A lot of people are afraid of people with mental illnesses because they think they're more violent but is that true?

I think it's wrong even though I don't really understand the pages I found on google. I found this article from University of Washington which states among other things that "the vast majority of people with mental illness are not violent". I understand this article but it is really short.

  • $\begingroup$ I don't think they are inherently more violent but probably more violent than sane people. Violence is natural and we (sane people) are educated to control it. Mentally ill people is not expected to understand/care about social conventions so their violence behavior is raw. $\endgroup$
    – Mandrill
    Apr 28, 2015 at 6:35

4 Answers 4


'Mental Illness' covers such a broad range of illnesses that it would be very difficult to answer your question. However, general consensus is that: people who are receiving an effective treatment for a mental illness are no more violent or dangerous than neurotypical/ 'healthy' people. People with a mental illness are more likely to harm themselves or to be harmed. There is a weak association between violence and mental illness, especially when alcohol or drug misuse is involved. Early treatment of mental illness is the key to preventing violence. This article covers the topic fairly thoroughly:

Link, B. G., & Stueve, A. (1995). Evidence bearing on mental illness as a possible cause of violent behavior. Epidemiologic Reviews, 17(1), 172-181.

This is from their conclusion on mental illness and public safety:

People with mental illness pose no greater threat to public safety than do such other "high risk" groups as men or young adults or people with less than a high school education.

  • 3
    $\begingroup$ "people who are receiving an effective treatment for a mental illness are no more violent or dangerous than neurotypical/ 'healthy' people" Isn't this more or less a No True Scotsman setup? A treatment which left the patient violent or dangerous would be considered ineffective, making the statement more or less circular. Answer is otherwise excellent, but this statement struck me. $\endgroup$
    – S McCrohan
    May 2, 2015 at 0:45
  • 1
    $\begingroup$ I think actually I could remove the 'effective treatment' part of that statement. The study cited mentions people who were once treated for a mental illness or were released from a psych facility, not necessarily those who are receiving ongoing treatment $\endgroup$
    – queenslug
    May 2, 2015 at 11:22

There are certain illnesses which are associated with violence, or have aggression as a symptom (which will not always result in physical violence). Especially the Cluster B personality disorders come to mind, with narcissistic rage and crimes perpetrated by people with antisocial personality disorder. Oppositional defiant disorder in children also includes aggression, as well as some addictions like alcoholism.

But these are only a few of the existing mental conditions. Basically, any single system in a human's brain can be messed up, including the ones responsible for aggression. The handful of conditions in which this happens is only a small percentage of all possible conditions. It is not that different from non-mental illness: there are hundreds of diagnoses, but only a small part of them are connected to a specific organ, say, the liver. Most people who suffer from a physical condition do not have a messed up liver; similarly, most people who suffer from a mental condition don't have a messed up aggression control.

If you are concerned with public safety, then it is quite possible that better methods for diagnosing and treating the aggression related mental conditions will result in a society with less violent crime. For example, this study found that 35% of a random sample of prison inmates has antisocial personality disorder, and goes on to mention that it is especially common in violent offenders, while I've seen numbers for the general population estimated at between 1 and 5% (sorry, no citation on that currently present, maybe somebody has better data). So if we knew how to heal people with APD, we could expect less violence in society, as they contribute so much to it.

But if you are concerned with your own personal safety, I would not recommend avoiding people who appear to have a mental illness. The conditions which have the most visible symptoms - the Down child with a creepy smile, the autistic boy who repeatedly jerks his head - are not the ones which are usually connected to aggression. The (potentially aggressive) people with cluster B disorders I mentioned above are hard to diagnose even for specialists, and appear outwardly normal at first. If you are dealing with somebody who has them, you are likely to notice a proneness to aggression which strikes you as unusually high long before you have even thought that this person could have a mental illness. And with some of them, you might not be aware of their condition even after spending years in a close familial relationship with them. So, you cannot use their condition as a heuristic to avoid them, because you are not aware of it at all. If you are on the street and see somebody who reminds you of a caricatural media portrayal of "mentally ill", the risk that he is prone to aggression is not that much higher than that of the normally looking passerbys.

If an acquaintance shares with you that they have a mental illness diagnosis, there is no reason to panic either. Research the specific illness they have so you can better understand it and support them in living with it, for most mental conditions you are not taking any risk for yourself in doing so. And if it is one of the few conditions where they can be violent, the research should easily uncover that, so you can make an informed decision.


Part of the very basic components of abnormal psychology suggests that any behaviour that could be dangerous to oneself or another individual could qualify for a diagnosis for mental illness. But, this is only when other factors are taken into account, for example the presence of dysfunctional behaviour that hinders appropriate functioning in society. (Comer, 2005)

Someone who is mentally ill may be more dangerous than someone who is not. However, you should note that only a small select few mental illnesses involve violence in their diagnosis, antisocial personality, etc.

Most mentally ill people are not violent.


Comer, R. J. (2011). Fundamentals of abnormal psychology. New York: Worth.

Neurodiverse people may still have the tools to manipulate people just fine. In fact both attention deficit disorder (ADD) and antisocial personalities are theorized to be characterized by propensities of stimulus hunger, as in low arousal theory. (Hare, 1970) This could suggest that ADD and Antisocial personality disorders may share fundamental characteristics.

Note also that Neurodiverse individuals may have issues with executive functioning that may impair social interaction, but also note that between Autism and Attention-Deficit disorder these impairments occur for different reasons. An individual with ADD may have trouble reading social cues because of working memory deficits and inattention. But with Autism it is shown to possibly be due to a difference of semantic processing strategy. (Moseley, 2013), a local processing bias (with global processing not in tact) (Drake, 2010) , as well as working memory deficits.

Neurodiversity covers a wide range of neurological disorders. Autism being one of the most prominent


Hare, Robert D. Psychopathy: Theory and Research: Theory and Research. New York <<[u.a.: Wiley, 1970. Print.

Moseley, Rachel L. et al. “Brain Routes for Reading in Adults with and without Autism: EMEG Evidence.” Journal of Autism and Developmental Disorders 44.1 (2014): 137–153. PMC. Web. 8 June 2016.

Drake, Jennifer E. et al. “`Autistic’ Local Processing Bias Also Found in Children Gifted in Realistic Drawing.” Journal of Autism and Developmental Disorders 40.6 (2010): 762–773. Web.

  • $\begingroup$ Again haunting you - referencing is strongly encouraged on scientific SE sites like this one. It's more of a comment now, without giving users the opportunity to background-read on your answer. Also, try to remove speech-language like So not in a way, not really most.... It doesn't make sense. $\endgroup$
    – AliceD
    Jun 8, 2016 at 20:48
  • 1
    $\begingroup$ Again, awesome work +1 Thanks for this $\endgroup$
    – AliceD
    Jun 8, 2016 at 21:16
  • $\begingroup$ Amazing work and references. $\endgroup$
    – Seanny123
    Jun 9, 2016 at 16:05

I love the way this question is structured. It naturally implies a prejudice and a hatred for those who're mentally disabled, which seems to imply that neurotypical people tend to be more violent.

And that's actually true. There's a history of neurotypicality resulting in greater violence and psychopathy than neurodiversity. The thing is? A neurodiverse person has a hard time being a sociopath because they don't have the tools necessary to manipulate people. So when you think of that person who snaps and goes around killing people? That's a neurotypical extravert.

Look up Ted Bundy and you'll see what I mean. Neurotypicals are far likely to snap and become serial killers. Not just that, but numerous studies have shown that neurodiverse people are far less prone to the kind of prejudices and bigotries that neurotyipcal people can have a proclivity for. Prejudices are especially common in neurotypical extraverts a they prefer to have 'their own kind' surrounding them.

So, no. Not really. If anything, the opposite is true. I feel safer being around neurodiverse people, because I know they won't snap and murder me in the most depraved way.

  • 3
    $\begingroup$ Welcome to CogSci. Could you add sources or references to your answer? It looks more like an opinion (i.e., a comment), rather than an answer. $\endgroup$
    – AliceD
    Jun 5, 2016 at 20:02

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.