Skip to main content

Verified by Psychology Today


New Study: Most Trolls Have Subclinical Sadism

Plus, how to minimize the harm they can cause.

Key points

  • Sadism is a new addition to the Dark Triad of narcissism, psychopathy, and Machiavellianism. So, it extends the Dark Triad to the Dark Tetrad.
  • Sadism refers to inflicting pain and suffering for the sake of dominating others and experiencing pleasure and enjoyment.
  • A systematic review of the link between everyday sadism and aggression finds that sadism is related to aggressive behavior, especially trolling.
Source: gracinistudios/Pixabay

A recent paper, published in the July-August issue of Aggression and Violent Behavior, reports the first ever systematic review of the link between everyday sadism and aggression, including trolling behavior.

Before discussing the study, I define sadism and explain how it relates to the Dark Triad and the Dark Tetrad.

Sadism and the Dark Triad and the Dark Tetrad

The personality traits of narcissism, psychopathy and Machiavellianism are collectively known as the Dark Triad. These traits share several characteristics in common. For instance, people high on these traits tend to be manipulative, deceitful, vain, callous, and aggressive.

Recently, some researchers have introduced the term Dark Tetrad. Specifically, they have added to the Dark Triad—of narcissism, psychopathy, and Machiavellianism—the trait of sadism.

Sadism refers to inflicting harm, cruelty, humiliation, suffering, or pain (physical, sexual, psychological), all for the sake of dominating others and experiencing pleasure and enjoyment.

Those with stronger sadistic tendencies engage in sadistic sexual behavior or even criminal acts (e.g., torture, animal abuse).

Those with subclinical sadism—sometimes referred to as everyday sadism—may express their cruelty in more socially acceptable ways, such as by playing violent games, watching disturbing movies, playing pranks, engaging in meddling, etc. But in what other ways might these individuals express their hostility?

For an answer, I discuss the findings of the systematic review and meta-analysis by Egan and Thomas.

Investigating Sadism and Aggression

Of the 627 records identified, 50 studies were included in the data synthesis (three in the narrative synthesis and 48 in the meta-analysis).

The studies used a cross-sectional design, were published between 2013 and 2020, and used a mixed-gender sample (except for four female-only and five male-only samples).

The total number of participants was 22,179. Whereas 26 investigations used community samples, 20 had student-only samples. Most were conducted in the US and Europe.

Sadism was assessed using four measures (sample items in parentheses):

  • The Short Sadistic Impulse Scale (“I would enjoy hurting someone physically, sexually, or emotionally.”)
  • The Comprehensive Assessment of Sadistic Tendencies (“I love to watch YouTube clips of people fighting.”)
  • Varieties of Sadistic Tendencies (“I enjoy mocking losers to their face.”)
  • Assessment of Sadistic Personality (“I enjoy humiliating others.”)

A variety of measures were used to evaluate the outcome of aggression. Specifically, for aggression perpetrated offline, the measures were:

Aggressive humor; anger towards others; antisocial behavior; bullying; conflict within intimate relationships; hazing; proactive aggression; radicalized behavior; reactive aggression; same-sex aggression; sexual aggression (e.g., partner violence, rape, sexual coercion); trait aggression; and violence and criminal attitudes.

For aggression perpetrated online: Antisocial dating behavior; cyberbullying; cyberstalking; intimate partner cyberstalking; online antisocial behavior; online trolling behavior; and revenge porn proclivity.


There was a moderate effect size for the relationship between sadism and aggression offline (pooled correlation coefficient, r = 0.30, 95 percent CI = 0.29, 0.32) and aggression online (pooled correlation coefficient, r = 0.40, 95 percent CI = 0.38, 0.41).

A strong effect size was found for trait aggression; an intermediate effect size was found for anger and proactive aggression. This means as the authors note, sadists “may also aggress against innocent others without warning.”

Analysis of data showed, furthermore, a strong association between sadism and trolling (i.e., disrupting, upsetting, or harassing people online) and other types of cyber-aggression. This agrees with previous studies. In fact, the authors of a 2014 paper stated, “Online trolls are prototypical everyday sadists.”

Finally, the review found that everyday sadism is associated with a variety of aggressive behaviors, including sexual aggression, sexual coercion, intimate partner sexual violence, cyberstalking, and radicalized or extremist behavior.

Source: thetruthpreneur/Pixabay


Previous research shows high levels of sadism are associated with aggressive acts—from torturing insects and animals to severe bullying, cyberbullying, and hazing.

Subclinical sadism, which is more prevalent, is less likely to result in extremely aggressive behavior.

Nevertheless, the present review concluded that even low levels of sadism are associated with violence. So, given the “right context,” many sadistic people (particularly men) behave aggressively. Indeed, analysis of data showed there was “a moderate relationship...between subclinical sadism and aggressive behavior, as defined by acts ranging from verbal to physical, and sexual aggression and violence.”

Furthermore, this review highlighted an even stronger “quantitative relationship between sadism and trolling behavior.”

An important question is what motivates sadists’ aggressive behavior. A recent study of Internet trolls provides some answers. It suggests sadistic trolls experience pleasure from seeing people suffer, have a more positive reaction to harmful scenarios (e.g., someone being punished or morally wronged), and tend to minimize the harm their actions cause.

How Can We Reduce the Harmful Effects of Sadistic Behavior?

In terms of aggressive online behavior, moderators of websites and social networking sites may need to use better screening processes to identify people with a predisposition to engage in trolling or other aggressive behavior (e.g., online stalking or bullying).

Similar screening processes could be used in work organizations or colleges in order to prevent bullying and hazing or to protect those vulnerable to radicalization.

As for preventing sexual aggression or protecting the victims of intimate partner violence from further victimization, better knowledge of offenders’ sadistic tendencies may be helpful in predicting who will likely offend or re-offend and for designing interventions to reduce recidivism.

Facebook/LinkedIn image: SpeedKingz/Shutterstock