A study by researchers at the University of Vienna finds that the attractiveness of online misogyny perpetrators does not influence how bystanders judge the posts. When the content is clearly hostile or abusive, people rely on the message itself rather than any appearance cue.
The work, published in Computers in Human Behavior, explored whether a troll’s physical appearance could sway readers’ judgments of misogynistic comments and how the wording of the hostility might shift responses.
The researchers frame their analysis in terms of the Elaboration Likelihood Model, a theory that distinguishes between central processing of information and peripheral processing that relies on cues such as a speaker’s looks when motivation or time are limited.
They note the attractiveness-leniency effect, a well-documented bias where physically attractive individuals receive softer moral judgments after wrongdoing, and question whether this bias extends to perpetrators of online misogyny in visually driven online environments.
The study recruited 884 adults in Austria to participate in a controlled online experiment. Participants viewed three mock social media profiles featuring male faces that had been categorised beforehand as highly, moderately, or poorly attractive.
Following the profile views, participants read three misogynistic comments attributed to those men. The comments were pre-selected by independent experts to represent incivility, intolerance, or explicit threats. After reading the material, participants rated each comment’s acceptability and its level of sexism, and reported their negative emotional reactions, support for platform moderation, and their likelihood of intervening.
Two forms of intervention were measured: indirect interventions, such as reporting or flagging a post, and direct interventions, including responding to the perpetrator or publicly supporting the victim.
The results showed a clear progression in moral condemnation. Participants judged intolerance and threats as less acceptable and more sexist than incivility, with a ceiling effect appearing between the two more severe categories in cognitive evaluations.
Emotional responses, however, rose steadily with the severity of the comments. Anger and disgust were higher for intolerance and highest when readers encountered direct threats.
In terms of action, there was a split between private and public responses. Indirect interventions became more likely as the hate escalated, and participants expressed greater support for platform moderation when facing threats. Direct interventions, by contrast, remained consistently low across all comment types.
The researchers highlight an asymmetry between moral condemnation and visible action, suggesting reputational and emotional risks associated with public confrontation help explain why people resist direct involvement.
Crucially, the study found that perpetrator attractiveness had no effect on any outcome. Responses did not differ whether the profile picture showed a highly attractive or a highly unattractive man, and the attractiveness result was not statistically significant.
The authors propose that when a moral violation is highly explicit, readers may engage in more central, systematic processing, allowing the content to override any superficial leniency typically afforded by physical attractiveness.
Limitations of the study include its use of static text and images rather than a fully interactive social media environment, and the focus on male faces with hostility directed at women. The authors note that reactions could differ if comments targeted other groups or if the perpetrator’s gender were varied.
The study, titled “Pretty privilege? Effects of perpetrators’ attractiveness and hate type on user responses against misogynist digital hate,” was authored by Rinat Meerson, Kevin Koban, and Jörg Matthes.
