Tattoos and piercings, what do they tell you about someone?

Visual cues of low intelligence and low mental health

Richard Hanania and I recently wrote on Twitter/X:

Naturally, my reply and his general level of trolling on the platform attracted a lot of angry, dumb replies from people with tattoos and piercings, seemingly confirming the observation. Nevertheless, there were some others asking for the evidence, and since it seems no one has compiled this, let's do that here.

Tattoos and mental health

I searched for academic studies using Google Scholar with the query "tattoo mental health". This gives us studies like these:

• Pajor, A. J., Broniarczyk-Dyła, G., & Świtalska, J. (2015). Satisfaction with life, self-esteem and evaluation of mental health in people with tattoos or piercings. Psychiatria Polska, 49(3), 559-573.

  • Nothing was found, Polish sample, but relatively few with tattoos ('body modifications')

• McCandlish, C., & Pearson, M. (2023). Tattoos as symbols–an exploration of the relationship between tattoos and mental health. The Journal of Mental Health Training, Education and Practice, 18(3), 217-227.

  • 2000 Americans from MTurk

  • "We find that the presence, number, and specific features of tattoos are positively correlated with two of the health-related outcomes (ever diagnosed with a mental health issue and trouble sleeping) and all three of the risky behaviors (P < .05). Magnitudes are larger for those with multiple, visible, and offensive tattoos."

• Khosla, V., Joseph, V., & Gordon, H. (2010). Tattoos: what is their significance?. Advances in psychiatric treatment, 16(4), 281-287.

  • Review of older studies

  • "Evidence of a relationship between tattoos and psychiatric disorders comes from studies of psychiatric pathology in tattooed individuals in non-psychiatric settings (Lander 1943; Measey 1972) and of tattooed psychiatric populations (Ferguson-Rayport 1955; Gittleson 1969; Raspa 1990; Inch 1993; Williams 1998). Both types of study report higher rates of alcoholism, drug misuse, impulsivity, risk-taking behaviour and personality disorders in both men and women (Lander 1943; Post 1968; Measey 1972; Buhrich 1982; Raspa 1990; Inch 1993). These individuals are also more likely to have a history of self-harm, possibly as a result of a disordered personality and increased impulsivity (Goldstein 1979; Buhrich 1982)."

• Romans, S. E., Martin, J. L., Morris, E. M., & Harrison, K. (1998). Tattoos, childhood sexual abuse and adult psychiatric disorder in women. Archives of Women's Mental Health, 1, 137-141.

  • "A random community sub-sample of women originally selected from the electoral rolls in the Otago province of New Zealand, first studied in 1989, was followed up six years later in 1995.", n = 400ish, but old data

  • "Women with tattoos were more likely to be younger, to drink more alcohol, to have more psychiatric symptoms and to show borderline personality features than were the non tattooed women. They were also more likely to report CSA." but no age control, so maybe younger women are just more crazy in that sample.

• Cardasis, W., Huth‐Bocks, A., & Silk, K. R. (2008). Tattoos and antisocial personality disorder. Personality and Mental Health, 2(3), 171-182.

  • "male psychiatric inpatients (N = 36)", so a tiny but highly mentally ill sample.

  • "more forensic psychiatric inpatients with tattoos had a diagnosis of ASPD compared to patients without tattoos. Patients with ASPD also had a significantly greater number of tattoos, a trend toward having a greater percentage of their total body surface area tattooed, and were more likely to have a history of substance abuse than patients without ASPD. Tattooed subjects, with or without ASPD, were significantly more likely to have histories of substance abuse, sexual abuse and suicide attempts than non-tattooed patients."

• Perrotta, G. (2021). Massive use of tattoos and psychopathological clinical evidence. Archives of Community Medicine and Public Health, 7(2), 079-085.

  • "a population sample of 444 people" I guess they are Italians

  • "On the basis of these data, it is reasonable to argue that as the percentage of body surface area covered by tattoos increases, so do the dysfunctional traits of a specific main disorder. In particular, the recurrent dysfunctional traits are anxious, phobic, obsessive, somatic and bipolar in subjects with less than 25% of the body surface covered by tattoos, while borderline, narcissistic, antisocial, sadistic and masochistic traits are more frequent in subjects with more than 26% of the body surface covered by tattoos."

• Stirn, A., Hinz, A., & Brähler, E. (2006). Prevalence of tattooing and body piercing in Germany and perception of health, mental disorders, and sensation seeking among tattooed and body-pierced individuals. Journal of psychosomatic research, 60(5), 531-534.

  • "large and representative sample of German citizens (N=2043)."

  • "Within the group of individuals aged between 14 and 44 years, unemployment and nonaffiliation to a church are positively correlated, tattooing is significantly correlated with the perception of reduced mental health, and both tattooing and body piercing are correlated with significantly increased sensation-seeking behavior."

These are all from the first page of results, so of the first 10 studies found, 7 studies with data, 6 which found the expected relationship, 1 didn't. The studies are to cumbersome to meta-analyze as they don't report the right metrics, so we will have to be satisfied with this summary. Nevertheless, I think the picture is fairly clear.

Tattoos and intelligence

There are not many studies on this (using this query), but I found these two:

• Požgain, I., Barkić, J., Filaković, P., & Koić, O. (2004). Tattoo and personality traits in Croatian veterans. Yonsei medical journal, 45(2), 300-305.

  • "one hundred Croatian veterans"

  • "The non-tattooed group achieved higher scores on the IQ test (IQ=100) than the tattooed group TQ=95). EPQ test showed results either above or below the norms on all scales that were applied. The tattooed group demonstrated significantly higher levels of impulsiveness, adventurism, empathy and neuroticism than the non-tattooed one (p<0.05)."

• Jennings, W. G., Fox, B. H., & Farrington, D. P. (2014). Inked into crime? An examination of the causal relationship between tattoos and life-course offending among males from the Cambridge Study in Delinquent Development. Journal of Criminal Justice, 42(1), 77-84.

  • "prospective longitudinal study of 411 British males from the Cambridge Study in Delinquent Development"

  • "Concerning the role of personality features, tattoos, and crime, our results suggest that personality traits (e.g., individual risk fac- tors) assessed in early childhood such as daring disposition, low nonverbal intelligence, nervous/withdrawn, extraversion, neuroticism, and psychomotor impulsivity, are associated with tattoos and crime over the life-course. As such, these personality features appear to be part of the causal explanation for why individuals may become both tattooed and involved in crime. These findings are consistent with the prior literature which has suggested that tattoos may be an indication/manifestation of an underlying per- sonality disorder and this underlying personality disorder may be the criminogenic mechanism (Raspa & Cusack, 1990; Scutt & Gotch, 1986)."

Checking the results using OKCupid data

Fortunately, we can look in the OKCupid dataset. In fact, that's the original reason I started thinking about this issue. About 10 years ago, I blogged some of the first OKCupid results, including this one:

This was a very crude analysis. We can do better now. Here's the results with a better analysis (IRT scoring, sample limitations, age and sex controls). For tattoos and intelligence:

And for tattoos and mental health:

The effect sizes are about 0.4 z for intelligence, and about 0.15 z for tattoos (p values are 0.001 or so, the intervals above are prediction intervals). But our measures are quite crude, so the real effect sizes will be larger. Converting these values into IQ points is fraught because of the severe sampling bias in who fills out hundreds of questions on a dating site, and who answers honestly in public about their mental health (all answers visible to other users!). Given these constraints, we still find these results, so we can be fairly confident in them, especially since they appear to align well with most of the other research as well as our casual observations. You always use your own stereotypes as reasonable informative priors!

As a bonus, since I also made the claim about piercings, let's check these as well. OKCupid doesn't appear to have a regular question about piercings, but it does have these questions about piercings:

Concerning extreme piercings and intelligence:

Extreme piercings and mental health:

"Below the belt" genital piercings and intelligence:

And mental health:

Overall, then, results are in line:

• Mental health relates negatively to having tattoos, especially large ones or many of them. This is consistent across many studies across time and place.

• Intelligence relates negatively to having tattoos.

• Having piercings out of the ordinary relates negatively to intelligence and mental health.

• We replicate these findings using the OKCupid sample, which is a very large sample of mostly English speaking young to middle aged adults.

The more interesting sociological question is: if tattoos and piercings signal lower intelligence and worse mental health, how did they become so popular? Why do people get them? The most obvious reason might be that we are becoming dumber and more mentally ill as a civilization. There is some evidence concerning the first, and much regarding the latter. But the cohort changes are massive, so this can't explain everything, at least not from the perspective of individual-level causation:

Notice how their educational attainment and non-heterosexuality results indirectly confirm our results, insofar as education is a proxy for intelligence, and non-heterosexuality is a proxy for worse mental health.

The explanation for this rapid increase must therefore mainly be something else. As usual, I will forward a social media copy cat effect, where people copy various generally mentally ill people they see on social media. The population mistakenly uses them as a proxy for what is 'more elite' to do, even though statistics tell us that actually smart and mentally healthy people don't generally modify their bodies so much in these ways. Social media and media in general cause issues with humans' ancestral adaptation to copy the behaviors of those they perceive to be higher social status since it presents people with an abnormal sample of high social status (or apparent social status) people to copy from. I don't know what to do about this problem, though. Maybe we can just invest really good methods for removing the 'permanent ink'. On the other hand, for those looking to avoid these tendencies in friends and romantic partners, it provides great quick and dirty visual cues. Insofar as this is utilized, it might increase assortative mating for the underlying traits, which will slightly speed up natural selection for or against these traits.