Disinhibition: A Source of Internet Aggression


The Internet is a fundamental tool in modern culture for communication. One distinct difference exists between traditional face-to-face interpersonal communication and online communication, that is the lack of physical social contact online. Various psychological influences impact online behaviors in ways that may not be present offline. Unlike in-person communication, online dialogues become permanent evidence of one’s behavior. With a mass prevalence of sharing, occasionally some of the things shared will incite anger and frustration. The ability to quickly communicate under a mask of anonymity can evoke temptation towards bad behaviors. It is intriguing to witness negative anti-social behavior online when those committing the act take no action to hide their identity; while distinctly obvious their activities are public. On Twitter alone approximately 10,000 (1 in every 15,000) tweets per day contain racist or ethnic slurs [1]. Some of these tweets are from accounts purposely intended to incite controversy that attempt to mask the users true identity [2]. Still some people publicly communicate in social networks as if they were private.

But not all negative online behavior originates from those with hurtful intent. A distinction should be made from those commonly referred to as “trolls” individuals with conscious ill intent to cause negative or hurtful emotions [2] and between those who cause controversy without necessarily intent to do so. Many publicly accessible social media accounts with well-defined associations with the users true identity have at some point produced less then prosocial content. Even some high-ranking public personalities in past years have had to resign due to socially unacceptable comments or images. Figure 1 is a screenshot of a tweet from a real woman who was fired as communications director at a New York based company because of the debatable nature of tweet she posted in December 2013 “Going to Africa. Hope I don’t get AIDS. Just kidding. I’m white!” [3]. What is it about digital environments that entice people to engage in negative behavior online?


Figure 1

Social psychology offers some explanations; as digital environments influence individuals’ perceptions of identity affecting the amount of involvement and exposure towards antisocial behavior. The Online Disinhibition Effect is a phenomena describing the behavioral influences that reduced inhibitions or self-control are caused by circumstances such as anonymity, invisibility, and minimization of authority [4]. Individuals with lowed inhibitions are more likely to engage in behaviors that they would not likely participate under normal conditions. Disinhibition can be categorized in two types, benign and toxic [4]. Benign disinhibition characterizes non-hurtful acts such as disclosing personal information like emotions and fears. Toxic disinhibition is more destructive resulting in increased activity in crime, pornography and aggression. It is this toxic disinhibition that is of interest regarding hostility.

Anonymity refers to the extent of one’s identity is detached from their actions. In more anonymous online environments sexual disclosure is more prevalent because others are not able to determine one’s true identity beyond a screen-name [5]. Being anonymous affords opportunities for experimentation, either sexual or boundary testing. However anonymity does not independently cause disinhibition. Invisibility though connected alongside anonymity differs because hiding one’s identity is not necessary. Invisibility, simply refers to not being able to see other people [4]. Eyes are an important property of interpersonal relations. One issue with maintaining the high levels of connectedness in a digital environment is the lack of real-time eye contact. It is this lack of eye contact that contributes to online disinhibition. In one study investigating the difference in communication between text only dialogue and webcam dialogue determined that the more anonymous and invisible correspondents were the more hostile people become [6].

Much of this disinhibition can be attributed towards differing levels of self-control and self-esteem. Social networks decrease self-control by enhancing self-esteem [7]. This heightened self-esteem is a result of self-presentation from social networks, which afford controllability over selecting particular information users can present publically. Certain forms of high self-esteem seem to increase one’s susceptibility to violence [8]. In addition those with higher self-esteem have been associated with greater reactivity leading to more anger, defensive justifications, selfishness and antisocial effects [9]. This heightened self-esteem when threatened can provoke damaging behavior.

Another phenomena closely related to disinhibition is deindivduation “the tendency for an individual within a groups to let go of self-awareness and restraint and do what the group is doing” [10]. Unlike disinhibition that is reliant on how a person fits within a group, deindivduation influences how group dynamics impact its members. Many classic experiments have been conducted on the effects of deindivduation, one such study examined the degree hooded students would administer electric shocks on another human versus non-hooded participants. The results showed that the participants who were deindividuated in the hooded condition administered twice as much eclectic shock then the non-hooded students [10]. Online communication is much like wearing a hood among others who cannot be seen nor identified, resulting in potentially more harmful conduct.

Group polarization “the tendency for an attitudes or beliefs to become magnified within a group after members discuss an issue among themselves” [10]. One study examined deindividuated and individuated participant responses within an online discussion forum. Those in the deindividuated condition reported greater feelings of within-group similarity. Deindividuated participants also shifted their attitudes to a more extreme stance after group discussions [11]. In face-to-face group division there is just as much of a tendency to polarize group ideology. Online when combined with anonymous and invisible group members, the baseline polarization per member is already skewed. When a group of disinhibited individuals congregate, the overall ideological bias many polarize more significantly. We see such effects within riots where rioters wear masks to hide their identity from authorities such rioters are also the most violent.

Members of online groups may overestimate the extent their beliefs are shared among the rest of the online community. Overconfidence offers the possibility to post something controversial with the belief of having more support in favor then against a position. False consensus effect “a phenomena that causes individuals to assume that everyone shares the same option they do” [10]. This overestimation can be seen between various online forums and sub- cultures. Among ideologically similar online groups, members are likely to overestimate support of their opinions by roughly five to seven percent [12]. Lack of exposure to other viewpoints is likely a cause of false consensus among Internet groups. Some of the favoritism towards topics of similar nature and exclusion of different opinions is partly unintentional. Eli Pariser in his book “The Filter Bubble: What the Internet Is Hiding from You” he exposes how internet search engines like Google, track user history and will tailor search results in a personalized manner to provide articles with similar ideological emphasis [13]. When unintentionally directed towards groups with similar viewpoints, views expressed become accustomed and the perception of the overall support is inflated due to the lack of dissimilar material.

Some results of disinhibited behavior and deindivduation is the easier lying becomes. Lying from a distance such as online elicits less negative emotions on the part of the liar then face-to-face. In one study participants were more likely to lie about themselves to another person over online text based formats, versus face-to-face [14]. Participants in an e-mail condition had the highest proportion of lies and participants in the face-to-face condition had the lowest percentage. When asked to deceive while given the option of media to use, a preference is towards modes that are non-reprocessable, or nonrecordable [15]. Either it be conscious or unconscious, it is innately known the best way to lie over any medium is thorough a method that has the least association concerning the liar. Greater the dissociations equals more potential to lie.

Certainly online environments influence behavior that would not ordinarily be present elsewhere under normal circumstances. It is not simply the reason that people are naturally mean and aggressive or have ill intent to harm. Rather deindividuated circumstances create false perceptions of reality pertaining to social norms and risk. It is also not the case that any single psychological influence is reasonable for disinhibited conduct. Each factor contributing toward disinhibited influences independently affects a person but the levels of multiple influences ultimately drive destructive online behavior.

Sources cited
[1]       Bartlett, Jamie, Reffin, Jeremy, Rumball, Jeremy, and Williamson, Sarah, “Anti-social media,” Demos, Feb. 2014.
[2]       P. Shachaf and N. Hara, “Beyond vandalism: Wikipedia trolls,” J. Inf. Sci., vol. 36, no. 3, pp. 357–370, Jun. 2010.
[3]       E. Pilkington, “Justine Sacco, PR executive fired over racist tweet, ‘ashamed,’” The Guardian, 22-Dec-2013.
[4]       J. Suler, “The Online Disinhibition Effect,” Cyberpsychol. Behav., vol. 7, no. 3, pp. 321–326, Jun. 2004.
[5]       W.-B. Chiou, “Adolescents’ sexual self-disclosure on the Internet: deindividuation and impression management,” Adolescence, vol. 41, no. 163, pp. 547–561, Sep. 2006.
[6]       N. Lapidot-Lefler and A. Barak, “Effects of anonymity, invisibility, and lack of eye-contact on toxic online disinhibition,” Comput. Hum. Behav., vol. 28, no. 2, pp. 434–443, Mar. 2012.
[7]       K. Wilcox and A. T. Stephen, “Are Close Friends the Enemy? Online Social Networks, Self-Esteem, and Self-Control,” J. Consum. Res., vol. 40, no. 1, pp. 90–103, Jun. 2013.
[8]       R. F. Baumeister, L. Smart, and J. M. Boden, “Relation of threatened egotism to violence and aggression: The dark side of high self-esteem,” Psychol. Rev., vol. 103, no. 1, pp. 5–33, Jan. 1996.
[9]       I. McGregor, K. A. Nash, and M. Inzlicht, “Threat, high self-esteem, and reactive approach-motivation: Electroencephalographic evidence,” J. Exp. Soc. Psychol., vol. 45, no. 4, pp. 1003–1007, Jul. 2009.
[10]     K. Duff and K. A. Peace, Think social psychology. Toronto: Pearson Education, 2012.
[11]     E.-J. Lee, “Deindividuation Effects on Group Polarization in Computer-Mediated Communication: The Role of Group Identification, Public-Self-Awareness, and Perceived Argument Quality,” J. Commun., vol. 57, no. 2, pp. 385–403, 2007.
[12]     M. Wojcieszak, “False Consensus Goes Online Impact of Ideologically Homogeneous Groups on False Consensus,” Public Opin. Q., vol. 72, no. 4, pp. 781–791, Dec. 2008.
[13]     E. Pariser, The Filter Bubble: What the Internet Is Hiding from You by Eli Pariser. Penguin Press HC, The.
[14]     M. Zimbler and R. S. Feldman, “Liar, Liar, Hard Drive on Fire: How Media Context Affects Lying Behavior,” J. Appl. Soc. Psychol., vol. 41, no. 10, pp. 2492–2507, 2011.
[15]     J. R. Carlson, J. F. George, J. K. Burgoon, M. Adkins, and C. H. White, “Deception in Computer-Mediated Communication,” Group Decis. Negot., vol. 13, no. 1, pp. 5–28, Jan. 2004.