the SD3. about the unique contributions of the Dark Tetrad. 1999) was to these four items (a = .80). and only the student sample completed rupt’’. five items.)? (even occasion- for overlap with psychopathy. Partial correlations (also displayed in Table 1) indicated and improved reliabilities.8 Narcissism Psychopathy 0. Measures identification with trolling and Internet subcultures.2 -0. we expected sadism to dominate personality ef. while the last item addressed ally?)’’. The latter par. cies (CAST. scientiousness (a = . SD = 13. ical Turk website (MTurk.1. etc. The CAST item set succeeds Dark Tetrad scores were positively correlated with commenting the VAST used in Study 1.

001 . in contrast. p < . . men had stronger GAIT It remained significant even when controlling for scores on the scores (M = 1.40. enjoyment. indirect effect (constructed with 10. Recall that.75. enjoyment was not significant. p = . were strongly associated with commenters’ trolling enjoyment.55). The di- t(530) = 7.56⁄⁄⁄ Vicarious Sadism . p = .44⁄⁄⁄ . Scores on extraversion and neuroticism were tion. p = .14 . p < .23⁄ .47. they displayed high levels of the Dark Tetrad scores.48.92) than did women (M = 1.40⁄⁄⁄ . ⁄⁄p < .02 . was unaffected by controlling for overall Internet use.001.53⁄⁄⁄ . ment.31. r(78) = . General discussion trolling enjoyment. b = 0.55⁄⁄⁄ Note. p < . It was sadism.004.001).25]. p < . p < .21.23⁄ .001).54⁄⁄⁄ .001.2. t(78) = 4.25⁄ . 5 The patterns of associations were comparable across the student and community p = .10. 95% CI = [0. in Quadrant II of the Interpersonal Circumplex (Wiggins.33⁄⁄⁄ . t(78) = 2. 1995).03. p = .20 . a = . An alternative ginally higher in openness (r = . GAIT scores distribution). p < . we ran a mediation analysis to examine if.09 . 95% CI = [ 0. Results were identical when omitting GAIT items overlapping with enjoyment.24. even when controlling for scores on the Dark Triad mea- (M = 0.46.27⁄ . indicating partial mediation.21.13.18⁄⁄⁄ Psychopathy . also significant.63.55⁄⁄⁄ Machiavellianism . 0.000 re-samples and a percentile Turning next to scores on the short trolling index.37 (see Table 1).71. but was instead positively correlated with ship between sadism and GAIT scores. ⁄⁄⁄p < .35⁄⁄⁄ . the associations between sadism and GAIT scores were so We ran an identical analysis for scores on the GAIT scale with data from the full sample.10. GAIT was positively associated with associated with rated trolling enjoyment. 95% CI for the standardized indirect effect = [0. b = 0. Personality scale Commenting frequency Commenting frequency (controlling Rated enjoyment: GAIT scores (h/day) for overall Internet use) Trolling Debating Chatting Making friends Sadism Total . Significance was tested with both Sobel’s while enjoyment of chatting was negatively correlated with test and a bootstrapped 95% confidence interval for the standardized psychopathy.43⁄⁄⁄ . the standardized indirect effect of GAIT on sadism via trolling unrelated. and identity. Participants with stronger GAIT scores rect effect of sadism was substantially reduced.100 E.10 . r(81) = . but remained sig- tended to be younger (r = .21 . positively associ- strongly with sadism (r’s > .09.74.19 .6 In other words.30⁄⁄ . t(734) = 2. as compared to their mediation analysis found no support for the opposite causal direc- low GAIT counterparts.28.001.27. were unique predictors of trolling that is. In other words.001.82. Mediation via trolling enjoyment scores were positively correlated with self-reported enjoyment of Because the associations between sadism and trolling were par- trolling. The present research was the first to examine comprehensive Machiavellianism.18. SD = 0.37⁄⁄⁄ . Also as expected.10. Sobel’s z = 3.61] with a BCa correction. GAIT scores ated with GAIT scores when controlling for sadism. GAIT scores were a composite of four items capturing trolling behavior. Narcissism.35⁄⁄⁄ .06 . p < . 6 sadism predicted stronger GAIT scores.41.45].01. SD = 0.65). narcissism was negatively associated with trolling enjoy- traits and a BFI profile consistent with those traits. all r’s > .53.001.001. p = .52⁄⁄⁄ . and psychopathy were entered as personality profiles of Internet trolls. b = 0.50. ⁄p < . p’s > .15 .62).001.78. and remained so when controlling for overall Internet use.35. r(78) = . when controlling for the other Dark Tetrad 2012).30. rated enjoyment of trolling explained the relation- with trolling enjoyment. Across all participants.43.001. sadism scores were strongly r(83) = . A regression analysis scores of all Dark Tetrad measures (see Table 1). p < . d = .08).12 . the personality projections of trolls emerged ism. Unique contributions of the Dark Tetrad to trolling To examine the unique contributions of the Dark Tetrad on 4. This analysis indicated that sad- measures of trolling.43. p < . but was positively associated with trolling of GAIT scores. In fact.23⁄ .14 .51]. lower in conscientiousness nificant when controlling for trolling enjoyment.1. Buckels et al. Total sadism.39⁄⁄⁄ . (r = .01. The pattern of association any of the personality measures.13 .001.10 . Dark Triad. psychopathy. we conducted a multiple regression analysis with data from the subsample of commenters. narcissism.68⁄⁄⁄ Direct Sadism . including those of the Big Five. b = and especially indicated that enjoyment of trolling was. p < . we sought enjoying debating issues important to them.34⁄⁄ .32⁄⁄ . SD = 0. and narcissism were not significant.30.40⁄⁄⁄ . p < .33⁄⁄ . among Internet use. / Personality and Individual Differences 67 (2014) 97–102 Table 1 Associations of Dark Tetrad variables with commenting behavior and enjoyment in Study 2. As was the case for trolling enjoyment.07 .07. even when controlling for overall ticularly strong. t(78) = 3.2.21 .65⁄⁄⁄ Direct Physical .23.2. and Machiavellianism.21 . Sobel’s z = 4. Psychopathy was also a unique (though weaker) predictor use.89. sadism. t(80) = 6.19 . Vicarious sadism to test the hypothesis that sadism leads to trolling because those was similarly positively correlated with enjoyment of debating. r(80) = .001.29. 0. and psychopathy was however. and Machiavellianism 3.33⁄⁄ .83) was positively correlated with overall Internet sures.001) and agreeableness (r = .34.03 .001. p’s > .23.23⁄ .E.36⁄⁄⁄ .39⁄⁄⁄ .002. p = . p = . p < . b = 0. The direct effect was samples. enjoyment. Among commenters. t(735) = 15. behaviors are pleasurable. The mediated effect of sadism through enjoy- p < .90. was not correlated commenters. and mar- t(80) = 5. 0.09.34⁄⁄⁄ Narcissism .07 . b = 0.39. Among commenters. commenting frequency p < . that had the most robust associations with trolling of unrelated to trolling enjoyment. Across two studies and two predictors of trolling enjoyment. The impact of sadism on troll identity/behavior was mediated by enjoyment: Sobel’s z = 2. 0. 95% CI = [0. 3.30].38⁄⁄⁄ . Critically.05.61.04 . 95% CI = [0. In contrast. all r’s > . ment was significant. leaving a ’’pure’’ measure of troll identity and behavior (2 items. All tests are two-tailed. among commenters. 0. High Agency and Low Communion (Jones & Paulhus.44.03. Sobel’s z = 0. b = 0.62⁄⁄⁄ Direct Verbal .17 . those associations. were positively correlated with commenting frequency. in turn. Machiavellianism enjoyment even when controlling for overall Internet use.

6. V. D. Computers in Human Behavior.. M. C. D. Measuring dark personalities. Boyle.. when controlling for sadism and the other Dark ‘‘bullies’’.org/10. and love to hate: the cyber. N. however idiosyncratic.. predicted trolling on both measures (trolling enjoyment and GAIT http://dx. measure of dark personality traits. 4. 0165551510365390.. A. & Gosling. 12–18. Personality profiles associated with different In the final analysis of Study 2. H. (2013). D. J.. Enjoyment of other online Buckels. The ideal elf: Identity exploration in Shachaf. References Paulhus. & Paulhus. Journal of World of Warcraft.doi. (2014). S.doi. 189–193. (2008). 2011. & Hara. Our findings add to accumulating evidence linking excessive Joiner. Carrier. Rest is not idleness: Implications of the brain’s default mode for human development and education. In L.. 16. sadism’s impact on trolling was cut nearly in half. technology use to antisociality (Carr.1556/JBA. K.9994.doi.12018. Norton & Company. Sadists just want to have fun . R. http://dx. Griffiths.1016/j. the case. Given that controlling for overall Internet use Furnham. Whaling. Davies. only sadism predictors of avatar-self discrepancy.1037/a0022400.1515/JPLR. John.doi. Unpublished measure. San Diego. Cyberpsychology. Social Psychological and Personality Science. K. http://dx. A. http://dx. Individual Differences. CyberPsychology & Behavior. N..2012. Behavior. From user discussions to academic definitions. Schrader.. Differentiating the dark triad within the ther to argue that use of Internet technology actually shifts users in interpersonal circumplex. 2004). Fanti. CyberPsychology & Behavior.0090.paid. Locke.. A.. Computers in Human Behavior. New York: Farrar Straus & Giroux.chb. C. & Hyland. Do antisocial persons use technol. Social and Personality Compass. reflecting both actual personality (Dunn & Guadag. Trickster makes this world.00335. 2013). 571–575. we found clear evidence that sa. K. trolling enjoyment. and ‘‘victims’’. J.. & Paulhus. pointing to high levels of consistency among the mea. 102–139).1016/S0092-6566(02)00505-6. Giammarco. and the psychology of embodiment. 2008. (2010).2012. Comprehensive assessment of sadistic Pervin (Eds. S. 2007).chb.11. 29. (2013). & Williams. Different provocations trigger aggression in motives? The findings of this study suggest that this is indeed narcissists and psychopaths.1016/j. 215– 78. L.1111/spc3. D. (2002). 1243–1254. / Personality and Individual Differences 67 (2014) 97–102 101 strong that it might be said that online trolls are prototypical Buckels.2013. (2012). S. & Singh.doi.12.003. & Singh. dencies of Internet use and familiarity cannot explain the findings. personalities.1177/ dx.doi. . N. Boduszek. (2002). Dearing. for those with sadistic self: Personality. and the indirect effect of sadism through enjoyment was substan.. & Paulhus. N. & Hammond.chb. Lenhardt.. CyberPsychology & Behavior.1746-1561. R. E. Immordino-Yang.doi. Study 2 also found strong positive relations among online com. 2012) and one’s ideal self (Bessière. M.. Psycho-social perspective. The troll persona appears to be a malicious case of a vir. L.08. L. seek out and explore one’s niche. T. 556–563. Behavior. Some authors go fur. Is Facebook between bullying behaviours and the dark triad: A study with adults.753.1037/a0033951. Herring.1089/ Behavioral Addictions. When controlling for Hardaker. 10. The dark triad of personality: Narcissism. 27–40. J.. A. Gender. S. 23. http://dx. R. 357–370. 249–267). In contrast. E. (2014). Suler. S... . Jones. University of British Columbia. K. A.doi. ‘‘bully-victims’’. Juvonen & Gross.). S. S.1089/cyber. D. E. C. (2013). Hyde.2011. 976–983.1089/cpb. Graham. Introducing the Short Dark Triad (SD3): A brief these associations is yet unclear. .2011. 52.doi. The Information Society. Hoboken. 97–106. Journal of for games. and remained significant when controlling for Managing‘‘ trolling’’ in a feminist forum. Washington. 36. Seay. http://dx. C. F.. D. Saklofske & G. smartphones & texting. Human & Boone. 8. J. Machiavellianism. Defining the virtual & Kiesler. sures. 396–409.020. & Paulhus. (2013).1177/1073191113514105. and theoretical perspectives. (2011).1016/ search on trolling. 2011. (2010). Strack (Eds.. L. http://dx. E. S.doi. 9. Matthews (Eds. Relationships Rosen.doi..2005. & Gross.1016/j. K. http://dx. (in press).2. D. such as chatting and debating. and the Internet is their Perspectives on Psychological Science. Gavin. motivations for playing World of Warcraft. Tetrad measures.0090.015.. The shallows: What the Internet is doing to our brains. cpb.. (2013). J. J. significant.2010.doi. 2012. 2011. & Cheever.x. The Dark Triad of personality: A 10-year review. http://dx. D. L. Jones. D. Searching for safety online: tial. often shared by their victims: Is there a dark side to victims? Personality and Online identity construction may be important to examine in re. Orr. (2013).1016/ lou. http://dx. (2012). 7. narcissism was actually negatively related to http://dx. D.... CyberPsychology and Behavior. associations were specific to (2011). 2013). Buckels et al.. In G. 2201–2209. attitudes and anxiety. 753–758. 738–743.. everyday sadists (Buckels et al. & Paulhus. ogy more than others because it facilitates their nefarious goals/ Jones. and Internet anxiety: Correlates of Internet use. R.doi. S. R. antisocial individuals have greater opportunities to con. S. N. http://dx. 21.019. L. Press. A.1016/ disorders and technology use. Psychology of Violence. K. Computers in behavior. Journal of School Health.. Christodou. Personality and social uszek & minary glimpse into the mechanism by which sadism fosters troll. (2013). Walters. Carr. John & L. The personality traits of workplace bullies are ‘‘self expression’’ than they did before the advent of the Paulhus.2013. & Guadagno. Conse. Trolling as provocation: YouTube’s agonistic publics. P. 1. & Kimonis. but more empirical work is needed. O... personality differences in broader ten. ing behavior. (2010).org/10.doi..2012. 371–384. http://dx. R. Both trolls and sadists feel sadistic glee at the distress Immordino-Yang Christodoulou. http:// dx. Rab. (2006).011. and Social Networking. http://dx. overlap with the Dark Triad scores.2007. P. D. 29. 199– Assessment.. (2011). 530–535. (1998). (2013). Internet identification.doi. Baughman. 24. P. L.. The causal direction of Jones.2012. Trolling in asynchronous computer-mediated communication: enjoyment. (2013). M.. & Power. McCreery. Center’s Internet & American Life Project. 28. tions were largely due to overlap with sadism. behavior.11.026. D. doi:1354856513501413. Rosen et al. Psychological Science. Beyond vandalism: Wikipedia trolls. and to pursue their personal brand of pewinternet. tendencies (CAST).2011. Personality and self-reported use of mobile phones behavioral addictions and the five-factor model of personality. E. Schrader. New Extending the school grounds? Bullying The Internet is an anonymous environment where it is easy to experiences in cyberspace. Linton. envy. http://dx. 10. (1999).). S. no. 2006. D. (2012). that ideal self may be a villain of chaos and mayhem O’Meara. NY: W... Note that the Dark Tetrad 10. These findings provide a preli. L. Phillips & Butt. Andreassen. Psychological Assessment. 3.. Ljepava.. Convergence: The International Journal of Research into New Media Technologies Boone. 90–99. Horowitz & S. trolling enjoyment. E. (2012). M. H. In O. J. A. 523–531. http://dx.1111/j. troll. International Journal of Criminology and Sociological 321–326. was unrelated to sadism. The online disinhibition effect. Job-Sluder. measurement. http:// Information Science. 371–378. New York: Guilford identity.doi.. K. 2012). D. A. Behavioral confirmation of everyday sadism. The relationships between Phillips.9. G. interpersonal theory and research (pp. & Scheckler. N.. and trolling behavior and Handbook of personality: Theory and research (pp. creating ‘‘iDisorders?’’ The link between clinical symptoms of psychiatric Personality and Individual Differences. 7. & Srivastava. 2012.01. P. menting frequency. L. (2007). http://dx. D. E. http://dx. j. The psychometric properties and – the online Trickster we fear. Handbook of an antisocial direction (Carr.1089/cyber. of others. 54. A. http://dx. 7. Subsequent analyses confirmed that the Dark Tetrad associa. tual avatar (Dunn & Guadagno. Journal of Politeness Research. CA: Academic Press (in press). N. N. Computers in Human j. Theory. N. A. R. did not affect these results. utility of the short sadistic impulse scale (SSIS). & Paulhus. 10. M. (2012). P. L. Our research suggests that. & Jones. L.chb. F. McCreery. & Duffield. P. M. D. 496–505. 18. McCreery et al. J.1177/0956797613490749. Dimensions of juvenile psychopathy distinguish scores). D. The Big-Five trait taxonomy: History.026. & Gjertsen. http://dx. L.2006.. The theoretical model of criminal social identity: Suler. R. My avatar and me – Gender and personality contributions were assessed in a multiple regression.. R. Bessière. DC: Pew Research quently. http://dx. 28. & McCosker.2008. S.doi. (2014). and psychopathy.1177/1948550609347591. P. E. Krach. & Kiesler. 352–364.doi. NJ: Wiley & Sons.doi.doi. D. (2004).1089/cpb. 36. Retrieved from <http:// nect with similar others. & Vernon.aspx>.8. (2013). Journal of Research in & Ross. 1602–1607.. Measures of personality and social psychological constructs. 2012. When their unique Dunn.1089/1094931041291295. http://dx. (2005). Krach. L. 604–615. Seay.. dists tend to troll because they enjoy it. A. especially in terms of antisocial identity (Bod.012.doi.).doi.1177/ playground! 1745691612447308. & Butt.371. 2007) and its role in trolling characteristics of Facebook non-users and frequent

Wiggins. (1995). FL: Psychological Assessment Resources. Buckels et al.. J. J. & Lachman. Criminal Justice and P.doi. Yahner. . The rate of cyber dating analysis of the lifestyle criminality screening form and psychological inventory abuse among teens and how it relates to other forms of teen dating violence. of criminal thinking styles. s10964-013-9922-8. 42. J. Odessa. (2007).doi. D. Interpersonal Adjective Scales: Professional Manual. / Personality and Individual Differences 67 (2014) 97–102 Walters.1177/0093854807307028.E. Dank.1007/ dx. The latent structure of the criminal lifestyle: A taxometric Zweig. G. M. http:// Journal of Youth and Adolescence. 34.102 E. S. M. (2013). 1063–1077. 1623–1637..