Violence

Violence is a possible concern for Virtual Reality. Because of presence and the immersive aspects of VR, many people worry that it would desensitize users to violence beyond the effects of TV, movies and video games. VR could cause certain users to become more disposed to violence. Many people also worry that users would become unable to differentiate reality and virtual reality and cause violence in real life.

VR games approximate violent behavior more closely than traditional games because users physically aim and fire virtual weapons. Laboratory studies have produced mixed answers. Some early experiments found that playing a violent game in an immersive system raised presence and aggressive feelings compared with a desktop screen, while later studies published in 2021 and 2022, including two pre-registered experiments with room-scale headsets, found no increase in hostile feelings or aggressive behavior from VR or from violent VR content. Researchers also use VR to study violence that would be unethical to stage for real, and to work with people convicted of violent offences.

Reviewed 27 September 2026. WHO definition, the Slater 2020, Madary and Metzinger 2016 and McIntosh and Allen 2024 quotations, study designs, sample sizes and findings of the aggression, desensitization, Milgram, bystander, offender and VRAPT papers and the 2026 review, Meta safety and IARC rating pages, the Steam content survey, and the UploadVR, CNN, PC Gamer and Ohio State quotations against their sources About review dates.

Definition

The World Health Organization's 2002 World report on violence and health defines violence as "the intentional use of physical force or power, threatened or actual, against oneself, another person, or against a group or community, that either results in or has a high likelihood of resulting in injury, death, psychological harm, maldevelopment or deprivation." In discussions of VR the word covers three different situations: violence depicted in games and films, violent acts that the user carries out against virtual characters, and hostile or abusive acts that other people direct at a user's avatar.

In their 2020 paper on the ethics of increasingly realistic VR and augmented reality, Mel Slater and co-authors summarized two arguments that Brey made in 1999 about why violence toward virtual characters raises moral questions. The first is that treating virtual characters badly may lead people to treat real people badly; the paper notes that this is a philosophical rather than an empirical argument. The second is that acting violently toward virtual characters may cause psychological harm to the people those characters represent. The authors add that in films "it is other people who treat other people badly whereas in XR it could be ourselves doing so", and that this can feel more real when the participant is fully embodied as the character.

Violence directed at users is a concern in social VR. A 2024 policy review by Verity McIntosh and Catherine Allen cites their earlier report on the prevalence of non-consensual touch and "simulated violence including sexual violence in social VR spaces", and argues that the immersive and embodied nature of these environments is likely to intensify the impact of harassment and abuse. Individual cases and the effects on victims are covered in the article on Psychological trauma.

Presence and embodied violence

Slater and colleagues argue that "presence, first-person experience and agency are very powerful cues to the brain that 'this is really happening,'" and conclude that "careful attention needs to be paid to the presentation of violence or abusive behavior in these contexts." Drummond and co-authors, in their 2021 study, point out that VR games "approximate violent behavior more closely than traditional games, for instance, with users physically aiming and firing virtual firearms."

Meta's health and safety warnings for the Meta Quest 3 tell users to "Choose your app and content carefully because virtual content can seem very realistic" and state that "Users may react to frightening, violent, disorienting, or anxiety-provoking content as if it were happening in the physical world." The section on children's use warns that "some apps can be intense, scary, violent, or anxiety-provoking."

Writing for PC Gamer in December 2018 about the Early Access release of the melee combat game Blade & Sorcery, Samuel Horti said he had never felt bad killing enemies in other games, partly because "it feels like it's my character, not me, who's dishing out the pain," and that "VR tears down that barrier by mapping your weapons directly to your hands and arms." He contrasted the game's human enemies, "with their changing expressions and dying screams", with the "cartoonish, exaggerated violence" of the VR melee game Gorn.

Desensitization

Desensitization is a possible outcome for extensive use of VR.

Desensitization occurs when the user becomes unaffected by acts of violence in real life. They may no longer show empathy towards the real victims of the violence. In media research, Carnagey, Anderson and Bushman defined physiological desensitization in 2007 as "showing less physiological arousal to violence in the real world after exposure to video game violence in the virtual world". In their experiment, participants who had played a violent video game for 20 minutes then showed lower heart rate and galvanic skin response while watching a 10-minute video of real-life violence. That study used conventional games such as Carmageddon, Duke Nukem and Mortal Kombat, not headsets. A later test of the hypothesis on a flat screen found no such effect: in a 2023 study in eLife, 89 male participants with little prior experience of violent games played either a highly violent or a non-violent modified version of Grand Theft Auto V over two weeks, and brain imaging and behavioral measures showed no reduction in empathy for pain and no desensitization to images of real-world violence. The authors added that their results "cannot preclude that effects could be found in settings with higher ecological validity, in vulnerable subpopulations, or after more extensive" play.

For VR, the concern is mostly stated as a risk rather than a measured effect. Slater and colleagues list "imitative behavior" among the possible harms of highly realistic VR: "it may be difficult for a person to carry out their first act of violence in XR, but eventually it becomes easy, and leads to a greater propensity for violence in reality". They also warn that unexpected horror could cause "a kind of post-traumatic stress response or, conversely, in desensitization for obscene sights". On realistic violent content in general, they write that its social consequences "seem to be more attributable to pornography and violence themselves and not so much to XR technology." For extreme depictions of physical or sexual assault, they state that whether such content would increase or decrease the behavior in real life "is not clear and is very difficult to assess experimentally." The same paper describes a "device-gap": putting on a headset marks a clear boundary between reality and VR, but that boundary may weaken with AR devices that people wear for long periods.

Research

Violent games and aggression

Most experiments on the topic compare the same violent game on an immersive display and on a conventional screen, then measure hostile feelings, aggressive thoughts and a laboratory measure of aggressive behavior. In the 2021 paper by Drummond and colleagues, the authors summarize meta-analyses finding that violent games explain "around 1% or less of variance in aggression", and note that researchers disagree about whether effects of that size matter.

Study	Setup	Participants	Main findings
Arriaga, Esteves, Carneiro and Monteiro (2008), Aggressive Behavior	Unreal Tournament (violent) or Motocross Madness (non-violent), played with a VR device or on a computer screen	148 undergraduate students	The violent game affected state hostility and aggression. Drummond and colleagues later grouped it with "early VR research" that "found no effect on aggression".
Persky and Blascovich (2008), Presence	Violent game on an immersive virtual environment platform or a less immersive desktop platform	Two experiments	The immersive platform led to more presence and more aggressive feelings and behavior; presence mediated the effect on aggressive feelings but not on aggressive behavior.
Lull and Bushman (2016), Psychology of Popular Media Culture	Grand Theft Auto IV played violently or non-violently for 15 minutes on a 17-inch 2D screen, a 96-inch 2D screen or a 96-inch 3D screen with 3D glasses (not a headset)	194 college students	Players in 3D felt more immersed, and that immersion was related to greater anger among those who played violently.
Drummond, Sauer, Ferguson, Cannon and Hall (2021), Journal of Experimental Social Psychology	Study 1: Arizona Sunshine (violent) or Portal Stories: VR (non-violent) on an HTC Vive, versus Call of Duty: Infinite Warfare zombie mode or Portal on a flat screen, 15 minutes; Study 2: Doom VFR in VR versus Doom on a flat screen	200 (Study 1) and 96 (Study 2)	VR produced slightly higher positive affect but no more aggression than comparable flat-screen games. Violent games raised aggressive cognition, but this did not translate into hostile affect or aggressive behavior.
Ferguson et al. (2022), British Journal of Social Psychology	Study 2: VR games differing in difficulty and violent content, the shooter Until Dawn: Rush of Blood and the racing game Driveclub	Two studies; VR used in Study 2	Violent content influenced neither hostility nor aggressive behavior; higher difficulty increased hostility.

Drummond and colleagues describe their 2021 experiments, which used ambulatory headsets that let players walk within a tracked space, as the first experimental tests of violent content in that type of VR. They concluded that their results were "early evidence against the notion" that such VR increases aggression compared with flat-screen games, while calling for replications with a wider variety of games. Presenting the earlier 3D result in 2014, Bushman said that "The combination of violent content and immersive technology like 3-D can be troublesome," and that this should be considered by "electronics manufacturers, video game developers, consumers, parents and content ratings agencies."

Violence as a research tool

Because presence makes virtual events feel real, VR lets researchers study violent situations that cannot ethically be staged. In a 2006 study, Slater and colleagues recreated Stanley Milgram's obedience experiment with a female virtual human who received increasing "electric shocks" whenever she answered a memory test wrongly and protested with growing discomfort. Of the 34 participants, 23 saw and heard her and 11 communicated with her only through a text interface. Although every participant knew that neither the woman nor the shocks were real, those who saw and heard her tended to respond at the subjective, behavioral and physiological levels as if the situation were real.

In a 2013 study, Slater's team showed a violent argument at life size in immersive VR to 40 male supporters of Arsenal Football Club. When the victim was presented as a fellow Arsenal supporter, participants made significantly more physical interventions than when he was not, which the authors took as support for a social identity explanation of bystander behavior.

Applications with violent offenders

VR has been tested as a way to change how violent offenders perceive their victims. In a 2018 study from the University of Barcelona and partner institutions, male domestic violence offenders and a control group without a history of violence wore a head-mounted display and experienced a virtual scene of abuse from a first-person view, with their bodies replaced by a life-sized virtual female body that moved with their own movements. Before the experience, offenders were significantly worse than controls at recognizing fear in female faces and tended to classify fearful faces as happy; after being embodied as the victim, their recognition of fearful female faces improved and that bias decreased. A 2020 follow-up with non-offender men found that the level of identification with the female avatar correlated with a decrease in prejudice against women, and that the first-person view tended to cause stronger behavioral and physiological reactions than watching the same scene as an observer.

Virtual Reality Aggression Prevention Training (VRAPT) uses VR role-play to practice the regulation of aggressive behavior. In a randomized controlled trial at four Dutch forensic psychiatric centers, published in 2020, 128 inpatients with aggressive behavior were assigned to 16 one-hour VRAPT sessions or a waiting list. VRAPT did not significantly reduce self-reported or staff-observed aggressive behavior, the primary outcomes. Hostility, anger control and non-planning impulsiveness improved at the end of treatment, but the improvements were not maintained at a three-month follow-up. A 2023 pilot study followed 14 male violent offenders who took part in VRAPT in a Swedish prison; individual results ranged from recovery to deterioration. A systematic review published in 2026 screened 1,317 articles and included 32 reports on VR in forensic and correctional psychiatry. It found VR interventions effective for purposes including reducing aggression and increasing empathy in intimate partner violence offenders, and rated the quality of the included studies as predominantly moderate.

Michael Madary and Thomas Metzinger's 2016 code of ethical conduct for VR saw "no problem at all" with voluntary use of empathy-inducing VR to rehabilitate violent offenders, but warned that governments and penal systems might adopt it as mandatory treatment, "calling to mind Anthony Burgess' A Clockwork Orange." They also noted that "Just as VR can be used to increase empathy, it can conceivably be used to decrease empathy," with obvious military applications in training soldiers "to feel no remorse about doing violence."

Industry and design responses

Content ratings

In January 2017 Oculus announced that the Oculus Store would move to age and content ratings from the International Age Rating Coalition (IARC), and that existing titles had to complete the IARC rating process by 1 March 2017 to avoid removal from the store. Meta states that all titles on its successor, the Meta Horizon Store, must show age and content ratings assigned through the IARC process. On Steam, developers complete a content survey whose mature content section requires disclosure of content "such as depictions of violence and sex"; Valve states that the answers determine how a game appears relative to each customer's preferences.

Developer and researcher views

In a March 2017 UploadVR article on violence in VR shooters such as Pavlov VR and Onward, Onward's creator Dante Buckley said: "Something that I've been thinking about lately is the ethics and the consciousness of violence in VR shooters." He suggested that as graphics improved, one answer could be to make such games more "casual", with less focus on realism. Arshya Vahabzadeh, a lecturer in psychiatry at Harvard Medical School, told the site that "Current clinical diagnosis of PTSD excludes exposures that occur through electronic media, including movies and pictures," but speculated that sufficiently realistic VR from which a person could not voluntarily exit might affect fear circuitry in a way similar to PTSD, calling it an area for further research.

In a 2018 CNN opinion piece, Jeremy Bailenson, founding director of Stanford's Virtual Human Interaction Lab, wrote that his argument "is not that virtual reality games are going to cause people to become violent", but that VR is an effective training tool because players use their arms and bodies in real combat movements. He proposed that shooting games give virtual bullets curved paths, replace realistic gun handling with abstract gestures, and use nonhuman targets such as robots, a practice he said "most, though not all, virtual reality games are adopting".

Slater and co-authors, drawing on Brey, suggest that designers ask what actions their application makes possible, how they are represented, and whether they are encouraged or discouraged: "In a game or another application in which killing is possible, for example, is this action encouraged or is it dissuaded? Is it rewarded or is it punished?" They also note proposals that developers disclose the potential effects of their content to users.

See also

Psychological trauma

Presence

Horror

Social VR