Understanding Incels’ Psychology, Ideology, and Networking

By Joe Whittaker (Swansea University), William Costello (University of Texas at Austin), and Andrew Thomas (Swansea University)

Involuntary Celibates (incels) have become a prescient security concern in recent years. This is, in large part, due to the handful of terror attacks conducted by individuals who are part of the online movement, who forge a sense of identity around their perceived inability to form sexual or romantic relationships and develop misogynistic attitudes. The exact number of attacks conducted by incels is contested, particularly given the lack of formal group membership as with other movements or ideologies. Hoffman, Ware, and Shapiro outline three “clear incel-motivated terrorist attacks” – Isla Vista in May 2014, Toronto in April 2018, and Tallahassee in November 2018.

Despite a huge growth in incel research since 2018, our study aims to fill several gaps highlighted by scholars in the field. Last year, Hart and Huber argued that there needs to be a greater emphasis on human participants to expand our knowledge of the incel subculture. Similarly, in a 2023 edition of this blog, Stephane Baele offers a range of paths forward for an incel research agenda, including cross-disciplinary work on the topic of mental health and neurodiversity between researchers of extremism and psychologists. Our research, the largest survey of incels to date (n=561), seeks to understand several key factors of incels’: psychological experiences; adherence to ideology; and their networks. We present here a (very) brief summary of some of the most salient findings.

Psychological Experiences

Incels in our sample displayed exceptionally poor mental health. On the PHQ-9 depression questionnaire, 39% scored over 15 points (which suggests “moderately severe” depression). Around one in five responded that they had suicidal thoughts on a daily basis. When given the Generalised Anxiety Disorder Assessment (GAD-7), 43% scored over 10 points, which is the cut-off point for moderate anxiety and for medical referral. Participants also scored highly on measures for loneliness and rejection sensitivity. When given the Autism Spectrum Quotient (AQ-10), slightly over 30% received a score of 6 or higher, which is the recommended point for referral to a specialist diagnostic assessment. These findings are congruent with existing research, which has consistently found that incels are neurodiverse and suffer from poor mental health. Other measures suggest a host of problematic attitudes. Incels scored highly on measures for: Hostile sexism; rape myth acceptance; angry rumination; and revenge planning.

Adherence to ideology

As with many movements, it can be difficult to pin down a consistent ideology among members. Sharkey highlights that beliefs and ideas go in and out of fashion on incel forums and there is not necessarily a coherent set of beliefs. This was reflected in by participants; only two-thirds believed that such a thing as “incel ideology” exists. To expand on this, we created a statement of ideology from an analysis of the existing academic literature that encapsulates some of the most common incels views around biological determinism and the idea that 80% of women only desire 20% of men, leaving those at the bottom destined for a life of celibacy and loneliness. Around four in five participants believed that most incels agree with the statement, while around two-thirds believed it themselves.

There was also general agreement that incels believed that they are discriminated against by society. When asked about who they deemed the “enemies” of incels, they responded with the following in order of most to least: feminists, the political left, wider society, women, non-incel men, incels themselves, the political right. Interestingly, each of these groups scored above average, demonstrating that while there are primary out-groups (such as feminists), incels blame several groups for their plight.

It is difficult to successfully gauge participant radicalisation in surveys; most studies opt for some kind of proxy indicator. We opted for a measure which asked whether violence could be justified against those that incels believe are causing them harm. Around 6% responded that violence was “often” justified and a further 19% responded “sometimes.” We should stress that this is an imperfect measure for understanding radicalisation because it measures attitude rather than behaviour and people often do not engage in behaviours that they intend to (often called the intention-behaviour gap), and therefore likely over-estimates violence. Moreover, one could make the case that social desirability bias may cause under-representation of violence-justification. Although roughly one-quarter of respondents justifying violence is alarming, this is (perhaps more distressingly) in keeping with societal baselines in opinion polling. Wintemute and colleagues found that around 1 in 5 people in the US thought that violence is sometimes, usually, or always justified, while Ipsos found a similar figure amongst French respondents.


In terms of communicated with like-minded people, anonymous social media and fora (e.g. the chans) was the most popular type of platform with over half (55%) stating that they used this to communicate with other incels, closely followed by pseudonymous platforms (e.g. incel fora with usernames) at 51%. Around a third said that they use Discord; slightly under one in five used messaging apps; with fewer than 1 in 10 using telephone or video calls. Much of the existing empirical literature focuses on analyses of incel fora, which seems justified given that our participants highlighted it as the most popular means of communication. However, our study paints a relatively diverse picture, supporting the recent incel ecosystem study conducted by Baele, Brace, and Ging, who find a range of platforms that have been used in recent years. One of the most surprising findings was that almost one in five (18%) incels had engaged in face-to-face communication with another incel. This is unexpected because incels are typically considered to be an “online only” phenomenon, making them an outlier compared to most other extremist-supporting ecologies, which tend to overate in both the online and offline domains. It is important to note that there is no reason to think that incels are necessarily mobilising offline in any meaningful way. It is more likely that participants answered this question affirmatively because they know someone who is also an incel, although this remains an open question for future research.

Predicting Harmful Attitudes

To develop a better understanding of how these measures contribute towards the development of problematic attitudes, we conducted a pathway analysis. We created a factor called “harm” which consists of: angry rumination, revenge planning, displaced aggression, hostile sexism, rape myth acceptance, and justification of violence. We then created factors made up of measures that pertain to: mental health and neurodiversity; adherence to ideology; and networking. The analysis demonstrated that all three factors had direct effects and therefore predicted “harm.” We also found that adherence to ideology and mental health were correlated, suggesting that poorer mental health (e.g. depression and anxiety) leads to greater adherence to the incel worldview (such as belief in the 80/20 rule and seeing feminists as the enemy) and vice versa. Therefore, these predictors work in tandem, exacerbating one another and strengthening harmful attitudes such as rape myth acceptance and condoning incel violence.

We then sought to predict whether attitudes predicted a respondent answering that violence was “often” justified (around 6% of the sample). The largest effects were related to misogynistic views, how one handles anger, and feelings of being discriminated against. One particularly interesting finding was political orientation – while the sample as a whole answered measures developed from Pew Research political spectrum questions as slightly left of centre, those that justified violence were more likely to be right-leaning.


Our study aimed to answer calls made by scholars to bolster incel research with human participants and drop cross-disciplinary drawbridges between extremism research and psychology. Some aspects of this survey were in line with expectations, such as the high levels of neurodiversity and poor mental health; the general adherence to biological determinism; and the framing of feminists as the primary out-group. Other findings were more of a surprise, like the one-in-five incels that have networked face-to-face; the similar number that justified violence; or the fact that the sample as a whole as politically slightly left of centre. Most importantly, when trying to understand what factors may lead to the development of problematic attitudes – a possible but imperfect proxy for radicalisation – we can see that it is a complex picture. All three factors of mental health, adherence to ideology, and networking play a role in driving it and, in some cases, reinforce and exacerbate each other. Fundamentally, it underpins existing research which paints radicalisation as a complex and multicausal phenomenon.

Joe Whittaker is a senior lecturer in Criminology, Sociology, and Social Policy at Swansea University.

William Costello is a PhD student at University of Texas at Austin

Andrew Thomas is a senior lecturer in Psychology at Swansea University.

Image Credit: Original image by Kevin Walsh via Wikimedia Commons, adapted by Lucian Stephenson.